Activities of the Open-Ended Online Forum (2014-2016)

POSTED ON BEHALF OF HELMUT GAUGITSCH

-----

Dear participants of the Online Forum,

I am pleased to have been invited to moderate this online discussion to review proposed revisions to the Guidance.

As you know, the AHTEG met face-to-face last November in Brasilia, and agreed on a number of revisions with a view to improving the Guidance. At that meeting, it was also agreed that four topics which were prioritized for further development of guidance materials could be incorporated into the existing Guidance, by creating text boxes or adding new text, as appropriate. These topics are:

     -> LMOs introduced in centres of origin and genetic diversity” and “LMOs intended for introduction into unmanaged ecosystems” (the two topics will be addressed together);

     -> “LMOs created through use of dsRNA techniques, engineered to produce dsRNA or dsRNA” and “LMOs containing RNAi” (the two topics will be addressed together);

     -> “Integrating human health into the environmental risk assessment” taking into account the topics “Nutritionally altered living modified plants” and “LMOs that produce pharmaceutical products”, as appropriate;

     -> “Synergistic impacts of different herbicides that are part of the technology package that accompanies certain LMOs”.

Following the AHTEG meeting, we had a discussion as open-ended online expert forum during 1-15 February 2016 to gather views, relevant guidance and sources of information on the four topics indicated above (http://bch.cbd.int/onlineconferences/onlineconferences/forum_ra/discussion.shtml).

Furthermore, following the AHTEG meeting, the Sub-group reviewed and addressed the comments from the testing that were still outstanding (http://bch.cbd.int/onlineconferences/RA_ahteg_subgroup).

For this discussion, what you have before you as the background document is a draft revised Guidance resulting from the work of the AHTEG sub-Group on both issues, the outstanding comments from the testing and the text boxes on the four issues mentioned above.  I would like to thank all the members of the sub-Group and the Secretariat for their hard work and dedication to this task. The background document is available in the page of our discussion (http://bch.cbd.int/onlineconferences/onlineconferences/forum_ra/discussion.shtml).

Briefly, the suggested revisions relate to the incorporation of text boxes on topics to the Guidance on the basis of feedback by the online forum, and further revisions on the basis of the comments provided through the testing of the Guidance. All suggested revisions are marked in “track changes” in relation to the draft revised Guidance resulting from the last face-to-face meeting of the AHTEG.

During the next two weeks, you are invited to review and provide feedback on the most recent revisions to the Guidance. In doing so, you are kindly requested to limit your comments to the new revisions since the last face-to-face meeting of the AHTEG (i.e. those marked as “track changes”). In providing your comments, please refer to specific line numbers or highlight the text directly in the Word document containing the Guidance and upload it together with your message.

I would also like to invite the members of the Sub-group to participate in the discussion and provide clarifications, if needed.

I am looking forward to two weeks of constructive and fruitful discussions which will bring us a big step closer to the goal of developing and improved version for the consideration of the COPMOP. I am looking forward to moderating this discussion.

The floor is open!

Best wishes
Helmut Gaugitsch

Dear Helmut, Dear all

First I would like to thanks all the group that worked on the update of the guidance document. I made a first reading of the document and in the following few lines, as an ice breaking activity, I will include some comments and suggestions.

Lines 200-203. I would prefer to replace those examples with the ones highlighted under the CBD to give an idea about the full range of protection goals. Concrete text proposal would be
“- Ecosystems and habitats: containing high diversity, large numbers of endemic or threatened species, or wilderness; required by migratory species; of social, economic, cultural or scientific importance; or, which are representative, unique or associated with key evolutionary or other biological processes;
- Species and communities which are: threatened; wild relatives of domesticated or cultivated species; of medicinal, agricultural or other economic value; or social, scientific or cultural importance; or importance for research into the conservation and sustainable use of biological diversity, such as indicator species; and
- Described genomes and genes of social, scientific or economic importance.”

Lines 305-312. I think that this paragraph is missing an important aspect. It needs to highlight that information needed depend on the specific case of risk assessment and its elements (the LMO itself, the parenteral organism(s), the intended use and the likely potential receiving environment) any change of those elements may implies that different/ additional information may be required. After this introductory statement we can go ahead with those further explanation and examples given in the para.

Lines 334-335. I think I have problems with the proposed amendments especially the footnote. It seems that all of a sudden we have chosen to adopt the EPA definition and approach towards handling the uncertainty. This idea was never on board in the discussions and I think that the original text reflects better the nature of the discussions we had. Furthermore, I don’t think it is appropriate to add some webpage as a footnote bearing in mind that the content may and will change with time.

Lines 350-352. It could benefit from a little rewording. Proposed text would be:
“Considerations and communication of uncertainty affect the level of acceptability of risk. However, it may also improve the understanding of the outcomes of a risk assessment, strengthen the scientific validity of a risk assessment and provide transparency in the decision making process.”

Lines 385-389
While the mother convention CBD gives those definitions
"Country of origin of genetic resources" means the country which possesses those genetic resources in in-situ conditions.
"In-situ conditions" means conditions where genetic resources exist within ecosystems and natural habitats, and, in the case of domesticated or cultivated species, in the surroundings where they have developed their distinctive properties.
I would prefer to develop some definition that is mainly based on those for definition before looking into other sources.

Lines 398-401
The language is quite weak in highlighting the importance of those centers. I would prefer a stronger language that is better in line with the CBD “noting the fundamental requirement for the conservation of biological diversity is the in-situ conservation of ecosystems and natural habitats and the maintenance and recovery of viable populations of species in their natural surroundings”

With regards to the whole box going from 385-409
I think the box is missing many vital elements such as the importance of in-situ conservation importance to indigenous and local communities. The para could benefit from some idea in the following statement
“Recognizing the close and traditional dependence of many indigenous and local communities embodying traditional lifestyles on biological resources, and the desirability of sharing equitably benefits arising from the use of traditional knowledge, innovations and practices relevant to the conservation of biological diversity and the sustainable use of its components.”
Another important missing aspect is establishing the linkage between this and the article of the CBD on the in-situ conservation (Article 8)

Lines 455-456
The proposed amendment does not reflect well the null hypothesis. I think we need to review that.

Line 476 we should add “including organic farming” after the word practice to capture the full range of ideas reflected in the original text.

Line 511 do not replace “do” with “will”

Box starting 584
I had a look at the box and was wondering why did we leave some relevant elements in the “use of term section” and did not merge all effects together in one place? In this regards, I would mention “unintended effect”. I think it is useful to check the definitions in that box and the one existing in the use of terms section of the document.
I would recommend to have a short statement here highlighting the type of effects that exists and leave the definition to the use of term section of the document
Box on LMOs developed through RNAi-based methods to reduce gene expression (something is wrong with the line numbering for that box)
I think the box should be introduced just after point to consider d as this point talks about the modification method so it is more relevant there.

Concerning the box starting line 791 on human health, I think the issue of human health needs more than just a box rather a separate section like the one on trees and abiotic stress due to the importance of the issue.
Secondly, I think it is missing a lot of important information and does not provide proper guidance on the issue of human health.
It is also important to highlight that human health is a cross cutting issue between environmental risk assessment, socioeconomic considerations and food toxicity studies should be reflected in a more appropriate way and how we are going to deal with that.

Para 958-964
I would prefer to keep the original text. The new text is misleading and gives the impression that some jurisdictions are not in compliance with the protocol.

Line 1084, I think we need to strike down “considered as single events and”

Finally for the issues that are still marked as “outstanding” in the provided text. I will request the chair to kindly provide clarification on how we are going to proceed with them.

Warm regards,
O.A.El-kawy

Dear Helmut, Secretariat and colleagues

I would like to thank the members of the Sub-group and the Secretariat for the work to incorporate text boxes with the suggestions on certain topics to the Guidance.
In my first reading of the document, it called my attention the kind of information required for small-scale releases and the definition of this concept. So, I reviewed the last bullet of the item ¨The relevance of information for the risk assessment¨ that is related with small-scale releases information and the text of the box below.

Line 309 to 312: I would like to note in relation to the focus of the required information for the risk assessment of LMOs small-scale releases, that beside the required information should be focused on certain specific hazard pathways, the relevant information that the risk assessor shall take into account with an experimental LMO, it is to know which are the confinement measures to prevent the escape of LMO beyond the limits of the trial and/or the elimination procedures of LMOs after the end of the trial. Since those LMOs are regulated materials. Therefore, I propose complete the paragraph as follow.
- The information available for the risk assessment of LMOs small-scale releases, such as field trials, is typically more limited and focused on specific hazard pathways, the containment requirements to prevent the escape of LMOs beyond the limits of the small-scale release and/or the procedures for elimination of LMOs after the end of a trial. Information collected during small-scale releases may provide useful information for further risk assessments of large-scale releases.-

Line 313: I suggest to call in a general way all the different objectives of a small-scale releases as ¨small scale releases¨ and then in the box explain what this term include. In this way I propose to withdraw the examples through the text of the box. 

Proposed text for box: Information requirements in the case of small-scale releases
 
Small-scale releases include field trials or experimental releases.  At early experimental stages or in the early steps of environmental releases of LMOs that are conducted in a step-wise manner, the nature and detail of the information that is required or available may differ compared to the information required or available for large scale or commercial environmental releases. The information required is related closely to the objective of the release. Typically, less information is required, or even available, for risk assessments where the exposure of the environment to the LMO is limited, as one of the objectives of such environmental releases is to generate information for further risk assessments. In such cases, the uncertainty resulting from the limited available information may be addressed by risk management and monitoring measures and, therefore, information on measures to minimize the exposure of the environment to the LMO is particularly relevant.
Therefore, some of the information identified throughout the Roadmap may not be known or be only partly relevant in the context of a small-scale release where the environment would have limited exposure to the LMO.


In the next days I will continue reading in detail the rest of the text boxes and the new changes on the Guidance to give more feedback.

Best regards,

Patricia Gadaleta

Dear Helmut, Dear all

Greetings from Helsinki

With regards to the text inside the boxes covering the 4 newly introduced topics. I believe structure harmonization is needed. The way the text is presented now is very narrative and not well structured.

I would suggest that text inside those boxes should be arranged into 2 subsections (not essentially with headings). One section serves as an introduction to the topic setting the scene to the topic and talking about why it needs special attention. The other section includes the elements for further considerations (additional points other than those already under the roadmap that are specific for the topic or may be emphasizing certain points that are present under the roadmap but from a different perspective ) in the form of bullet points.

Regards,
O.A.El-Kawy

Dear colleagues,

as moderator of this online discussion I would like to thank those of you very much, who have already provided first inputs to our discussion! The discussion we are having is very important with respect to the overall goal of improving the Guidance and I therefore invite all of you to share your reflections on the new elements in the current draft Guidance and provide your input during the coming days. The discussion is still open for another week and it will close on 9 May at 1:00 am, which means in fact coming Sunday (8 May).

With respect to Ossama`s question of clarification to the Chair, raised in his first intervention a couple of days ago, I would like to provide the following answer: we will take up the outstanding issues at the AHTEG Meeting in July 2016, but those participants who wish to do so, are welcome to provide very concrete text proposals to help achieve the compromise that is needed on those issues.

I am looking forward to reading your postings and to our further discussions.
Thank you and best wishes

Helmut Gaugitsch

Dear all,

Thank you very much for the opportunity to contribute to the discussion and all the efforts made for the revision of the guidance (GD).

As indicated previously, I do not consider that the addition of boxes to cover specific LMOs (i.e., RNAi-based GMPs) or areas of concern (i.e., centres of origin) is the most sensible way forward for the following reasons. The first of the reasons is that it diverts the focus of the discussion on issues of secondary importance. In my view, there is a lack of clarity on: (1) the overall objectives of the GD, (2) its target audience, (3) key messages to convey; (4) the level of detail given to specific risk assessment (RA) considerations, and (5) its structure/organisation. Perhaps it would therefore be more valuable to discuss these points first, so that further revisions can be tailored accordingly.

In its present form, the GD is shaped as an extensive review of all possible RA considerations to account for when risk assessing LMOs, but fails to make a clear distinction between what is nice and need to know under which circumstances, and to explain how these considerations should be implemented in practice. No concrete criteria to use for implementation purposes are given, though this should be the main purpose of the GD, in my view. The GD is therefore not practical. Plus, several recommendations are too theoretical/conceptual, and do not reflect the LMO RA experience gained over the last three decades.

The second reason is that the examples given in the boxes fail to meet their purpose, complicating the practical use of the GD even more. In my view, the examples given in the boxes should be used as case studies to test the robustness of the GD during the revision process. Based on the outcome of this exercise, specific RA considerations could be given in the core text of the GD to account for the specificities of the specific cases considered, where appropriate. In this way, one would develop a GD that is sufficiently robust to capture all relevant cases. Consequently, it would not be necessary to exemplify those cases in boxes.

The final reason is that some of the cases outlined in the boxes are typically/routinely considered in RAs of LMOs, and are thus redundant and not necessary.

I also suggest the information on protection goals and problem formulation provided in boxes to be improved. The box on protection goals could indicate that policy protection goals are often too generic and vague for use in RAs, and therefore need to be translated into operational ones (also termed assessment endpoints or specific protection goals). It could also be explained that operational protection goals have to delineate the environmental components that need protection, where and over what time period, and the maximum impact that can be tolerated. There would be added value to provide guidance on how to translate policy protection goals into operational protection goals, and to clarify how risk managers and risk assessors could interact to achieve this challenging goal. In this context, the below references provide relevant thoughts that could be considered further:

-Devos Y, Romeis J, Luttik R, Maggiore A, Perry JN, Schoonjans R, Streissl F, Tarazona JV and Brock TCM, 2015. Optimising environmental risk assessments – Accounting for biodiversity and ecosystem services helps to translate broad policy protection goals into specific operational ones for environmental risk assessments. EMBO Reports, 16, 1060–1063.
-EFSA (European Food Safety Authority), 2010. Scientific Opinion on the development of specific protection goal options for environmental risk assessment of pesticides, in particular in relation to the revision of the Guidance Documents on Aquatic and Terrestrial Ecotoxicology (SANCO/3268/2001 and SANCO/10329/2002). EFSA Journal, 8, 1–55. Available from: http://www.efsa.europa.eu/en/scdocs/doc/1821.pdf.
-EFSA (European Food Safety Authority), 2010. Report on the PPR stakeholder workshop Protection goals for environmental risk assessment of pesticide: What and where to protect? EFSA Journal, 8, 1–46. Available from: http://www.efsa.europa.eu/en/search/doc/1672.pdf.
-EFSA (European Food Safety Authority), 2014. EFSA Scientific Colloquium XIX summary report “Biodiversity as Protection Goal in Environmental Risk Assessment for EU agro-ecosystems”. Available from: http://www.efsa.europa.eu/en/supporting/doc/583e.pdf.
-EFSA (European Food Safety Authority), 2016. Guidance to define operational protection goals for environmental risk assessment at EFSA, in relation to biodiversity and ecosystem services. EFSA Journal (forthcoming).
-Garcia-Alonso M and Raybould A, 2014. Protection goals in environmental risk assessment: a practical approach. Transgenic Research, 23, 945–956.
-Gray AJ, 2012. Problem formulation in environmental risk assessment for genetically modified crops: A practitioner’s approach. Collection of Biosafety Reviews, 6, 10–65.
-Gray AJ, 2014. The policy chicken and the science egg. Has applied ecology failed the transgenic crop debate? Transgenic Research, 23, 923–932.
-Maltby L, 2013. Ecosystem services and the protection, restoration, and management of ecosystems exposed to chemical stressors. Environmental Toxicolology and Chemistry, 32, 974–983.
-Maltby LL, Duke C and van Wensem J,2016. Ecosystem services, environmental stressors and decision making. Integrated Environmental Assessment and Management (forthcoming).
-Munns WR, Rea AW, Suter GW, Martin L, Blake-Hedges L, Crk T, Davis C, Ferreira G, Jordan S, Mahoney M and Barron MG, 2015. Ecosystem services as assessment endpoints for ecological risk assessment. Integrated Environmental Assessment and Management, DOI:10.1002/ieam.1707.
-Munns WR, Poulsen V, Gala WR, Marschall SJ, Rea AW, Sorensen MT and von Stackelberg K, 2016. Ecosystem services in risk assessment and management. Integrated Environmental Assessment and Management (forthcoming).
-Nienstedt KM, Brock TCM, van Wensum J, Montforts M, Hart A, Aagaard A, Alix A, Boesten J, Bopp SK, Brown C, Capri E, Forbes V, Köpp H, Liess M, Luttik R, Maltby L, Sousa JP, Streissl F and Hardy AR, 2012. Development of a framework based on an ecosystem services approach for deriving specific protection goals for environmental risk assessment of pesticides. Science of the Total Environment, 415, 31–38.
-Olander L and Maltby L, 2014. Mainstreaming ecosystem services into decision making. Frontiers in Ecology and the Environment, 12, 539.
-Pollock CJ and Hails RS, 2014. The case for reforming the EU regulatory system for GMOs. Trends in Biotechnology, 32, 63–64.
-Sanvido O, Romeis J, Gathmann A, Gielkens M, Raybould A and Bigler F, 2012. Evaluating environmental risks of genetically modified crops – ecological harm criteria for regulatory decision-making. Environmental Science & Policy, 15, 82–91.
-SEP (Science for Environment Policy), 2015. Ecosystem Services and the Environment. In-depth Report 11. Available from: http://ec.europa.eu/environment/integration/research/newsalert/pdf/ecosystem_services_biodiversity_IR11_en.pdf.
-Suter GW, 2000. Generic assessment endpoints are needed for ecological risk assessment. Risk Analysis, 20, 173–178.
-Suter GW, Rodier DJ, Schwenk S, Troyer ME, Tyler PL, Urban DJ, Wellman MC and Wharton S, 2004. The US Environmental Protection Agency’s generic ecological assessment endpoints. Human and Ecological Risk Assessment: An International Journal, 10, 967–981.
-Wickson F, 2015. Environmental protection goals, policy & publics in the European regulation of GMOs. Ecological Economics, 108, 269–273.

The box on problem formulation is too brief and should be expanded, because problem formulation is the critical first step of the RA process that helps to: frame the process; determine the type and extent of the data that are useful in conducting RAs; and identify scientific uncertainties that may limit the assessment. Key elements of problem formulation could be outlined and explained in more detail, in line with relevant scientific literature (see below). It would also be useful to address the types of information sources that can inform problem formulation.

-Gray AJ, 2012. Problem formulation in environmental risk assessment for genetically modified crops: A practitioner’s approach. Collection of Biosafety Reviews, 6, 10–65.
-Gray AJ, 2014. The policy chicken and the science egg. Has applied ecology failed the transgenic crop debate? Transgenic Research, 23, 923–932.
-Johnson KL, Raybould A, Hudoson MD and Poppy GM, 2007. How does scientific risk assessment of GM crops fit within the wider risk analysis? Trends in Plant Science, 12, 1–5.
-Layton R, Smith J, Macdonals P, Letchumanan R, Keese P and Lema M, 2015. Building better environmental risk assessments. Frontiers in Bioengineering and Biotechnology, 3, 110.
-Raybould A, 2006. Problem formulation and hypothesis testing for environmental risk assessment of genetically modified crops. Environmental Biosafety Research, 5, 119–125.
-Raybould A, 2007. Ecological versus ecotoxicological methods for assessing the environmental risks of transgenic crops. Plant Science, 173, 589–602.
-Raybould A, 2010. Reducing uncertainty in regulatory decision-making for transgenic crops. More ecological research or clearer environmental risk assessment? GM Crops, 1, 1–7.
-Raybould A, 2013. Can science justify regulatory decisions about the cultivation of transgenic crops? Transgenic Research, 21, 691–698.
-Tepfer M, Racovita M and Craig W, 2013. Putting problem formulation at the forefront of GMO risk analysis. GM crops and Food: Biotechnology in Agriculture and the Food Chain, 4, 1–6.
-USEPA (US Environmental Protection Agency), 1998. Guidelines for ecological risk assessment. Washington (DC), USA: USEPA Risk Assessment Forum. EPA/630/R-95/002F. Available from: http://rais.ornl.gov/documents/ECOTXTBX.PDF.
-Wolt JD, Keese P, Raybould A, Fitzpatrick JW, Burachik M, Gray A, Olin SS, Schiemann J, Sears M and Wu F, 2010. Problem formulation in the environmental risk assessment for genetically modified plants. Transgenic Research, 19, 425–436.

I suggest excluding small-scale field trials (deliberate releases under part B of the EU Directive 2001/18/EC) from the scope of the GD, as the GD is not currently appropriate for these trials. The text given in the respective box acknowledges that not all considerations would apply to small-scale field trials, but does not provide more concrete guidance on which considerations would apply or not.

I am sorry to raise these critical points in a rather blunt manner, but hope they will foster further discussion. Please do not hesitate to contact me if you wish to have further clarifications. I remain available for any inquiry you may have on this matter.

Please note that the views expressed above are mine and do not necessarily represent the official position of my current employer.

With thanks,
Yann

Dear Helmut, Dear Ones-

I would like to express my genuine appreciation to the Chair and to the Subgroup of five (or is it four?) who have strived to revise the guidance based on the comments from the testing.  However, I would like to take a moment to profess my frustration in the process for reviewing these revisions to the Guidance.

For one point, during the last face-to-face meeting of the AHTEG in Brasilia, I specifically requested that the online forum be given more time to respond than has previously been allowed.  I note that we have been given two weeks to respond this time, which is indeed better than one week.  However, the greater problem is that we did not know, until the day before the discussion started, on which revisions we would be asked to comment.  I have been spending a considerable amount of time just trying to understand what changes have been made and why.  I still am formulating my response, but time is running out.  Perhaps this is why, after one whole week of the open forum, only 5 responses have been posted from the participants.  I do not think you can possibly make the claim that over 200 ‘experts’ have contributed to the discussion of the Guidance when only a handful have ever commented.  The lack of posts should certainly not be taken as a general approval.

On another point, we have been instructed to comment only on the most recent (tracked) changes in the document, which includes the new ‘boxes’ of information. 

This is problematic for at least a couple reasons:

One is that the online forum has not been afforded the opportunity to comment on all of the many other revisions that have been made to the document after MOP7 and up to this point, and in particular whether these revisions have addressed the comments submitted following the testing, or if problems with the Guidance remain.  Surely this sort of feedback from the ‘experts’ in the online forum is necessary, but I do not see that we will have this opportunity.

The other reason this poses a problem is that we are asked to comment mainly on the most recent information that has been placed in the ‘boxes’.   However, during the online discussion that took place regarding these ‘boxes’ during Feb. 1-15, a number of participants expressed concern that it is premature to work on this ‘additional’ guidance until the problems with the roadmap noted during the testing have been addressed and agreed.  I do not think we are yet at that point.   Yann Devos in his post to the current discussion [#7780] did an excellent job of capturing the reasons why the addition of these boxes is not ‘a sensible way forward’.  I completely agree.

In the spirit of cooperation, I do intend to give some ‘concrete’ proposals in response to the most recent revisions upon which we have been asked to comment.  I will share these in my next posts.

Regards,
Karen

Dear Helmut, dear all,

Also from my side I would like to express my sincere thanks to the AHTEG and the subgroup for their work. They do not have an easy task.

Like Karen Hokanson [#7781] I was somewhat confused, since I expected we could comment on the revised Guidance as a whole. There have been quite some changes since the version that was originally tested by the Parties and these changes are not limited to the text marked in tracked changes. Although we are not supposed to comment on the complete revised version, I agree with Yann Devos [#7780] with the points he brought up. The Guidance in its present form still lacks a clear focus and does not help risk assessors to ‘guide’ them through the risk assessment process. As was indicated during the testing of the previous version of the Guidance by the testers, it would have helped to include examples of real cases to show what information is need-to-know to consider in a risk assessment and how one can come from step 1 to step 5 in a focused way without drowning in all 'elements for consideration'. 

As also indicated by Ayako Yoshio [#7782] the included boxes do not help to streamline and support the process. They even increase confusion, especially since the text in the boxes is not clearly linked to the main text. If the text in boxes are decided to remain as part of the Guidance, a suggestion would be to include links in the main document that will refer the reader to the information in the boxes.

Having said this, here my main comments on the text indicated in tracked changes.

- Lines 305-312 (on difference between large-scale and small scale releases with respect to field trials): this text does not add any relevant information. Every risk assessment should be focused, on a case-by-case basis, on specific hazard pathways (if any). It does not matter if it concerns a field trial or not or what the size is (small-scale or large scale). The Guidance could be improved if it would only focus on commercial releases and not on field trials, since the purpose of field trials is different from that of commercial releases (irrespective their size).
- Lines 385-409 (centre of origin and of genetic diversity). It is unclear why this box is here. A suggestion is to include a link to this information in the text on lines 675-679, this puts it into the right context.
- Lines 643-665 (LMOs developed by RNAi-based methods). This box confuses the reader and is too detailed to be part of the main text. A suggestion is include a link to this information in the text in lines 633-638.
- Lines 726-747 (LM crops and use of herbicides). This detailed text in the section ‘potential adverse effects resulting from the LMO’ gives the impression that this one of the main adverse effects that has to be taken into account in the risk assessment. Given the fact that in most legistations potential adverse effects of herbicides are assessed by other regulation, it is strange to give it so much attention in a Guidance on LMOs. Suggestion is to leave this text out, or to include a link to this information in the text of lines 721-724.
- Lines 791-805 (human health). This text does not explain how ‘human health’ can be taken into account in the environmental risk assessment, but this text describes how food safety can be assessed. As far as I understood, the Cartagena Protocol does not cover food safety. This should be clearly stated in the text.

I hope this helps.

Kind regards,
Boet

Dear All, 

First of all my thanks to the subgroup and to the Secretariat for their undiminished energy in working on the guidance document, and to Helmut for agreeing to moderate this discussion, on top of his role as chair of the AHTEG.

While I can understand Helmut’s disappointment in between the lines of his second posting that so few people have reacted after one week, I think that the analyses of Karen Hokanson and Boet Glandorf that this process is not easy, hits the nail on the head. I endorse Karen’s observation that one cannot possibly make the claim that over 200 ‘experts’ have contributed to the discussion of the Guidance when only a handful have ever commented in substance.  Sharing also the other concerns raised by Ayako Yoshio and Boet Glandorf, I believe that this process of commenting on a growing number of interrelated documents that are in different stages of development has become so difficult that perhaps the MOP would be well advised review the process.

Like Patricia Gadaleta, I managed to finish a first reading, and I share for now some general observations, and plan to come back with more specific comments later this week.

First, I whole heartedly share the concern expressed by Yann Devos and supported by Boet Glandorf, that in its present form, the GD is mostly an extensive list of possible points to consider, but without any guidance or examples as to which considerations would be relevant in which type of cases, and how these considerations can be implemented in practice. As Yann said, the GD is not practical.

Second, in line with the previous observation, I support the comments by Patricia that much of the information in the GD will not be relevant for the risk assessment for small-scale, confined releases. The GD should make this very clear on several places and furthermore - in line with the suggestion of Yann - it should make clear that it does not apply to small-scale confined field trials.

Third, the text now mixes protection goals of very different levels, ranging from overall protection goals such as “conservation and sustainable use of biodiversity” (which is in fact the scope of the risk assessment) to operational protection goals and related assessment end points, to be considered in case specific risks have been identified.

Fourth, I agree with Ayako and Boet about the text boxes.

Regards to all and working on more detailed comments

Piet

Dear colleagues,

thank you for taking up my suggestion to intensify the discussion in the online forum in this important discussion on the revised version of the Draft Guidance.

Reflecting on the postings over the last couple of days, in particular those by Yann Devos (7780), Karen Hokanson (7781), Ayako Yoshio (7782), Boet Glandorf (7783) and Piet van der Meer (7784), as moderator of this discussion I would like to share with all of you the following explanations and observations:

I fully understand that the task in front of us is challenging as it addresses the whole revised Guidance document after an intensive period of work by the sub-group members, who have taken into account all the relevant tasks resulting from the previous AHTEG meeting and the subsequent online discussions. I therefore thank you very much for your efforts during this 2-weeks period of online discussion. Let me stress that it is for ease of reference, that the changes introduced since the last AHTEG meeting are marked in track changes. I would also like to stress that the Online Forum is not being asked to comment on changes already agreed by the AHTEG. We have to take stock of already achieved results and move forward on the outstanding issues in order to avoid going in circles.

The bulk of the changes relate to the introduction of text boxes on four topics as agreed upon at the AHTEG meeting. A few additional changes to the text relate to requests from the testing of the Guidance, including the introduction of examples, more details to emphasize the difference in the information required for field trials and commercial releases, a definition of uncertainty from the US EPA to replace a sentence stating that there is no agreed definition of uncertainty, and editorial changes to improve the text.

Taken together, the changes on which feedback is being sought add up to a manageable amount of text. You as participants are as a whole given 18 days since the document was made available to consider the changes and provide comments. I acknowledge the fact that participation in the forum requires time and dedication while many of us are extremely busy with other tasks, but I think that 18 days seems an appropriate amount of time for providing comments.

I recognize that most of the recent posts relate to procedural issues. While these views are duly noted and will be submitted for consideration by the COP-MOP, I think it would be very useful to hear comments related to substance and how the text could be improved. I want to voice many thanks to those of you who have already provided a few concrete suggestions.

A few issues were highlighted as still lacking in the Guidance. For example: its overall objectives, the target audience, key messages to convey, the level of detail given to specific risk assessment considerations, and its structure/organisation. Concrete text proposals to address these and any other remaining gaps in the Guidance would be very useful and I invite you to make suggestions along those lines as well.

I also note that some participants are of the view that it is premature and not a sensible way forward to develop text boxes on additional topics (human health, centres of origin, etc). These views will be forwarded to the AHTEG for their critical consideration along with text proposals drafted in response to those who wish to have such text. Ultimately, the AHTEG will make recommendations to the COP-MOP on how to reconcile the different views. In this context I want to emphasize that, since a discussion held beginning of this year, I recognize that a number of participants are against the introduction of these text boxes. However, the reason why the sub-group went ahead with the development of draft texts for these boxes is because the AHTEG agreed to do that in response to the request in decision BS-VII/12, and other participants in the online forum supported their development.  Again, if it is not possible to reconcile the different views at the level of the online forum, all concerns will be submitted to the AHTEG along with the text proposals for their consideration at its next meeting in July this year.

I hope that I could clarify most of the points and provide some guidance for the remaining discussion. I look forward to more concrete text proposals as well as to reading posts from other participants.

Thank you and best wishes

Helmut Gaugitsch

Dear colleagues,

Following up on my earlier post with some general observations on process and content, I have tried to go through the text in the way suggested by Helmut, i.e. to only look at the changes introduced since the last AHTEG meeting, and not to comment on changes already agreed by the AHTEG.

I found this very difficult to do, in part because the contributions from some AHTEG members suggest that the suggested agreement on text may not be as firm as it was presented, and in part because newly inserted text may have all kinds of combinatorial, cumulative, antagonistic, additive or synergistic effects on existing tex.

With that perspective, I share the following additional observations:

1) support the comment of Ayako Yoshio and others that “Centers of origin and centers of genetic diversity” (para.385-409), adds nothing else but “it is important area for diversity”

2) support the comments of Luciana Ambrozevicus and others that the attempt to incorporate every possible topic in this guidance is disrupting the process

3) support the concern that the Use of Terms section was never part of the on line discussion.

4) support the changes of the title proposed by Karen Hokanson

5) support the comment that the Cartagena Protocol does not cover food safety

6) disagree with Ossama’s comment that the guidance document introduces basic concepts of risk assessment ; the basic concepts are the ones in Annex III and it is up to the MOP to introduce more, if need be.

7) support the comment by Patricia Gadaleta taht the text doesn’t reflex the relevant outcomes from the testing

8) support the comment by Sol Ortiz García that the AHTEG has not provided a justification for the suggestions that were not taken on board, neither a justification on the comments and suggestions taken on board that modified previous texts

9) support the comment by Monica Garcia-Alonso and others on the box on LM crops and the use of herbicides

10) support the comment by Hector Quemada that the box on RNAi offers nothing new that is not already considered in the assessment of other types of LMOs.

11) support the comment by Maria Mercedes Roca, that it may be appropriate to note the disagreement on what the guidance should be, and how useful and practical it is

12) support the comment by Felicity Keiper that the document is difficult to read as it continues to grow larger, more detailed, more repetitive, and difficult to follow, even for experienced risk assessors. We have reached the point that we need a  GPS for the roadmap. Support also Felicity’s suggestion for Lines 1041-1051

13) support the comments by Andrew Roberts, and in particular that changes are made from one version to the next, sometimes with little or no explanation and frequently not in keeping with suggestions made in the online forum. 

Finally, I believe that the shovel analogy that Karen Hokanson presented hits the nail on the head. The same goes for Paul Keese’s observation that “it is inevitable that our bias blind spot finds fault in the work of others with far more alacrity than the critiquing of our own efforts”.

As Karen said, the real question here is whether 'some guidance is better than no guidance', or whether  'no guidance is better than bad guidance'. 

Wishing everyone a good night!

Piet

Dear Helmut and other colleagues,

Thank you for providing the improved version and giving me this opportunity to share my points of view. I strongly believe that it is too difficult to track all the changes from the previous version in this short period as pointed out by Karen [#7781]. I also agree with the comment from Yann [# 7780] that we should first discuss about efficacy of adding boxes.

First, to avoid confusion of readers, “text boxes” should be used in a coherent manner throughout the document. In this version, some of them are used for explanation of important words (para. 196-221 and 451-459), for tips (para. 313-326, 385-409, 643-665, 726-759 and 791-805) or examples of particular situation (para.584-603 and 680-690). Such a lack of uniformity makes this document more confusing.

Second, regarding “LM crops and the use herbicides” (para.726-759) as well as para.474-477 and “Risk assessment of LMOs and human health” (para.791-805), I would like to draw your attention to that the lots of comments posted in the previous forum state that there is other legislative framework either for risk assessment of pesticide (including the change of using way/amount caused by introduction of LMOs) or for safety assessment for human health. Therefore, I think, these issues should be addressed in accordance with the existing relevant regulatory frameworks. 

Third, regarding “Centers of origin and centers of genetic diversity” (para.385-409), it seems that there is no specific consideration points rather than “it is important area for diversity”. Therefore, it should be reconsider whether these texts are necessary in addition to the (i) of elements for consideration of step1 (para.678-679).

Fourth, regarding “Types of adverse effects” (para.584-603), it should be left to the “use of term section”. Furthermore, we haven’t reach a consensus on the definition of each term. I also point out that the terms “immediate effect” and “delayed effect” are rarely used in the document, so it’s meaningless to explain the definition of the terms there.

Fifth, the redundancy should be avoided. With a short glance, I acknowledged the followings;
Most concepts in the box of “RNAi-based methods” (para. 643-665) are already covered in -“Elements for consideration regarding characterization of the LMO,” and it just makes the GD redundant and complex to repeat these concepts in the box. I think just consideration about bioinformatics and ‘omics’ technologies (para 662-665) could be new information for RNAi., and
-"Elements for consideration” in each step are not well-organized. They seem just random lists, and some of them are overlapped or not related to the steps. For example, many elements for consideration in Step 2, such as “information on the location of the release and the receiving environment” (para 816-817), should be considered and are already covered in Step 1.

Last but not least, it is undesirable situation that the results of discussion at previous on-line forum have not been reflected in this latest version of the guidance. I would like to ask AHTEG to equally consider all of the comments posted in every related one-line forum, and secretariat to conduct related works in a more stepwise manner.

Best regards,
Ayako

Dear participants,

First I would to say thank you for the AHTEG members and the chair for this new Guidance version. Although I appreciate the work I have to say that the outcome is not we expected after so many years. I don´t know but I have the impression that “the attempt” to incorporate every possible topic in this guidance is really disrupting the process. And I think is even worse because the incorporated topics were choosen based on the opinion of a few on line forum participants and those are not representing Parties positions.

Having said that I would like to agree with some general points raised by other participants:

- it should be an opportunity for the on line forum to comment on the revisions made in the document after MOP 7 (#7781),  that actually was the main task and part of the TOR in the Decision BS-VII/12

- there is a lack of efficacy of adding boxes to support RA for the reasons already well explained by others (#7780, #7782, #7783)

- the Use of Terms section was defined by the AHTEG and was never submited to the on line discussion. Due to the importance of this terms it should be discussed in the on line forum.

Some specific suggestions regardind the text indicated in tracked changes:

Title: totally agree with the post # 7789 and I suggest the title “Guidance on Risk Assessment of Living Modified Organism” (monitoring is presented in a specific guidance as much as other subjects while the RoadMap is about Risk Assessment)

Line 196-221 and Line 451-459: all the important terms should be defined in the Use of Terms section. A paragraph should be included explaining  in details the problem formulation (the references in the #7780 could be used for that) as this is the first and critical step for RA. The box could be used only to present examples of how protection goals and assessment endpoints are used to inform problem formulation.

Line 309-312 and Line 313-326: in the current form I do not agree that the Guidance can be used for field trial or experimental release RA as well explained by others (#7776). It should be clear in the Guidance that it can only be used for commercial release RA. 

Line 333-335: Although a sentence from EPA about Uncertainty was included I understand that the question raised by many countries is still a problem in this section. The text is still misleading impression that all uncertainties are relevant and should be dealt with instead of only the uncertainty that impact the overall risk and affect decision making.

Line 385-409: the inclusion of a box describing center of origin and of genetic diversity did not present any further or practical information beyond what is already described in the Guidance. All the points to consider in this box are already mentioned in different parts of the Guidance related with likely potential receiving environment. If the text is kept I think it should not be presented in a box and I agree with Boet that it should include a link in the text lines 674-679 (#7783).

Line 510-514: the paragraph should be deleted, I do not consider appropriate to make considerations about future technique applications and the appropriate comparator without scientific basis.

Line 584-603: all the terms should be defined in the Use of Terms section. Since the first round of Guidance test we do not agree with the use of terms such as cumullative and combinatorial effects in the LMO RA context without an agreement among Parties. 

Line 643-665: the box should be deteletd as all the Points to Consider are already in the Guidance. The sentence “Bioinformatics and ‘omics’ technologies may be useful in assessing the homology of the transformed genetic elements in the LMOs using RNAi” does not add any value for the reasons already presented by other (#7797, #7801)

Line 726-747: “LM crops and use of herbicides” – use of herbicide and their synergistic interactions it´s not part of LMO risk assessment and it´s an aspect considered under other legal framework for pesticide approval. This box has two mains points to consider, both already mentioned in the Guidance. Due to those considerations the box should be deleted.

Line 791-805: Cartagena Protocol does not cover food safety, CODEX is the specific body for that and it should be kept in their scope. I agree with other reasons well presented in the post #7801.

I would also like to agree and emphasize the post #7788 - although I also focus my comments in the instructions for this discussion it does not mean, by any way, that I agree with all the other revisions (or the absence of that as the case for categories B and D). I also like to clear express the fact that those categories should be incorporated during the guidance revision as those were important points presented by Parties during the test phase.

I also do not agree with any other specific guidance (GM fish and Synthetic Biology) before Parties agree about the necessity of those guidances. In the SBSTTA 20 last week  the brazilian delegation express in the contact group in a very clear way that the SynBio guidance is not necessary at this moment as there was no technical justification for this guidance  (actually we do not have even a definition for SynBio as it was put in brackets) and the mention to the guidance was deleted from the SBSTT

I´m afraid that after so many intervations in this on line forum that are not being considered along so many years, after all the time and resources we spent testing the RoadMap without the incorporation of our main suggestions, after having an agreement with compromisse solutions and those words being used to conduct the process  in a biased way, we are very distant from the balanced tune necessary for the consensus decisions in the COP-MOP.

Best regards,
Luciana P. Ambrozevicius / Ministry of Agriculture - Brasil

Dear all,
I would also like to thank the AHTEG and the subgroup for the work they have done, to address all comments submitted and to improve the Guidance in many aspects.

I support Yan Devos [#7780] on his comments made for problem formulation and protection goals. Problem formulation is only briefly explained in the box. Since this is the first and critical step in ERA, efforts should be made to improve this part. It would also be useful for risk managers and risk assessors to be provided with a more explicit view of how protection goals are linked to specific measurement end points and how limits of concern are set.

Regarding field trials, I would suggest like Boet Glandorf [#7780] and Yan Devos [#7780] excluding them from the scope of this Guidance.

Kind regards,
Maria

Dear Helmut and All,

Thank you to Helmut for his clarification [post #7785] on the points in my earlier post [#7781]. 

I still am of the opinion that the Forum should have the opportunity to respond to ALL of the revisions to the Guidance based on a more transparent justification for how it has been changed - or not as the case may be - and say whether these changes have in fact addressed the concerns that were raised in the many comments that were submitted as a result of the testing of the Guidance.  This would not be going in circles, which is the concern expressed by Helmut.  In my opinion, this would bring the discussion to a logical end.

The Subgroup of 5 (or is it 4?) has worked very hard to make changes in the document based on ‘some’ of the comments.  Their changes have mainly been in response to the comments in the category of ‘Suggestions for substantive changes to specific sections of the Guidance’ (Category E).  However, it is not at all clear how even many of these comments in this category were addressed, and why some were not addressed at all. 

Furthermore, the many comments that were in the categories of ‘Overall evaluation of the Guidance (Category B)’ or ‘Suggestions for substantive changes without a specified location in the Guidance (Category D)’ have been left largely unaddressed, or it is not at all clear how they were addressed.  Yet, the most important critiques of the Guidance can be found among the comments in these last two categories.

I also respectfully disagree with Helmut’s assertion that 18 days is enough time to respond to the many revisions of the Guidance with any level of attentiveness.  It is not even enough time for most of us to respond to the few most recent revisions that the Forum has only been invited to review in this discussion, as evidenced by the lack of participation in this discussion with only 5 days to go. 

I agree with the post of Piet van der Meer [#7784] to suggest that perhaps the MOP would be well advised to review the process, and the others who have expressed concern with the process. 

However, in the spirit of cooperation, I do still intend to submit some ‘concrete’ suggestions on the revisions to the Guidance – if I can find the time.  I will try to focus on the most recent revisions per the instructions for this discussion, but I may stray here and there into ‘other’ revisions to which we have not been asked to respond.   Please also note that because I am only responding to these points at this moment does NOT mean that I therefore ‘agree’ with all of the other revisions to which I am not responding.  (Absence of a response does not indicate a level of agreement.)

Thanks,
Karen

Dear all,

Here, a concrete text proposal:

TITLE: Specifically the addition of the word ‘MONITORING’ in the title. 

The word ‘MONITORING’ does not belong in the title of this document.  The mandate was never to develop guidance on risk assessment and monitoring. The core of the Guidance is the ‘Roadmap’.  The section on monitoring (Part III) is ‘additional guidance’.  The AHTEG should concentrate on revising the ‘Roadmap’ before it turns to the ‘additional guidance’.   Work on ‘additional guidance’ is premature.  There is no agreement on what is in Part III on Monitoring, including whether it is consistent with the protocol or with the many years of experience of monitoring within the context of risk assessment of LMOs.   This has been stated many, many times in our discussions.

Therefore the TITLE of the document should be: ‘GUIDANCE ON RISK ASSESSMENT OF LIVING MODFIFIED ORGANISMS’, as it was before. 

Or perhaps: ‘GUIDANCE ON RISK ASSESSMENT AND RISK MANAGEMENT OF LIVING MODIFIED ORGANISMS’.

The TITLE should NOT be: ‘GUIDANCE ON RISK ASSESSMENT OF LIVING MODIFIED ORGANISMS AND MONITORING WITHIN THE CONTEXT OF RISK ASSESSMENT’.  That’s just wrong.

[This is for Chan:  It is true that the word ‘Monitoring’ occurs 129 times in the Guidance document, but 100 of those occurrences are in the ‘additional guidance’ section specifically on ‘monitoring’ (PartIII).  The word ‘Monitoring’ only occurs 29 times in the whole rest of the document including only 15 times in the ‘Roadmap’. 

However, the word ‘Management ‘occurs 69 times in the whole document including 59 times outside of Part III, and including 35 times in the ‘Roadmap’ of the document. Therefore, the word ‘Management’ is clearly more suited for the title of the Guidance than ‘Monitoring’, if anything. ]

Thanks,
Karen

Dear All,

Here, another concrete text proposal, regarding field trials:

LINES 314-327 - should be deleted and replaced by the statement:
‘This Guidance on Risk Assessment of LMOs does not apply to field trials or experimental release.’ Or something along those lines.

The focus of this guidance should be solely on commercial/general releases.  This point has been made many, many times over the years that we have been working on this guidance. 

It has already been made many times in the limited posts to the current discussion, including by Patricia Gadaleta [#7776], Yann Devos [#7780], Boet Glandorf [#7783], Piet van der Meer [#7784], and Maria Kammenou [#7787].

Thanks,
Karen

Dear all

I have been following the discussions closely and would like to highlight the following points:

- I respectfully disagree with the claim that the AHTEG’s mandate was never to develop guidance on risk assessment and monitoring and I quote the following text from the annex of the COP-MOP decision BS-IV/11 describing the terms of references of the first AHTEG
“(ii) Taking into consideration the identified need for further guidance on specific aspects of risk assessment, including particular types of (i) living modified organisms (for example, fish, invertebrates, trees, pharmaplants and algae); (ii) introduced traits; and (iii) receiving environments, as well as monitoring of the long-term effects of living modified organisms released in the environment, prioritize the need for further guidance on specific aspects of risk assessment and define which such aspects should be addressed first, taking also into account the need for and relevance of such guidance, and availability of scientific information;
(iii) Define an action plan to produce, prior to the second meeting of the Group, modalities for development of the guidance documents on the specific aspects that were identified as priorities and for testing of the roadmap. This action plan should include the details of a process for monitoring and reviewing the progress in each of the specific aspects;”
It is quite clear that the issue of monitoring was at the heart of the AHTEG mandate from the beginning.
The AHTEG started working on a primary guidance document covering some of the important topics and monitoring was among the topics that was on the list for next phase. And of course since the primary document did not include a monitoring section the title did not include the word monitoring. At a later stage, the monitoring part was developed then this was reflected in the title.
Having said that, I oppose the deletion of the word monitoring from the title.

2. The guidance document is already at a very advanced stage and has been tested and re-tested several times over years. The process even involved more than one ATHEG. In the last testing, the guidance document received a very high rating. However, as with any other document that undergoes  a review process there are and there will always be comments that is why The COP-MOP in its decision BS-VII/12 established the methodology and the mechanism on how to take on board the comments of the final testing. As mandated by the COPMOP this is the methodology to be followed:
“Methodology
1. Taking into account the results of the testing process, established in decision BS-VI/12, the Guidance on Risk Assessment of LMOs shall be revised and improved in accordance with the following mechanism:
(a)After the seventh meeting of the COP-MOP, the Secretariat will group the original comments provided through the testing of the Guidance. The grouping will be done in the form of a matrix based on the following categories: statements that do not trigger changes; editorial and translational changes; suggestions for changes without a specified location in the Guidance; and suggestions for changes to specific sections of the Guidance (sorted by line numbers);
(b)The AHTEG shall review the grouping of comments done by the Secretariat and work on the suggestions for changes;
(c)The AHTEG shall streamline the comments by identifying which suggestions may be taken on board and providing justification for those suggestions that may not be taken on board. The AHTEG will also provide concrete text proposals for the suggestions to be taken on board with a justification where the original suggestion was modified;
(d)The Open-ended Online Forum and the AHTEG shall subsequently review all comments and suggestions with a view to having an improved version of the Guidance for consideration by the COP-MOP at its eighth meeting.”

I believe that any proposal to advise the COP-MOP to review the proposed review process is just a thinly disguised attempt at keeping the guidance under a non-ending revision process. This undermines developing country governments needing guidance to establish and practice quality risk assessment to meet their obligations under the Protocol. Those government who, in the last testing, found the Guidance so useful and are waiting for it to be adopted.

3. It is well highlighted that the guidance document introduces basic concepts of risk assessment rather than providing detailed guidance for individual case-specific risk assessments. And that the amount and type of information available and needed to support risk assessments of the different types of intentional release into the environment will vary from case to case. Having said that I believe that the guidance document can be used and is of a great value for all types of environmental releases of LMOs, including those of limited duration and scale such as field trials or experimental release.

Finally I would like to thank the chair for his able leadership and the subworking group for their dedication and commitment.

O.A.El-Kawy

Dear all,

Here a couple more concrete text proposals, regarding ‘uncertainty’:

LINES 329-382:  This entire section on ‘uncertainty’ should be DELETED from the document, as there has been significant disagreement about whether this is consistent with the protocol or with the years of experience with risk assessment of LMOs.

At the very least, this entire section should be moved to a ‘box’ (Although it is not clear why something goes in a box, as expressed very well by Ayako Yoshio in her post [#7782]).

If this section on uncertainty is placed in a ‘box’, it should at the very least, make this concrete change in the text:

LINES 334-335:  The sentence

--‘There are no internationally agreed guidelines to determine ‘scientific uncertainty’, nor are there internationally agreed general rules or guidelines to determine its occurrence”

which has been deleted in the current version, should be reinserted.

The sentences that have been inserted in the place of the sentence that was deleted,

--“Uncertainty may be defined as lack of knowledge and concerning an event, state, model, or parameter.  Uncertainty, unlike variability, may be reduced by research or observation.“

and the addition of the reference to ‘United States Environmental Protection Agency (http://www.epa.gov/caddisindex.html)’, does not eliminate the need to include the sentence that was deleted.

I believe Ossama AbdelKawy in his post [#7775] also had problems with the inclusion of this reference. This reference to the USEPA ‘Causal Analysis/Diagnosis Decision Information System’ does not seem appropriate here. I am not familiar with the CADDIS document, but I did look it up.  It does have an extensive section on data analysis.  I perused the section on data analysis and, while it is a thorough treatment of basic statistics, I could not find a definition of “uncertainty”.  Of course, there is a discussion of statistical analysis which implies putting confidence intervals around a value based on probability and assessing results and causality within a statistical framework.  But I do not think this represents how the USEPA treats uncertainty within the context of risk assessment of LMOs.  Certainly this CADDIS does not represent an “internationally agreed guideline to determine ‘scientific uncertainty’…”

Therefore, the deleted sentence should be reinserted. And probably this reference to CADDIS should be deleted because it will only add to the confusion about uncertainty.

Thanks,
Karen

Dear Helmut, Secretariat and fellow forum participants

I want to thank the sub-group and Chair for their hard and good work preparing suggested revisions. I also want to acknowledge the important interventions that have come before mine.

I would like to register the following concrete text proposals. I anticipate posting more later.

1. Box Protection Goals
This always includes human health.
“‘Assessment endpoints’ define, in operational terms, the environmental and human health values that…”

‘Measurement endpoints’ is [DEL: a quantifiable] an observable indicator of change..."
Observation does not always rely on quantification. Eg, many colour indicators of gene expression are simply on or off.

2. I support the deletion of the sentence: There are no internationally agreed guidelines to determine “scientific uncertainty”, nor are there internationally agreed general rules or guidelines to determine its occurrence.

The statement adds nothing to the guidance. There are many things for which there are no international rules or guidelines. This alone does not inform how a State might choose to conduct its risk assessment. The AHTEG Guidance provides a definition for a concept that is part of any risk assessment framework. Do we agree on that definition? If so, then that is the point. The text immediately following the definition satisfactorily alerts the reader to a diversity of opinion in how to handle uncertainty.

3. I support this modification to the text: “Considerations and communication of uncertainty may improve the understanding of the outcomes of a risk assessment, strengthen the scientific validity of a risk assessment and provide transparency in the decision making process.”

4. A small grammatical suggestion (hypotheses for hypothesis) in this sentence: After identifying the potential adverse effects, conceptual models are developed as working hypotheses to describe how the LMO may have adverse effects on the assessment endpoints.

5. The box on 'RNAi'
i. LMOs developed through RNAi-based methods [DEL: to reduce gene expression]. Rationale: It may be used for any number of reasons and sometimes results in up-regulation of genes. In pesticide applications, it may not be used to reduce gene expression in the LMO.
ii. Revised text:  RNA interference (RNAi) refers to a process that alters gene expression and is triggered by double-stranded RNA (dsRNA) molecule(s). The dsRNA is commonly processed into small interfering RNAs (siRNAs). The smaller dsRNA molecules then are converted into guides for protein complexes that bind to particular sequences of messenger RNA (mRNA), leading to either the degradation of mRNA or the inhibition of mRNA translation into protein, or bind to DNA sequences resulting in modification of genes (methylation) that usually inhibits transcription.
iii. Revised text: Several LMOs have been developed using RNAi, or engineered to produce dsRNA to induce RNAi. The molecular characterization of these existing LMOs during a risk assessment may include: 1) specificity of the molecules as predicted from the sequence(s) of the DNA used to generate the novel dsRNA in either the LMO or organisms that may be exposed to the LMO; 2) length (size) and amount of the dsRNA that may cause adverse effects to other organisms; 3) the stability of the expression and function of the modified elements; 4) expression level and target and non-target locations (tissues, organs); 5) effects on target and off-target genes; 6) effects on target and other organisms that may be exposed to the dsRNA or its derivatives.
iv. Revised text: An assessment of the interaction of LMOs using RNAi, or engineered to produce dsRNA with the receiving environment and organisms in the receiving environment may include considerations of the following: 1) the effects of changed environmental conditions; 2) horizontal transfer of the modified genetic elements (including the DNA used to create the dsRNA) and the(ir) products to other organisms; 3) exposure to other non-target organisms through persistence in natural media (e.g. soils) and food webs, that resulted in non-target effects, e.g. epigenetic inheritance.
v. Revised text: Bioinformatics and ‘omics’ techniques may be useful in making predictions based on the sequence similarities of the intended novel nucleic acids in the LMOs, or engineered to produce dsRNA in other organisms. These techniques may be used to monitor the expression of the modified genetic elements in target and non-target organisms and compare that expression with target and off-target genes. However, bioinformatics at this time has not proven effective at predicting off-target effects and thus should be supplemented with other techniques and experimental approaches.

Best regards
Jack

Dear All.

First, thanks to Jack.  I know that you were not at the last face-to-face meeting of the AHTEG, so I want to share a few more thoughts on the sentence that I support reinserting in the Guidance:
“‘There are no internationally agreed guidelines to determine ‘scientific uncertainty’, nor are there internationally agreed general rules or guidelines to determine its occurrence.”

Really, I am surprised to see this sentence deleted from the Guidance in this version.  I distinctly remember discussing this sentence at the AHTEG face-to-face meeting because the subgroup had first elected to delete this sentence at that time, based on a comment that was submitted as part of the testing process from Norway (Comment R266).

Some members of the AHTEG agreed then that this sentence should NOT be deleted, and I noted in the version of the revised Guidance with the report from the AHTEG meeting (Annex 2) that this sentence had then been reinserted as requested.

Why delete it now???

Actually, I think we could have responded to Norway’s comment suggesting deletion of this sentence in the first place by saying:

‘No changes. The current text is the result of lengthy discussions and represents a delicate balance between different points of view.’ 

This is the justification that was provided by the subgroup for quite a few other comments in Category E where they elected to make ‘no changes’.  The AHTEG never discussed any of these comments where the Subgroup elected to make ‘no change’.

In fact, this was the justification given by the Subgroup for ‘no change’ to an original comment (R224) from Malaysia on this same section on uncertainty in the roadmap, but the Subgroup elected for ‘no change’ to that ‘comment’ from Malaysia, and the AHTEG did not discuss it at the face-to-face meeting.

Comment R224 from Malaysia read:
“Line 266-274 [in the originally tested version of the Guidance] Identification and Consideration of Uncertainty – Rather lengthy explanation of uncertainty.  Suggest putting as appendix.”

This is my idea too – put it in a box!  Why not??

This does illustrate the difficulty in this process we are going through, doesn’t it?  It is very hard to follow which comments were accepted and which were rejected, and it is also hard to follow which changes were ‘agreed’ to by the AHTEG so that we will not be revisiting them – and which we will still revisit.

Thanks,
Karen

Dear All,

Thank you very much for the opportunity to contribute to the revision of the guidance.

I am agree with Ms. Boet Glandorf, [#7783], where she refers to the box “LM crops and the use of herbicides” (lines 746-780), leaving this text out. Currently the risk assesment of plaguicides is executed and take in count by other regulations and agencies, so the sugestion, like Ms. Boet mention is to leave this text out or consider this possibility just when the herbicide dosis of aplication in the LM crop would be greater than indicated in the herbicide´s label.

Another point is in reference to lines 406-407 that mention that any adverse effects identified in the centers of origin and genetic diversity will be considered significant. Therefore, I suggest include in the text that gene flow is not per se a significant adverse effect and the risk assesment of the consecuences of the gene flow must be   analized.

Best regards,
Rafael Romero

Dear Colleagues

Here is my second contribution of concrete text suggestions, following on from those posted earlier.

1. Box on herbicides. I remain an advocate for inclusion. Specific suggestions:
i. Minor grammatical - LM crops and the use of herbicides 
ii Minor grammatical -  In the case of LM crops that are resistant to herbicides...
iii. Suggest deletion of 'synergistic' in this sentence: "Furthermore, the use of multiple herbicides in the same area, applied either simultaneously or in sequence, may result in synergistic effects that are more toxic..."

[synergy means different things depending on context. In some contexts, synergy is >> than additive and I don't think we mean to exclude additive effects]

iv. Revise: The risk assessment of LM crops may also include considerations of potential consequences arising from the use of multiple herbicides and that are relevant to assessing  effects from herbicide mixtures. In this context, potential effects of multiple herbicides may be taken into account during the assessment...
v. Revise: Effects on non-target organisms may be different due to effects of mixtures of herbicides and additional studies may be needed to both identify and assess those risks;
vi. Revise: Effects on biodiversity may be different due to mixture effects e.g. changes in weed populations, composition and diversity.
vii. Revise: Detailed information on agricultural practices and the herbicide regime that will be applied along with the cultivation of the LM crop with herbicide resistance are needed in order to identify the differences in relation to conventional practices and to identify possible mixture effects.
viii. Revise: What is known of their fate and behaviour in the environment and are there effects amplified by mixtures?

2. Box on human health
i. Revise: The kind of experimental studies needed to assess potential toxicity and allergenicity is determined on a case-by-case basis depending on the nature of the product(s) synthesized by the transgene(s). For statistical rigor, the number of such studies may also vary according to the case. In some cases...
ii. Insert new last paragraph: There may be indirect effects on human health that are not caused by toxins or allergens. For example, people may be exposed to potential microbial pathogens that were on the LMO when it was sprayed with a herbicide. Effects of the combination of LMO trait and management might have implications for treating an infection, should it occur.

Best regards to all
Jack

Dear Colleagues,

I am sure that the sub-group members have had an intensive period of work before this new version of Guidance document has been arisen. I would like to give my acknowledgment to their work; however, as far as I am concerned the text doesn’t fully reflex the relevant outcomes from the testing, the previous AHTEG meeting and the subsequent online discussions.
I find that in this round of discussion there are two tasks to accomplish in this new version. One is to revise the general changes in the text and the current boxes with gathered views, information and references given in the testing and its analysis in categories, particularly those that haven’t been taken into account yet. In my humble opinion this should be done first, in order to complete the task of improving the Guidance. The second task is to see how the subgroups have taken into account the outcome of the discussions regarding the four topics raised of the analysis of issues that the Parties expressed to need further guidance, hold 1 – 15 February 2016. The AHTEG agreed, with reserve of some members, that some topics could be incorporated into the current Guidance as boxes or texts depending on the issue, the relevance for improving the Guidance and the knowledge so far, mainly if these topics are applicable to the environmental risk assessment of LMOs.  It is highly arguable how the topics were assembled and also the meaning and significance for the environmental risk assessment of some of them.
Despite the good intention and hard work of the subgroup, I have found the new text in the boxes premature, trying to put some text for each topic without a justification or taking into account if the topic is relevant or not for this Guidance, as it was discussed in the last forum, adding confusion to the existent text of the Guidance still in discussion.

I agree with other people that it is difficult to track all the changes from the previous version, and also deal with the analysis of the new boxes. I am spending time of my duty on this task but I find distractive to analyze and give some feedback on the new boxes when the main text of the Guidance doesn’t fit very well with the current way to perform the environmental risk assessment. As an example, I have given feedback in my first post about the box on field trial or small scale regulated releases and the text related to this issue in the Guidance, but I don´t feel confident with the text, and I have also asked to other colleagues in my office that work as regulator in this issue that agree with me, that the Guidance doesn’t give clear tools for this kind of regulated releases. I would like to hear other Parties with experience in field trial their voices.

With regard to the new boxes, I would like to express my feedback about the text into the boxes, and also about the rationality of inclusion or not of the boxes of each merged topics:

Particularly for “Synergistic impacts of different herbicides that are part of the technology package that accompanies certain LMOs¨ I think that the assessment of this issue is performed by a different authorities in many countries, using international standards produced by international bodies. The information into this box is very because addresses to questions that are out of the scope of the Guidance. In my opinion there is no need to further text or boxes for this issue.

Conserning to “Nutritionally altered living modified plants” as was pointed out in the last discussion on February and also was expressed; it is unlikely that they pose special considerations for the environment different to other traits since they are specifically developed for nutritional or health benefits. So, I believe that the current Guidance for the risk assessment of LMOs can cover the risk assessment of nutritionally altered plants without having to add specific information in the Guidance. So I suggest taking out the box because, if we incorporate a box for each new trait in the Guidance, we will be losing the focus on a wide applicable Guidance and Roadmap. The same scenario is for “LMOs that produce pharmaceutical products” as well as “LMOs created through use of dsRNA techniques, engineered to produce dsRNA or dsRNA” and “LMOs containing RNAi”, these are specific types of LMO, so in my opinion adding more examples of specific LMOs to the guidance in ‘boxes’ suggests that each type of molecular product or mechanisms that give a new trait need its own specific risk assessment. This is very confusing taking into account that in previous boxes, the Guidance addresses to a general accepted way to tackle the ERA following a problem formulation procedure in a case by case basis. On the one hand the text in the box about LMOs developed through RNAi-based methods to reduce gene expression (line 643-665) is very informative about what RNAi technology, and doesn´t say anything new about how to tackle the environmental risk assessment. On the other hand, the utility of ‘omics’ technologies in environmental risk assessment is very controversial (line 662).

Regarding “LMOs introduced in centers of origin and genetic diversity” and “LMOs intended for introduction into unmanaged ecosystems”, I believe that in the previous discussion, it has been pointed out that this could be addressed by asking relevant questions about the receiving environment (such as presence of wild relatives, probability of gene flow, etc.) in the problem formulation process. These facts have been already considered in other part of the text of the Guidance. Then, I consider that the box position (line 385) is not appropriated. I suggest moving only the two last paragraphs (line 402-409) of the text into the box as a bullet in ¨Elements for consideration regarding the intended use and the likely potential receiving environment:¨on line 667.

I am attaching the Guidance document with some comments and observation already expressed and some additional ones.

Thank you for the opportunity to comment in this round of discussion.
Best regards,
Patricia

Dear Helmut and All,
Thank you for the efforts made on the revision of the guidance.
I want to acknowledge all colleagues contributing to this interesting discussion.
I like to state that I agree with previous comments:#7780, #7781, #7782, #7783, #7784, #7788, #7797 in relation to the reviewed process of the guidance and the need of a more transparent mechanism on how the comments derived by the testing process and the previous online forum were addressed, and why some comments were not addressed at all.
As recalled by post #7791, the COP in decision BS- VI/12, included in the methodology, among other things a mandate to the AHTEG that states:
“(c) The AHTEG shall streamline the comments by identifying which suggestions may be taken on board and providing justification for those suggestions that may not be taken on board. The AHTEG will also provide concrete text proposals for the suggestions to be taken on board with a justification where the original suggestion was modified”.  To my understanding the AHTEG has not provide a justification for the suggestions that were not taken on board, neither a justification on the comments and suggestions taken on board that modified previous texts. This step in the methodology is important because it gives context to the Online forum members for the time when we get the opportunity to comment on the whole document (or parts of it).
That is why this mandate in c) was supposed to occur previously to the step d) of the methodology where the work on the Online forum focus on “review all comments and suggestions with a view of having and improved version of the Guidance..”. I think that having the justifications is important to avoid frustration and misunderstanding during this long process of the revision.
Maybe the work on having the justifications will happen during the next face to face meeting of the AHTEG, and then there will be another Online forum. Could we have some clarification on this?
Finally on issues about process, I agree that 18 days is not enough time to look carefully to the guidance, but I understand that it may be complicated to extend the deadline.
I am hoping to have time to comment on the proposed text boxes and new texts in other post.
Kind regards
Sol Ortiz

Dear Helmut, Dear Colleagues

I would just like to recall the recent past discussion on the inclusion of human health topics in the Guidance (see eg posts [#7660] and [#7664] http://bch.cbd.int/onlineconferences/onlineconferences/forum_ra/discussion.shtml?forumid=17454&threadid=7578#7659). Human health risk assessment guidance is a valid part of the Guidance in my opinion. Moreover, it should be broader than that issued by Codex Alimentarius because not all current or future exposure pathways will necessarily be through ingestion (i.e. food). Not all GMOs will be intended to be used as food and therefore not subject to the review of regulators who rely on Codex guidance.
For example, a type of GM grass developed in the USA has not required a risk assessment there. GMOs not intended to be used as food, eg many kinds of trees, would therefore likely not attract agencies that relied on Codex guidance. According to Nature Biotechnology, the grass was just the start: "The grass set a precedent (Nat. Biotechnol. 30, 215–217, 2012). Since then, various developers have inquired about at least 20 different biotech plants. In all but three of those cases, the agency has agreed that the plants do not require oversight, according to federal documents posted on USDA of those case reviewed by Nature Biotechnology."
Furthermore, the revised Guidance very nicely describes indirect effects. Neither are these to be routinely captured by a food safety assessment.

I support the retention of dedicated text on human health RA and reference to the Codex guidance for matters of risk associated with ingestion.

Best regards
Jack

I wish to thank the subgroup, secretariat and Chair for all their hard work on making these suggested changes, and to everyone who has been part of the discussion so far.

Please note that it appears that different people may be referring to different line numbers for the same section of text and this may be due to the different settings on different recipients’ computers causing the text in the word document to wrap-around differently.  The line numbers I refer to below are those that appear on my computer.  I apologise if they are different to others’ line numbers.

Lines 200-203.  I agree with Ossama AbdelKawy [#7775] that these lines need improvement.  In particular, I have problems with “red list species” as this term may mean different things to different people in different countries.  I agree with Ossama AbdelKawy’s suggested changes in wording, except that I would like to change his words “large numbers of endemic or threatened species” to remove the requirement for large numbers of endangered species to be present, because some endangered species may have only a few individuals left.   So I suggest changing his words to be: “…containing high diversity, threatened or endangered species or large numbers of endemic species, or wilderness…”

Lines 305-312.  I agree with Ossama AbdelKawy [#7775].  I disagree with various people who have suggested that field trials should be excluded from the Guidance, on the basis that field trials are clearly a release into the environment.

Lines 334-335. I agree with Ossama AbdelKawy [#7775].

Lines 348-350.  Considerations of uncertainty and communication of that uncertainty can be two different things.  Therefore, an improvement to the text would be to split them apart as follows: “Considerations of uncertainty may improve the outcomes of a risk assessment and strengthen its scientific validity. Relevant considerations include the source and nature of uncertainties, and focusing on uncertainties that can have a significant impact on the conclusions of the risk assessment.  Communicating the uncertainty may improve the understanding of the risk assessment process and transparency in the decision making process.”

I support Jack Heinemann’s view[item 2 in his post #7793] about the deletion of the sentence on “scientific uncertainty”.

Lines 396-399, being “Due to the importance of centres of origin and centres of genetic diversity, as highly genetic diverse and repositories of wild relatives and landraces, taking into account both their spatial and temporal characteristics, any potential adverse effect identified during the risk assessment can be considered significant”.  I found the wording unclear and confusing.  Have words been inadvertently left-out or repeated?  Should “as highly genetic diverse” be “highly genetically diverse areas”?  My suggested re-wording is: “Due to the importance of centres of origin as repositories of wild relatives and landraces, and the importance of centres of genetic diversity for their diversity, any potential adverse effect on these centres that is identified during the risk assessment can be considered to be significant.”

Lines 401-402.  “and what would the consequences be” should be changed to “and what the consequences would be”.

Lines 403-406, being “In order to properly answer this question, availability and access to adequate baseline data, models to understand and simulate gene flow, and methods to identify and quantify possible consequences related to the introduction of LMOs in centres of origin and centres of genetic diversity are fundamental to enable sound risk assessments and to inform decision making.” Would be better expressed as: “In order to properly answer this question and to enable sound risk assessments and to inform decision making, it is fundamental to have: adequate baseline data available and accessible, models to understand and simulate gene flow, and methods to identify and quantify possible consequences related to the introduction of LMOs in centres of origin and centres of genetic diversity.”

My suggested changes to the box on “Centres of origin and centres of genetic diversity”, above, are simple word-smithing of the existing text to make it easier to read, if the text doesn’t change much.  However, I need to add that I agree with Ossama AbdelKawy’s suggestions [#7775] that the text should be strengthened, and in the way that he has suggested.

Lines 633-655, being the box on RNAi.  I strongly agree with the suggested changes made by Jack Heinemann, as they more thoroughly capture the reality of these LMOs.

Lines 716-749, being the box titled “LM crops and the use herbicides”.  I strongly support the inclusion of this box in the Guidance and strongly disagree with various people (e.g. #7783 and #7795) who wish to remove it.  LMOs that are designed to be herbicide tolerant are designed to work with herbicides as a LMO-herbicide package and it would be negligent in my view to ignore a significant part of the package as if it did not exist.  Please note that there is a typo in the heading.  It should read: “LM crops and the use of herbicides”.  Another lies in Line 718, which states  “In the case  LM crops that are resistant to herbicides…” and would be better written as “For LM crops that are resistant to herbicides..” 

I agree with Jack Heinemann [#7796] that the use of “synergistic” by itself is a little unusual in the sentence “Furthermore, the use of multiple herbicides in the same area, applied either simultaneously or in sequence, may result in synergistic effects that are more toxic and harmful than when the individual herbicides are used alone.”  And that we should not exclude additive effects.  While his suggestion that we should remove “synergistic” is a good one, an alternative would be to just replace “synergistic” with “additive and synergistic” or  “additive or synergistic” depending on the context.  For example, the first of these sentences could be changed to: “Furthermore, the use of multiple herbicides in the same area, applied either simultaneously or in sequence, may result in additive or synergistic effects that are more toxic and harmful than when the individual herbicides are used alone.”  Subsequent sentences could be similarly changed.

Lines 781-795.  Risk assessment of LMOs and human health.  I agree with Ossama AbdelKawy [#7775], that a risk assessment of human health from exposure to an LMO is a complex issue and one that can’t be properly contained in three paragraphs of a text box.  I therefore support a separate section in the Guidance as he has suggested, in order to better address the issue.  That is, the decision of the previous AHTEG meeting in Brasilia for this topic to be “incorporated into the existing Guidance, by creating text boxes or adding new text, as appropriate” [Helmut Gaugitsch, #7774] would, in my view, be best enacted by adding new text in the form of a separate section.  Important considerations could then addressed, such as: different routes of exposure (eg dermal contact, inhalation of dust, flour or pollen, and consumption via food or water) may result in different effects; consideration of possible allergic, toxic, reproductive and carcinogenic effects; the role of toxicity studies compared to the more common animal production studies to assess health outcomes, whether to rely on animal studies or to also require studies on humans for a certain LMO, a discussion of possible additive or synergistic effects of herbicides on human health which may not be fully covered in the current Guidance, etc.  Furthermore, Jack Heinemann’s sound suggestions [#7796] for additions to this section could be given proper space if this part of the Guidance was given a separate section.  While some may argue that consumption of the LMO in food may be covered by a separate agency in a given country and hence should not be in this section, such an agency would not consider the entirety of additive exposures through dermal, inhalation and consumption routes (via food and water) as a whole. 

Also please note that the term “statistical significance” in the current text box is a statistical term that usually means that the p value resulting from a statistical test is less than 0.05, yet the text reads as if that is not what is meant.

Lines 1073-1074.  I agree with Ossama AbdelKawy’s view [#7775] that “considered as a single event” should be removed.

Lines 2424-2426 says: “Crop wild relative – Crop wild relatives are species closely related to crops (including crop progenitors) and are defined by their potential ability to contribute beneficial traits to crops, such as pest or disease resistance, yield improvement or stability.”  Would be better expressed as: “Crop wild relative – Crop wild relatives are species closely related to crops (including crop progenitors).  They may have the ability to contribute beneficial traits to crops, such as pest or disease resistance, yield improvement or stability.”  The reason for my suggested change is that the existing text suggests that if a plant species has no currently-known ability to contribute a beneficial trait to a crop, then it may not be classified as a crop wild relative.  That is, the definition seems to rest on how the wild relative may be able to improve a currently-used crop, rather than the wild relative being genetically linked to the crop per se.

Dear Colleagues

This is an addendum to my previous post which included comments on the box on herbicides. I've been thinking about the interventions by others that have indicated that (at least in their or some countries) herbicide effects are considered under other regulation or regulators (eg #7795, #7782, #7783). I've been searching the evaluations in my own country and I cannot find an example of our regulator yet considering mixtures of herbicides of different active ingredients (as might be routinely or at least potentially used on some stacked LMOs). My colleague from New Zealand may be able to correct me on this point if I have overlooked it.
I did see discussion on Europe about introducing such evaluations. But to my knowledge, this is still at the discussion stage.
A contact in the US said that EPA may at times require some limited acute toxicity testing of mixtures.
Coupled with the need to be relevant to all member countries and without perfect knowledge of what their risk assessment frameworks might be, I find it compelling to include this advice on assessing the risk of herbicide/LMO combinations.

Best regards
Jack

Dear all,
First I would like to thank the group that worked on the update of the guidance document. I was reading the document and the comments in this forum and I want to put in consideration the following aspects:
I agree with O.A. El-Kawy in lines 200-203 “to replace those examples with the ones highlighted under the CBD to give an idea about the full range of protection goals”, I also like his text proposal.
Lines 207-208: I suggest eliminating the term mortality because it is not an attribute that must be protected and it may create confusion.
Lines 213-216: the concept “measurement endpoints” is confusing as is in the document, it is not possible to measure “adaptation” or “growth”, it is possible to measure indicators or variables.
Lines 385-409: Box “Centres of origin and centres of genetic diversity”, I think that it could be included near or in step 3. As O.A. El-Kawy said, it is relevant to talk about the importance of these areas for indigenous communities, but I believe that it is also very important to flag that one of the most relevant issues for these areas is the necessity of the maintenance of the domestication process (which is still on going).
Lines 584-603: there are different ways of classification, and it could be relevant to clarify that the the ones  included in the box come from different classifications:
1) If the effect is caused by direct contact with the GMO or not: direct or indirect effects.
2) If you are talking about the moment when the effect is observed: Immediate or delayed.
3) Depending on the frequence or dose of the “entity” (i.e. substance) and/or the interactions with other entities: cumulative or combinatory.
Lines 736-739: text proposal “In this context, potential synergistic effects may be taken into account during the assessment of LM crops containing  herbicide resistance genes if they are likely to be cultivated in areas where other herbicides are used”
The box “risk assessment of LMOs and human health” is not necessary because this is a topic which has been developed sufficiently by other initiatives and is regulated in different ways. I suggest to include “Principles for the risk analysis of foods derived from modern biotechnology” (http://www.codexalimentarius.org/download/standards/10007/CXG_044e.pdf) and “Guideline for the conduct of food safety assessment of foods derived from recombinant-DNA plants” (http://www.codexalimentarius.org/download/standards/10021/CXG_045e.pdf), as a foot note.
I think that it is important taking into account the ideas of Yan Devos related to extend the topics “protection goals” and “Problem formulation”.
Finally I agree with the opinion of Ayako Yoshio about the use of the “text boxes” because currently some of them talk about relevant topics (i.e. centers of origin and diversity or that one about RNAi) meanwhile others are related to definitions (i.e. effects) or examples of a particular situation that maybe could be included as part of the section “Use of terms”.

Thank you for the opportunity to comment in this round of discussion.

Best regards,

M. Andrea

Dear colleagues,
first of all my thanks to the subgroup and to the Secretariat for providing the improved version and giving this opportunity to share points of view.

My name is Werner Schenkel. I am working as a risk assessor at the Federal Office of Consumer Protection and Food Safety (BVL) which is responsible for the authorisation of genetically modified organisms, plant protection products and veterinary drugs in Germany.

I very much agree with points already brought up by others before, e.g. Yann [# 7780], about the efficacy of adding boxes and how to improve the text in order to stringently guide risk assessors through the risk assessment process. If it is decided to keep the text boxes as part of the Guidance they should be used in a coherent manner – see Ayako [#7782]. That could be done by singling out descriptions that are either too explicit or too detailed to be part of the main text. There should be a link between these text blocks and the respective passages in the main text – see Boet [#7783].

These are my comments on text in tracked changes:

Lines 385-409 (centre of origin and of genetic diversity). It should be clarified why this text box is placed in the chapter “planning phase of the risk assessment”. The statement in lines 400, 401: “any potential adverse effect identified during the risk assessment can be considered significant” is suggesting that the context is “identification of criteria for the acceptability in the planning phase of the risk” (Line 416). Here the link to the text box should be included. It is also suggested to bring the text box in the context of Step 1 of the risk assessment (lines 675-679).
The statement in lines 400, 401 should be rephrased to: “any potential adverse effect identified during the risk assessment is to be considered significant” – as if it is significant there are no options.

Lines 643-665 (LMOs developed by RNAi-based methods). If the box has been established to facilitate the risk assessment of RNAi-based LMOs, it failed this goal. The box has been written in a general and imprecise way, so it doesn’t provide the assistance, rather it raises more questions. Most aspects of LMO risk assessment which are addressed in the box are already considered in the guidance. Title and text of the box need some clarification in order to not confuse the readers.
The title should be changed to “LMOs using RNAi to reduce gene expression”. The present title is misleading as the use of “RNAi-based methods” does not create a “living organism that possesses a novel combination of genetic material” (LMO definition in the Cartagena Protocol (Article 3(g)). This definition of the term LMO lies clearly in the novel combination of genetic material. Therefore the modification of gene regulation or the metabolism of an organism via dsRNAs does not create a LMO if no novel combination of the genetic material of the organism is created.
The text has to be more precise in order to facilitate the specific risk assessment of RNAi-based LMOs:
- Lines 651, 652: change “homology to other species” to “homology to sequences of other species”
- Lines 652, 653: “2) length (size) and amount of the dsRNA that may cause adverse effects to other organisms;”
Question 1: Which adverse effects and other organisms are meant in connection to length and amount of dsRNA? The assessment whether the dsRNA may cause adverse effects to other organisms is not part of the molecular characterization. It should therefore read: “2) on a case-by-case basis, depending whether there is reason to assume that the dsRNA may cause adverse effects to other organisms, length (size) and amount of the dsRNA”.
Question 2: Why not “length (size) and amount of the siRNA” as siRNAs represent the active component of the system?
- Line 654: Is there meant to be a difference between “4) expression level” and 2) “amount of”? If not, the wording should be harmonized.
- Lines 656 to 661: The considerations in this paragraph are not specific for LMOs using RNAi. Points 1) to 3) are valid for all LMOs and are already sufficiently covered in the main text of the guidance. The paragraph can therefore be deleted.
In contrast to necessary precisions I can see no reason to add comments to this box which hold to any LMO and are summarized in the main text already. The extensions proposed by Jack [7793] on e.g. effects on changed environmental conditions, horizontal transfer of genetic elements and exposure to non-target organisms are redundant therefore.
I agree with Patricia [7797] in that “the utility of ‘omics’ technologies in environmental risk assessment is very controversial (line 662)”. The value of omics to risk assessment is doubtful as the variation by season, location and genetic background usually hampers an interpretation of the data obtained. Therefore, they usually do not add to full risk assessment or even hazard identification and should not be asked for in the first place.
I also agree with Boet [#7783] to include a link to this information in the text in lines 633-638.

Lines 726-747 (LM crops and use of herbicides). I agree with Boet [#7783]: “This detailed text in the section ‘potential adverse effects resulting from the LMO’ gives the impression that this one of the main adverse effects that has to be taken into account in the risk assessment.” In order to not confuse risk assessors the context should be clarified. Therefore I suggest to replace the first sentence (lines 727 to 728) of the text block by the following text:
“In most legislations the assessment of potential adverse effects of plant protection products (PPP) as e.g. herbicides used during cultivation of a crop is regulated by specific PPP acts. If such regulations are missing a national legislation for the introduction of LMOs may also require an assessment of the environmental impacts of the use of respective PPP. The aspects of the assessment described in the following are not specific for LMO but apply accordingly to PPP used on conventional crops.”

Lines 791-805 (human health). Two problems detract the text box from being a valuable addition to the guidance in its present form:
1) The first two paragraphs just pick up few aspects of a LMO food/feed safety assessment which is more comprehensively and in detail described in the Codex documents cited below. In the light of the given reference the text box information remains not only superficial, but redundant. There is no need to pinpoint bits of information on this topic here if there is a stand-alone UN document on food safety available.
2) In addition, the information in the two paragraphs is partly incorrect:
First paragraph: Comparison is to be done with a near-isogenic comparator, as non-modified parental organisms are no suitable comparators e.g. when dealing with stacks because of their different genetic background.
Second paragraph: Statistical significance is not a reason for designing different experiments to assess for toxicity. Rather should each experiment be designed in a way that it obtains robust statistical power. Toxicity is not assessed in long-term studies. Long-term studies can be used to address questions of carcinogenicity and teratogenicity.
In order to address human health aspects in respect to food safety, it would be sufficient to include the last paragraph and references therein into the text. It may be useful to cite the respective OECD documents on toxicity assessment as well: OECD guidelines for the testing of chemicals, section 4 (http://www.oecd-ilibrary.org/environment/oecd-guidelines-for-the-testing-of-chemicals-section-4-health-effects_20745788).

Further aspect to consider:
Lines 474 to 480: The text changes caused the loss of the original statement addressing the comparison of effects of the LMO and the non-modified recipient organism on NTOs under different conditions reflecting differences in management that are expected to be applied in practice (e.g. different pesticide types/application regimes).

Thank you very much.

Kind regards,
Werner

Dear all

Thank you for the interesting discussions and to the Chair for clearly setting out the tasks before us. I would like to pick up on a few of the points raised so far in the discussion.

1) I strongly disagree with the proposal [#7792] to delete the whole section on uncertainty. This is a critical section integral to any risk assessment, and is consistent with the Protocol, which addresses aspects of uncertainty in its paragraphs 4 and 8(f) of Annex III, as well as Articles 10(6) and 11(8). Neither the sentence that was deleted nor the two sentences that was added (Lines 334-335) add to the narrative. The sentences that were added (referring to one particular definition), in my view do not do justice to the substantial literature on scientific uncertainty.

2) Regarding the text on centres of origin and centres of genetic diversity (Lines 389-415) I would suggest that some concrete links be made with Article 8(j) of the Convention.

3) The box on ‘types of adverse effects’ (Lines 602-628) is a useful addition.

4) I support the inclusion the box on ‘LM crops and the use [of] herbicides’ (Lines 751-785). As stated by others, LM crops that are herbicide-resistant are engineered to work in conjunction with those herbicides, so the latter should not be disregarded in the biosafety context. A text box in this case is a suitable modality.

5) I agree with the various suggestions to provide stronger and more thorough treatment to the issue of ‘risk assessment of LMOs and human health’ (Lines 817-831). Reference to the Codex Alimentarius Commission’s Principles and Guidelines is appropriate for foods derived from LMOs, and exposure through ingestion. However, given that the Protocol’s reference to human health is not limited to this one pathway, it would also be necessary, as others have suggested, for this Guidance to include human health issues that arise from different exposure pathways (direct and indirect) as well as considerations beyond what is addressed by Codex. The suggestions made in #7800 are very useful in this respect.

kind regards
Lim Li Ching
Third World Network

Dear all,

Considering many of the comments related to include herbicides in the Guidance on risk assessments of LMO. I just want to pick up some of the group's opinions concerning avoid duplication in herbicide´s regulations (with the exception of the cases where countries do not have regulations about pesticide control).

Nowadays, resistance is a regular condition in the use of a product. The adecuate use and handling of the product determines the development of resistance, so, it is not determine by the biotech product itself (that is the focus of the assessment), because of that, I consider useful the box of herbicides just when a country lacks herbicide regulations, otherwise I think we are upregulating the technology.

Thanks a lot
PMC

Dear Colleagues

Whereas I agree with Werner that the treatment of RNAi is not optimized using the boxes (and I think it should be a topic in Part II of the guidance), we have as a group thoroughly covered that discussion and settled on a box. Thus I would like to help make that box to be the most effective box possible.
In a previous forum I suggested that we provide a definition of dsRNA that would help us to avoid the philosophical debates on whether only DNA is genetic material or what constitutes 'new combinations'. [([#7595] in
http://bch.cbd.int/onlineconferences/onlineconferences/forum_ra/discussion.
shtml?threadid=7577)]. Clearly, when RNAi is instigated by creating a novel combination of DNA there is no debate on the organism being an LMO. The question of concern is whether using RNA itself to instigate an RNAi effect creates an LMO. I say that it does because it is a nucleic acid molecule created using in vitro nucleic acid techniques and introduced into a cell where it creates a novel combination of nucleic acids (at least the regulatory dsRNA that otherwise would not be there). Moreover, RNAi effects can be heritable (at least in some organisms, and certainly via cell division).

Unfortunately, I see this discussion as turning into a battle of the definitions and that is something for the MOP to work out, if it feels that the definitions could be improved. At this stage, in my opinion the existing definitions are sufficient to justify our offering guidance to Parties on RNAi created by whatever means. I think we should offer them guidance. If they choose to not take it, that is also their prerogative.

Best wishes to all
Jack

Dear all,

In response to the last post [#7805], it seems that some may hold the view that 'some guidance' is better than 'no guidance', although I believe there are many among us who hold the view that, in fact, 'no guidance' is better than 'bad guidance'.  That is at the heart of this debate. 

This notion that we can offer guidance when we do not agree that it is good guidance, and 'they' can choose to take it or not, is unjust.

You may take whatever concrete suggestion you wish away from this.

Thanks,
Karen

Dear Jack,

A  comment to your posting about RNAi.

"The question of concern is whether using RNA itself to instigate an RNAi effect creates an LMO. I say that it does because it is a nucleic acid molecule created using in vitro nucleic acid techniques and introduced into a cell where it creates a novel combination of nucleic acids (at least the regulatory dsRNA that otherwise would not be there). Moreover, RNAi effects can be heritable (at least in some organisms, and certainly via cell division)"

Your argument would makes sense if we agree to regulate the process (introducing dsRNA that was not there before), and  not the product (a healthy plant).

From a viroloy point of view, we could compare  a plant with  a virus  infection (say a papaya  with  Ring Spot Virus or a bean plant with Bean Golden Mosaic Virus) to a healthy plant  without the virus. We want the healthy plant.  From a regulatory point, we want a safe plant  for the environment and human health. Which one is the comparator?  Both healthy and virus-infected exist in nature. I would say the comparator should be  the healthy plant the farmer wants to grow,  and we want to eat.    We often eat plants with viral infections because there is no alternative. Nobody bats an eyelid about it, because it is inevitable.

A GM papaya or  GM bean (considered an LMO)  produced by RNAi will not be able to express one of the viral  capsid proteins, the virus cannot encapsidate,  and thus the plant will be free of virus - healthy - and  identical to the comparator? a relevant biosaefty question would be:  which molecule poses more risk to the environment and human health -  all the expressed viral genes and proteins including whole viral particles in an infected plant that we gorw and eat, or a dsRNA used to silence one gene?

I think this would be useful for  a novice risk assessor  to grasp when evaluating a product developed by RNAi. A crop develop by RNAi to be resistant to a virus - ie. a healthy plant  with no virus -  is much better for the farmer and healthier for the consumer. Would  it be useful to put this in a box too? Maybe it would, to give perspective.

However, this explanation requires too much text and an understanding on plant virology and molecular biology and I would not advocate to include more text in boxes, as this makes the guidance far to complex and thus not "practical and useful".

I agree with statements from earlier fora and  present discussions that including text boxes complicates the guidance for novice risk assessors.

Best regards,

Maria

Dear colleagues

I would like to thank the Secretariat, the moderator and the members of the AHTEG for their hard work,  as well as for the possibility to participate in this forum.

Concerning the new text additions to the guidance, I would like to share the following suggestions.

1) I think it would be useful to introduce a small paragraph  in the “Objective and scope of this guidance”  section , in order to briefly expose the purpose of the “text boxes”(thinking on readers in the future who might not be aware of  the present discussions).  It would also be helpful to include references to the boxes inside the text where the come into account, maybe giving a number to each box.

2) In the box “Protection goals, assessment endpoints and measurement endpoints  (L 196-221) The relationship (or the difference), between the “specific attribute” of an assessment endpoint (eg honeybee mortality), and the corresponding measurement endpoint (eg fitness, growth behavior and development) doesn’t seem to be explained clearly enough, at least  in terms of the examples chosen. In order to make these concepts easier to understand it might be useful to recheck the definitions and use related examples to illustrate “assessment endpoints” and “measurement endpoints”.  

3) It might be  helpful to check if the places where the boxes are inserted along the text are the more suitable. For example the box on Centres of Origin (which is by the way, a relevant issue to be developed and included in the roadmap, due to the importance for humankind of  the genetic pools present and continuously evolving  in these geographic regions),  could be maybe inserted a little farther in the text  for example after line 417 or line 447.


Best regards,

Caroline

Dear all,
I would like to thank all those involved in making these suggested changes; it is obvious that a lot of work has gone into them. However, there are some key issues with the roadmap that still need to be resolved, one of the main ones is problem formulation. Including problem formulation is something that was suggested during the testing of the previous roadmap version as this would help integrating the “Planning phase of the risk assessment” section with the “Conducting the risk assessment section” making it easier to link the four steps described in the assessment. Including problem formulation as an integral part of step 1 would also help establishing what information is necessary for the assessment, so risk assessors would have a framework that guides them through the process of deciding which information they would need for each case. The need to incorporate problem formulation in the roadmap has also been raised by Yann [#7780], Maria [#7787], Patricia [#7797], Maria Andrea [#7798] and Wernel [#7801]. Thus, I believe that just adding a box on problem formulation is not enough. In any case, the current box, if it remained, it would need thorough revision.

Within problem formulation, the selection of relevant protection goals for each case and their translation into assessment endpoints that focus the risk assessment is crucial. Thus I would suggest revisions to this text. The box on protection goals correctly points out that it is important to distinguish between assessment endpoints and measurement endpoints, but then gives an example that is incorrect, thus confusing rather than clarifying. This has also been raised by Maria Andrea [#7798] and Caroline [#7808].

Regarding the lists of points to consider, now called “elements to consider”, I agree with comments made by Yann [#7780], Ayako [#7782], Boet [#7783] and Piet [#7784]. I think that these should be revised as they distract from the flow of the document. Also every element for consideration suggested should be put into context, explaining why it should be considered and why it is relevant to the risk assessment. Again problem formulation would guide assessors in the selection of the relevant elements of consideration, so I am wondering if indeed these lists should be in boxes, rather than problem formulation being in a box?

Regarding the newly added boxes, some of them would not be necessary if the problem formulation approach was followed, as Patricia pointed out ([#7797]). For example, for the box regarding the centres of origin, one of the protection goals selected for the risk assessment could be the protection of natural populations of wild relatives. This would make sure that specific concerns associated with centres of origin would be taken into account in the risk assessment following the same process described for any other LMO. I agree with Yann’s comment [#7780] that the boxes distract from the flow of the document and as Boet [#7783] pointed out, their position within the document should be revised. In addition, as Ayako [#7782] mentioned, they should be consistent and contain relevant information and examples.

I support the many other comments made regarding the deletion of the box on LM crops and the use of herbicides (Ayako [#7782], Yann [#7780], Boet [#7783], Rafael [#7795], Werner [#7801] and Pedro [#7809]). There are other regulatory frameworks that look at the effects of the proposed herbicide use and rate over a crop (whether that crop is GM or not).

Given the number of comments I have made in the document, I thought it would be more practical to upload a version including my proposed changes in tracked mode.

I understand that it must be frustrating for those trying to make progress with the changes to re-visit those changes. However, I have to agree with the comments made by Karen Hokanson, in that there has been very little chance to comment on any amendments made since the testing took place. Many parties invested a considerable amount of time testing the guidance and providing comments, hence, it would seem fair if those were given the opportunity to make sure their concerns regarding the document were addressed and that the new text is something they can agree to. At the end of the day the purpose of this exercise should be to have a roadmap that is useful to all that endorse it and does not compromise current practices that have a proven track record of working.
I hope these comments are useful.
Best regards
Monica

Dear Helmut, dear all
With regards to the four newly introduced topics I believe the following need to be discussed throughly:
- What are the elements that needs to be included under those topics?
- What is the structure and the outline of the introduced text? how to harmonise the text under those topics.. I think there is a need for a least two sections one includes the rational for having those topics in a separate form and what are the points to consider with regards to those topics in the context of risk assessment
- Where and how to introduce that text?
for example to answer the questions: should each topic be introduced in the form of single box or multiple boxes at different place and where?
        Is just having a box appropriate or we need separate section under the specific type of LMOs?
Regards,
O.A.El-Kawy

Dear Helmut, Dear All-

I realize that these are not ‘concrete suggestions’, but I wish to agree with the suggestion from O. A. Kawy in the most recent post [#7811] that it would be important to thoroughly consider those questions regarding the text that has been placed in boxes.  I think this is very much in line with the points raised in earlier posts by others regarding the confusion about the purpose of the boxes.

I would add two more questions to this list in post [#7811]:

-Would the additional guidance that is already found in separate sections of the guidance be better placed in boxes? What is the difference between ‘boxes’ and ‘additional guidance’?

-Would most if not all of the ‘elements to consider’ be better placed in boxes?

Here are some concrete suggestions:
On my first question above, I do recall during the face-to-face meeting making a similar suggestion (but it somehow did not get considered then) to put the additional guidance into boxes, and save the actual separate guidance for organisms other than plants. 

In this case, the separate guidance would be for LM mosquitos and now LM fish.  And stacked genes or traits, abiotic stress, LM trees (which are plants after all), monitoring, and synthetic biology (IF the AHTEG will be developing guidance on synthetic biology – should this be separate guidance or put in a ‘box’?)
Much of what is now in the additional guidance is redundant to what is in the roadmap, and so streamlining that guidance into boxes would cut down on that redundancy.

On my second question above:
I think putting all of the ‘elements to consider’ into boxes would address a fundamental concern expressed in many of the comments about the testing (Categories B and D) that it is hard to follow and seems more like a checklist than guidance.   And I can’t see how the ‘elements to consider’ are different than the information that has now been placed in ‘boxes’.

I also believe that the revision from ‘points to consider’ to ‘elements for consideration’ is not sufficient to make the roadmap seem less prescriptive or less like a checklist, although I think this was the justification for this change.  I do recall a comment based on the testing from South Africa [Comment G36 – from the ‘General Comments’ section of the testing questionnaire] suggesting to change ‘points to consider’ to ‘points that may be considered’. I fully agree with this comment.  It should be made very clear that the ‘points’ or ‘elements’ are not ‘requirements’.  Placing them in boxes could also serve this purpose.  Further revisions on the guidance could then focus on the rest of the roadmap that is more about the process of risk assessment.

Thanks,
Karen

Many thanks for Helmut, Secretariat and Sub group for their work.
I totally agree with Osama's explanation in his post 7791and also appose the delition of the monitoring from the title and see the Guidance title as it is now. Conserning protection goals and problem formulation I agree with Yan Devos post 7780 and think that we need to work on the text in the context of the above.
With regard to new introduced topics and boxes added I think that we need harmonize information as now its divergent. Osama in his post 7777 made a good proposal to arrange the text inside boxes into 2 subsections - introduction to the topic and elements for further considerations (additional points other than under the roadmap or may be emphasizing certain points that are present under the roadmap but from a different perspective ). And revise than again the Road Map and boxes not to reduplicate elements and not to confuse readers. For better understanding for future readers it will be good as it has been proposed in the post 7808 to introduce a  paragraph  in the “Objective and scope of this guidance” which briefly explane purpose of the boxes and to include references to the boxes giving a number to each box post.
And concerning the proposal to remove risk assessment for field trials ( small or large-scale). I don't see rational reason why we need to remove it from the context of the Guidance. During any release  could be risks, assessment endpoints, protection goals (including human health). For GM plants which produce farmaceuticals or BAD it could not be only exposure through ingestion (someone accidently ate GM fruite) but other pathways including pollen if constitutive promoter used in construction and because the end product of such plants could possess allergenic or toxic potential. In case of our legislation risk assessment should be made twice before field trials and before State registration of new variety for commercial release. And existing Guidance was wery helpful to focus on all potential environmental risks before field trials. And it will be very helpful for our risk assessors if it will be provided stronger guidelines or points to consider for risk assessment to human health within the scope of this Guidance.  In this context I agree with the post of Lim Li Ching 7802.
Best wishes, Galina

Dear forum participants,

In order to provide concrete recommendations, my comments pertain to the boxes under consideration:

As a background to my specific recommendations, I think the inclusion of boxes gives the impression that these issues or types of LMOs deserve specific attention over others not mentioned, which is not correct, and only serves to detract from the application of the principles being elaborated in the guidance.  The boxes not provide help or illustrations regarding how the principles in the guidance could be applied, which is in my opinion what boxes should do.  My opinion on this aligns completely with Jann Devos’ posting #7780, where he elaborates these ideas more thoroughly and extensively.

Regarding the box on RNAi:  The text of the guidance in the box offers nothing new that is not already considered in the assessment of other types of LMOs.  As we all know, risk assessments have been conducted quite well already with RNAi-based methods, using the same considerations as other LMOs.  Therefore, we can add those examples as references relevant to Step 1, and leave it at that.  To me, the box, as written, doesn’t need to be included.

Regarding the box on human health:  The box as written provides nothing new than what is already well-covered in more detail by the Codex Alimentarius Commission. In fact a more succinct way of saying what is in the box is “Refer to Codex Alimentarius on this matter.”  Whether or not we think that human health is something that needs special attention under the Cartagena Protocol, or whether it is out of scope of the Protocol, the box as it stands provides no new guidance that is not already available elsewhere.  It doesn’t even explain, as Boet Glendora has pointed out, how “human health” is taken into account from the perspective of conservation and sustainable use of biological diversity.  As with the box on RNAi, whether we agree or not that there should be a box there, this particular box doesn’t serve the purpose and should not be included.

Dear all,

I would like to first thank Helmut for chairing this important discussion, the members of the AHTEG for their work and the Secretariat for organizing this. I also thank the forum members for their contributions to the process.

I have a few general comments to the revised document and they follow the order of the guidance document. I leave the grammar and semantic suggestions or clarifications to my English-speaking colleagues.

1) I support the additional text clarifying the relevance of information from field trials. For countries like Brazil, which the adoption for the technology occurs in continuous millions of hectares, the contextualization of field trial information is highly relevant.

2) I support the original text of the section on uncertainty and so the maintenance of the text as it is a strong component in RA, something reflected in the Protocol.

3) I support the inclusion of a clarification textbox for center of origin and I support the link between this text and Article 8(j) as suggested by colleagues #7802. In fact, I think this box is highly relevant because it recalls that not only wild relatives might be include in protection goals but also landraces, like the case of maize in Mexico. In that case, teosintle (the maize wild relative) is not the only relevant species. Therefore, I disagree with other colleagues that suggested it deletion. Also, the box adds clarity to the concept of ‘centers of diversity’ too, something neglected by some regulators. Centers of diversity are as important as centers of origin for the maintenance of agrobiodiversity worldwide.

4) I support the inclusion of the text line 526 that mentions the real possibility of lack of comparators in future techniques.

5) I support the addition of the ‘types of adverse effects’ box and the Norwegian Gene Technology Act is a very good reference of it.

6) I strongly support additional information on dsRNA methods.  This topic deserves additional attention and it is something also reflected in the EFSA Workshop on RNAi-based plants, which has been followed by a tender to gather relevant literature information. I agree with #7775 to move the box to ‘methods of modification’.  I also agree to delete ‘to reduce gene expression’ and other text revision as per suggestion #7793. I also support the definitional ground that RNAi-based methods do create an LMO using ‘nucleic acids’ because RNA is a nucleic acid and, therefore, I support #7805 clarification. This discussion has extensively covered in earlier forums. I also support the information on the potential utility of ‘omics’ technologies to complement the molecular characterization step of RA. This is something also reflected in EFSA special issue EFSA Journal 2012;10(10):s1008. Nonetheless, bioinformatics based on omics is a crucial step in the search for off-target effects of new dsRNA sequences and should be therefore mentioned.

7) I support the box on ‘LMO crops and the use of herbicides’ and I am also of the opinion that this information is fully compatible with the different regulatory regimes we have for both LMOs and herbicides, in fact, it helps inform regulators about the possible interplay in the risk assessment of both products.

8) I support the box on ‘LMOs and human health’. I also agree that the reference to Codex Alimentarius is appropriate and that it should be treated in a broader context by the Guidance because not all current or future exposure pathways will necessarily be through ingestion as expressed by colleagues #7799 #7800. It could also be better fitted into a separate section for a more dedicated space.

Overall, I would like to thank the AHTEG members and the Secretariat for their work and also the forum colleagues for their insightful comments to this discussion.

Best regards,

Sarah

Dear all,

It may be appropriate to note at this final stage of this discussion, that the disagreement on what the guidance should be, and how useful and practical it is, has been prevalent since this process started and has not diminished to date. As Sol recently stated, it would not be fair to characterize the document in front of us as representing agreement, and we see that many substantial matters including the title are not agreed upon by AHTEG members.  Thus, my heartfelt thanks and admiration to Helmut and the subgroup for undertaking this rather complex task.

Although I will not offer specific suggestions to delete or change parts of the text, I am in agreement with  Yann´s, Boet,  Karen´s and Hector´s  comments, specifically about including text boxes. We need to prepare "teaching material" that is easy to understand and follow. As a educator,  I know that too much information is often counter-productive for novice learners. 

My understanding, and please correct me if I am mistaken, is that the objective of the guidance is to provide novice risk assessors from countries that still lack a mature biosafety framework (mainly developing countries) to conduct a coherent risk assessment, with good material to guide them through this rather complex process. 

One analysis of the testing results indicated that some developing countries found the guidance useful.  In those cases, success has been achieved.  However, this has not been our experience within the Central American community were many countries abstained from the testing.  As Karen noted, not speaking is not the same as agreeing and one has to ask why so many countries abstained from the testing. My informed guess is that they found the exercise dauntingly complex and time-consuming. Nevertheless, there is nothing preventing those countries that find the guidance useful from adopting it and putting it into practice.  We must recall that this guidance is not prescriptive and imposes no obligations on Parties.  My concern would be for those countries that did not do the testing because they don´t grasp the full complexities for risk assessment, and adopt a guidance presented to them  “by default”.

Given the text before us that the years of debate that have led to it, adoption of a final guidance at COP may be some way off,  and may require the Parties to carefully examine the process.   I fully recognize what is at stake in producing a coherent document that help those Parties that need it the most  (developing countries ?) and fully acknowledge the hard work and valuable contribution  everyone in this group, especially Helmut and the Subgroup,  are putting to follow a transparent process.

Warm regards,

Maria Mercedes

Dear participants of the Online Forum,

Thank you for the opportunity to contribute to this discussion. With due respect to the many people that have devoted significant time and effort to this long-standing project, I do have several criticisms, which I hope can contribute to improving the Guidance. I have reviewed previous versions of the Guidance over the years, and unfortunately I do not think it has been improved, as mandated by the Parties at MOP7. Rather, it continues to grow larger, more detailed, more repetitive, and at more than 100 pages now, difficult to follow, even for experienced risk assessors. I agree with the comments of #7780 that there is an apparent lack of clarity in the overall objective of the Guidance, its target audience, necessary messages and level of detail, and its structure – and this is where efforts should be focused. Instead, the new boxes add unnecessary detail and complexity and for the most part should be deleted. It is also concerning that the boxes appear to attempt to impose protection goals (e.g. centre of origin box), expand the scope of LMO risk assessment beyond the principles of Annex III (e.g. herbicide box, human health box), and include unsubstantiated statements (e.g. herbicide box). Also, as mentioned by others before me, it is a challenge to properly review this substantial document in the timeframe of an online discussion, therefore my review of changes to the text (other than the boxes) is not comprehensive.

Specific comments on the text:
Lines 196-221 (protection goals etc box): this box should be deleted. The terms here are already defined in the ‘Use of Terms’ section (from line 2402) and do not need repeating here; ‘measurement endpoint’ is not in the Use of Terms and could be added. The content of this box seems excessively detailed and inappropriate for the Guidance.

Lines 305-312: this newly inserted paragraph is not necessary – it repeats the content of lines 293-296 that state that the relevant information will vary according to the scale of the release, the steps outlined in lines 226-237.

Lines 313-326 (field trial box): I agree with others that this should be deleted. As noted in the comment above, it is already stated in the Guidance that the relevant information for risk assessment will vary according to the scale of the release.

Line 328 onwards (uncertainty section): I do not generally agree with the inclusion of this section - as pointed out before me there is no consensus on how to manage uncertainty. This is reflected in the Guidance on lines 335-338, with it ‘subject to much discussion’, and the ‘importance assigned to uncertainty and the determination of its occurrence, are dealt with differently under different regulatory frameworks’, and this factual statement should be retained. This section should be consistent with the Protocol and not give directions to risk assessors and decision-makers. The paragraphs in lines 355-360, 361-366 and 372-376 should be deleted, and lines 352-354 (from ‘Relevant considerations…’) and 370-371 should be deleted. Lines 367-369 repeat lines 341-342 and should be deleted.

Lines 385-409 (centres of origin box): this box should be deleted. The content is inappropriate in that it appears to be expanding on the provisions of the International Treaty on Plant Genetic Resources for Food and Agriculture and combining them with that of the Protocol to impose an obligation to protect centres of origin and prescribe how a risk assessment should be interpreted; e.g. text such as ‘must ensure that no negative effects … with a degree of certainty’ (lines 395-396), ‘any potential adverse effect identified during the risk assessment can be considered significant’ (lines 400-401). This is inconsistent with text, e.g. the paragraph below the box (from line 410, from line 418, and elsewhere in the Roadmap) regarding ‘protection goals, assessment endpoints and risk thresholds’ being determined according to a country’s own environmental policies.

Lines 462 onwards (choice of comparators section): the three paragraphs in lines 481-491, 492-499 and 500-509 should be combined and simplified as follows:
“The choice of an appropriate comparator will depend on the characteristics of the LMO being assessed, on a case-by-case basis. For example, in many cases the comparator selected as most appropriate may be the non-modified genotype with a genetic background as close as possible to the LMO, e.g. a (near-)isogenic line. In some cases, the closest genetic background may be provided by a negative segregant derived from the LMO. Additional comparators such as non-modified reference lines, or alternative comparators such as other cultivated varieties, including other LMOs, may be most appropriate for providing relevant information for the risk assessment.”
This reduces the emphasis on near-isogenic lines, which are a requirement imposed by regulators and not the Protocol, and puts the focus back on the Annex III principle of case-by-case determination of what is relevant (consistent with the statement in lines 470-471 that one or more comparators may be relevant).

Lines 576-578 and lines 604-608: combine as one paragraph.

Lines 579-580: delete this sentence. The paragraph in lines 563-569 covers the relevant types of adverse effects.

Lines 584-603 (adverse effects box): delete this box, the level of detail is not appropriate for the Guidance. The terms that require defining should be briefly defined in the ‘Use of Terms’ section.

Lines 643-665 (RNAi box): this box should be deleted. It is too detailed and it is unnecessary considering that, as pointed out in #7813, LMOs developed for the purpose of silencing genes via RNAi mechanisms can be (and are) assessed according to the same considerations that apply to other types of LMOs. The Guidance should be broad in scope and not make specific reference to types of techniques or engineering approaches.

Lines 726-759 (herbicide box): I agree with others that this box should be deleted. The herbicide regimes used by growers, and assessment of them, is beyond the scope of the risk assessment of LMOs and Annex III, and as pointed out in #7782 this is the remit of other regulatory frameworks – where assessors have the appropriate expertise. These frameworks include assessment of synergism where relevant and it is not appropriate for the Guidance to invite LMO risk assessors to take this into account. Further, as it reads, the language is not neutral in this section and could have the effect of biasing LMO risk assessors. 

Lines 791-805 (human health box): recommend deleting this box. I agree with the comments in #7783 that the text reflects food safety assessment and this is outside the scope of the Protocol, and #7782 that this is within the scope of other regulatory frameworks. If this box is retained, it is important to refer to the Codex Guidelines, and the text should be carefully revised to ensure it is consistent with internationally recognized standards for assessment of allergenicity and toxicity, particularly in regard to the statement about long-term toxicity studies (lines 797-798).

Lines 1001-1003: recommend deleting this point. It is confusing and reads as though there needs to be a risk assessment of the risk management strategy (‘taking into account that the proposed risk management strategies may introduce different risks’).

Lines 1018-1029 (related issues section): recommend the following revision to the text:
"Risk assessment contributes to the decision-making process for LMOs, and other factors may be taken into account, as appropriate, and in accordance with a country’s policies and regulations. Issues that are mentioned in other articles of the Protocol include: (list from lines 1022-1026).”
This is more descriptive rather than prescriptive, and example of other factors that are not within the scope of the Protocol (e.g. ‘co-existence, ethical issues’ in brackets on line 1027) should not be listed.

Lines 1041-1051 (Part II): recommend the following simplification of the introductory paragraphs:
“The guidance contained in this section, Part II, complements the Roadmap for Risk Assessment of LMOs in Part I, giving emphasis to issues that may be particularly relevant when assessing the risks of the specific types of LMOs and traits included in this section. As for the Roadmap, Part II should be considered in the context of the Cartagena Protocol on Biosafety, with Article 15 and Annex III of the Protocol applicable. Only those considerations that may be particularly relevant to the specific types of LMOs or traits included in this section are elaborated, without repeating considerations that may be more broadly applicable to different types of LMOs already described in the Roadmap.”

Lines 1198-1199: delete this point – as stated above for the herbicide box (lines 726-759), the scope of the risk assessment of the LMOs does not include pesticides.

Lines 1210 onwards (crossing and segregation of transgenes): this section is problematic. It reads as though crossing is a foregone conclusion where there is sexual compatibility, when the reality is far more complex. Item (a) of the ‘elements for consideration’ (line 1230) should be revised to:
“Presence of other single-event and stacked LM plants of the same species and likelihood of crossing considering the plant breeding system”.
This section also includes ‘intentional’ crossing (e.g. line 1232) ‘mediated by man’ (line 1221), when surely it is not within the scope of Annex III for risk assessors to anticipate any possible intentional activities of ‘man’. The sentence starting line 1221 should be revised to delete ‘either be mediated by man or’, and line 1232 revised to delete ‘intentionally or unintentionally’.

Lines 1239 onwards (detection of stacks): this section is also problematic. The sentence starting line 1258 (‘in the future…’) is hypothesizing about future technological applications and this is not presently relevant to the scope of the Guidance. It is noted in the ‘preface’ that the Guidance is a ‘living-document’ that can be updated as necessary, and this version should focus on currently relevant information, and the paragraph in lines 1256-1261 should be deleted. Lines 1262-1264 should also be deleted, it is not necessary for ‘each and all individual transgenes’ to be detected when the section is concerned with detection of ‘transformation events’. In the ‘elements for consideration’, item (a) should be deleted, as (b) is most relevant for risk management.

Line 2312: revise point (i) for greater specificity: “findings from previous or ongoing monitoring activities with the LMO, if any, and other relevant monitoring activities”.

Best regards,
Felicity Keiper

Dear participants in the forum, dear chair,

Thank you for the opportunity to contribute to the revision of the draft guidance and to share our views in this on-line forum.

Before providing specific comments to the text as requested by the moderator, I would like to join others on the forum that have expressed concerns about the overall quality of the document. After having read the full draft guidance document, I am left with a feeling that in its current state, the document is unlikely to deliver on its  primary objective – to be a useful reference for designing and planning risk assessment of LMOs. Substantial editorial effort would be necessary throughout the document in order to provide clarity, consistency and focus and to lift off the document from its current state of being primarily a compromise compilation of opinions.

Below are several specific comments for the latest edits of the document:

lines 385 – 409, Text box "Centers of origin": Delete the text box as its presence in the section of “Establishing the Context and Scope” is not appropriate, especially because the rest of text in this section provides a broad list of items that should be taken in to account.
Recommend to add the term ”Centers of origin”  in the section “Use of terms” starting line 2402

lines 474 – 477: The current edit of the sentence removed the needed comparison with the management practices of the non-modified recipient and the following edit is suggested by adding the text “in comparison to management practices applied to the non-modified recipient:
Moreover, an assessment of the potential adverse effects of an LMO (for example, a Bt crop) to beneficial organisms (for example, honey bees) must reflect the standard management practices that are expected to be applied to the LMO (for example, different pesticide types/application regimes) [ in comparison to management practices applied to the non-modified recipient].

lines 584 – 603, Text box "Types of adverse effects": Because the body of text of the guidance already contains the same information, I suggest that the text box is removed and the terms “direct effects”, “indirect effects”, immediate effects”, “delayed effects” are added, after revision to the “Use of terms” section , starting line 2402.
I do not understand the definition of “cumulative effects” and would like to ask the forum to further discuss the appropriateness of the current definition. My own view is that cumulative effects surely cannot be  linked to the “presence of multiple LMOs or their products in the receiving environment”, as currently started in the text.
The entry of “combinatorial effects” needs revision too as the focus is on the identification of adverse effects on the environment and human health, while what is described  under this point in the text refers to  of gene expression, and gene interactions  within the LMO!

lines 643 – 665, Text box “LMOs developed through RNAi-based methods to reduce gene expression”: The information of relevance for the risk assessment present in the text is repeating the more general points made elsewhere in the body of the text of the guidance. Suggest that the text box is deleted.

lines 726 – 759, Text box “LM crops and the use of the herbicides”: Suggest to delete the textbox, as this topic was discussed in the forum and clearly the majority view was that the risk assessment of the herbicides is out of the scope of the LMO risk assessment and it is covered by other regulations in place. In addition, the assumption presented in the text that herbicides are mixed during application is entirely speculative.

line 1176, section on  “Combinatorial and cumulative effects”, and as mentioned in multiple places in the this section. It is not clear what exactly is to be considered,  especially because the focus is placed on LM plants, while a broader scope seems more appropriate when considering that the guidance is intended to cover RA of LMOs in general.

line 1369 “omics” technologies: Edit the sentence by adding “if validated for regulatory use” :
On a case by case basis, available information from “omics” technologies, for example, “transcriptomics” and “metabolomics”, [if validated for regulatory use], may help to detect phenotypic and compositional changes (e.g., the production of a novel allergen or anti-nutrient) that cannot be detected using a comparison with field grown plants under suboptimal conditions.

With best regards,
Ana

Dear Helmut, dear colleagues
Let me extend my appreciation to the sub-group on their Herculean task to seek out harmonious text from a confluence of opposing philosophical positions, worldviews and values. It is inevitable that our bias blind spot finds fault in the work of others with far more alacrity than the critiquing of our own efforts.
I will not dwell on my strong support for the comments #7780-7784, which attempt to answer the important, but unasked question “has the new text given adequate voice to all of the comments raised during the roadtesting?” Instead, I will take instruction from our esteemed moderator and confine myself to concrete text proposals on the new text.
Lines 310-311. The reference to “hazard pathways” is jargon that means little to the uninitiated. I would suggest replacing this sentence with “The information required for small scale releases of an LMO, such as confined field trials, is typically more limited and focused on the effectiveness of proposed [risk management] controls to limit the spread and persistence of the LMO or its genetic material.” This suggested text highlights a more critical, unaddressed issue, namely declaring the starting assumptions for risk assessment of a confined field trial: either 1) presumption that no controls or limits exist (ie it is equivalent to an unrestricted commercial release) or 2) consideration of proposed controls and limits that are typically outlined in an application for regulatory approval (eg limits on duration, area, number of locations, access, use as food or feed; and controls such as isolation distances, pollen traps, storage, transport and disposal, cleaning of equipment, contingency plans, reporting). This latter setting provides a more realistic context for risk assessment of confined field trials as currently practiced.
Lines 311-312. This sentence (“Information collected during small-scale releases may provide useful information for further risk assessments of large-scale releases.”) gives us a warm inner glow that our regulatory approach is measured and sound. Unfortunately, the reality based on past experience is that field trials are primarily a standard tool for breeders to test agronomic performance but have not revealed much from a risk assessment perspective. The only indicator of any potential use is indications of substantial equivalence as a proxy for unintended effects. However, given the allergic reaction of the AHTEG(s) to the term ‘substantial equivalence’, I would kindly suggest deletion of this sentence unless real world examples can be provided.
Line 385. After the title of this box there should be a clarifying statement that indicates that the focus is on release of an LMO into a centre of origin or centre of genetic diversity for the same species or a sexually compatible species. Otherwise, the rest is problematic as it raises new issues such as the release of LM cotton in a centre of genetic diversity for cowpea.
Line 390. Replace ‘irreplaceable’ with ‘valuable’ as the CGIAR centres have placed great emphasis on maintaining extensive seed banks of crop varieties that are capable of ‘replacing’ losses from greater threats such as urbanisation, land clearing and agricultural intensification. In addition modern biotechnology now offers us the bold opportunity to resuscitate ‘original genotype[s]’ and replace some of the devastating losses of genetic diversity over the last 4 billion years (see comment on lines 510-514).
Lines 392-394. Delete these two sentences. Firstly, a centre of origin is not a protection goal, but a scientific inference. Secondly genetic diversity is not a universal protection goal. ‘Universal’ protection goals to reduce genetic diversity include elimination of smallpox and polio viruses. In addition, there seems to be determined efforts to reduce genetic diversity of Ebola, HIV, mosquitoes, Striga as well as numerous other pests, weeds and pathogens; despite all of these organisms providing ‘ecological functions’.
Lines 395-397 should be deleted as it imposes obligations outside the mandate of the AHTEG. “Human intervention in centres of origin..…” has no qualifier and could imply any human based non-LMO activity. Does the reference to “stability of original genotypes” now require culling of any rogue individual with a mutation that differs from the parental genotypes? This would seem to defy natural biological processes and even contrary to our celebration of genetic diversity.
Lines 399-400. “spatial and temporal characteristics” is unclear. Does it encompass considerations such as climate change, land clearing, changes in land use (eg conversion of agricultural land to urban) etc?
Line 401. The use of ‘significant’ is this context suggests pre-judgement of the consequence and likelihood assessments. It should be replaced by ‘evaluated’.
Lines 405-408 could be simplified to “Baseline data, models to understand and simulate gene flow, and methods to identify and quantify possible consequences related to the introduction of LMOs in centres of origin and centres of genetic diversity may be useful information elements for the risk assessment”. These elements are not ‘fundamental’. For example, a conclusion of no substantive adverse effects from 100% introgression may render any consideration of gene flow rather moot.
Line 455. Delete ‘working’ as redundant. I don’t think we intend  to generate ‘non-working’ hypotheses. I support Jack’s intervention (#7793) that hypotheses be preferred to hypothesis.
Lines 474-480. Here I do support the sentiments expressed that we tread carefully with this advanced text that has been agonised over many years and retain the original version. This new text obscures the important point of the original text that more clearly expresses, namely, the need to assess adverse effects under relevant management conditions (eg two sprays per season on LM cotton as opposed to up to 14 sprays of broad spectrum chemical sprays on non-LM cotton).
Lines 510-514. Another example where I support retention of the carefully crafted original text. The new text presents a rather problematic perspective as alternative comparators are required with current (not necessarily future) techniques. For example, one application to the Australian regulatory system proposed reconstruction of a putatively novel virus from genomic fragments present in a bat genome. The non-LM form of the virus may not occur in nature, and if extinct, we have no certainty of what the non-LMO sequence was.
Lines 512-513. I respectfully suggest that “In such situations, the characterization of an LMO may be similar to that carried out for alien species” be replaced by “In such situations, the characterization of an LMO may be similar to that carried out for biosecurity risk assessments” as native species, not only alien species, can be problematic and subject to risk assessment.
Lines 584-603 are a rather idiosyncratic interpretation of “Types of adverse effects”, which I have not encountered in LMO risk assessments in the Americas, Africa, Asia or the Australia-Pacific region. Part of my difficulty is the disappearance of anything relating to ‘adverse’ in the text box (other than the title). Indeed the text would seem to be more relevant to effects within a causal pathway, rather than an outcome at the end of the causal chain that is adverse (an assessment endpoint if my interpretation of difficult to understand text earlier in the guidance is correct). This problematic omission of ‘adverse’ in the text may help to explain my additional comments below.
Line 586-587. Direct [adverse] effects always require a causal chain of events, even if it is a rather simple chain such as contact (exposure) with the LMO by a desirable species, consumption, followed by expression of toxic effects. The only non-causal chain of events would be if the release of an LMO is in itself an adverse effect, which may be the case if your values object to the mere presence of an LMO. But in these cases the risk assessment is rather trivial as the existence of the LMO is immediately unacceptable and therefore does not rely on the detailed guidance provided here.
Lines 588-590. Indirect [adverse] effects would make more sense if used to describe the situation where the adverse effect does not require direct exposure of our object of protection to the LMO (eg an LMO that increases the intensity or frequency of fires; or restricts movement of water such that hydroelectric power to a remote location is impaired as demonstrated by water hyacinth).
Lines 591-595. Typically regulatory legislation refers to consider adverse effects in the short and long term. These are more accessible descriptors.
Lines 584-603. In summary, I would like to boldly suggest that types of adverse effects might be better served with using text from Article 7 of Burkina Faso’s Decree on risk assessment of LMOs, which more clearly provides a linkage to ‘adverse’ and is more consistent with regulatory approaches used worldwide, namely:
The following elements are considered potential adverse effects:
o Damage to health and human and animal safety, including:
• toxicity or allergenicity;
•diseases ;
• ailments or injuries;
o Damage to environment including:
• toxicity to non-target organisms;
• loss of biodiversity, including loss of species diversity and genetic diversity within   species;
• the harmful effects of pathogens, pests and bad most important herbs;
• adverse effects of new species of pests;
• disruption of biotic communities;
• degradation abiotic environment.
Lines 644-655. Suggest replacing “RNA interference (RNAi) refers to a process that silences gene expression and is triggered by double-stranded RNA (dsRNA) molecule(s).” with “RNA interference (RNAi) refers to a process that silences RNA translation that is triggered by double-stranded RNA (dsRNA) molecule(s) that contain sequences complementary to short segments of the target RNA.” Part of the reason is that gene expression in terms of RNA transcription is unimpaired and some indication of the basis for target specificity should be included.
Lines 650-655. The elements 1) to 5) do not necessarily need to be evaluated as molecular changes do not constitute an adverse effect indeed are a poor proxy for phenotypic changes that provide a more direct information on the potential for an adverse effect.
Lines 656-661. The point of this section is somewhat lost on me as it seems no different from risk assessment of any other type of genetic modification.
Lines 662-665. The extensive uncertainty on the specificity of binding with respect to number of nucleotide matches, number and location of mismatches tolerated, effect of nucleotide bias, influence of secondary and tertiary structures renders any regulatory interpretation of ‘bioinformatic’ data as highly speculative.
Line 726. This box provides a misleading impression only LMOs are resistant to herbicides, rather than a natural occurrence that has been selected for in non-LMO breeding programs. Therefore, this box could be replaced with a simpler statement that “An LMO resistant to herbicide(s) should be managed according to national regulations for the use of herbicides.”
Lines 746-747. What is the target organism implied by this sentence? Unless clarified, it should be deleted.
Line 791. This box neglects to provide sufficient guidance on the full range of potential adverse effects to human health. In addition to toxicity, allergenicity, reduced levels of key nutrients, increased levels of anti-nutrients there is also production of structures that cause physical injury (eg thorns), endocrine disruptors, psychotic effectors, products that give rise to contact dermatitis etc.
Many apologies for adding to the burden of the sub-group, but I do wish the AHTEG all the best and a useful face-to-face meeting in Mexico.
Paul

Dear all,

I state my sincere thanks to everyone who has provided thoughtful comments and views.  In particular, my thanks go to Helmut for his willingness to continue to chair and moderate this challenging process. 

First, I wish to support the numerous concrete suggestions from a range of experts who are knowledgeable and experienced in risk assessment of LMOs.  In particular, I support interventions: 7780, 7781, 7782, 7783, 7784, 7801, 7808, 7810, 7812, 7813, 7815, 7819, 7820 and 7821. There is enough commonality among these proposals to form an excellent basis to improve the guidance and perhaps approach a wider consensus. 

However, longer-term we must confront the realities that (a) the text is not improving after many years, and (b) the process may be the root cause.  I counted at least 11 submissions that noted that the text is not improved.  In addition, interventions 7781, 7782, 7783, 7810, 7817, 7818 and 7821 all to some degree point to problems with process.  This, in no way, is a reflection on the chair and his exemplary efforts to make progress.  It is, however, a critical point that needs to be conveyed to the MOP for their considerations. 

Finally, and with all due respect to Helmut, I must correct the statement that the text used to create this version reflects "agreement" within the AHTEG.  This point was also noted in interventions 7817 and 7818 by a former and current member of the AHTEG.  Again, after multiple years, the text is most accurately characterized as "that which was present when time ran out".  Thorny, if not intractable, issues remained present despite that efforts of the chair.  Many of us agreed to drafts with the understanding that we would return to points of disagreement at a later date.  It was with this understanding that drafts were produced. 

I thank everyone for the continued rich discussion on approaches to and the substance of guidance on risk assessment for LMO's. 

Tom

Dear Colleagues

I would also like to thank Helmut who has for a large number of years now ably borne the burden of chairmanship of the group. The documents upon which we work come not from a process that is perfect - but no process could be - but from a group that can be highly polarized on some issues.

I believe that we all want excellent guidance (and not guidance for the sake of it) and are still committed to the process. While Tom recalls those voices who see defects in the process, I'd like to add my voice to the side of the ledger that believes this is perhaps the best possible process we could achieve in a flawed world, under the challenging circumstances of a broad group of experts with strong and sometimes divergent views.

I also want to thank the Secretariat who, over the many years, have been so dedicated to helping the Parties develop the Guidance they sought. The testing confirms that it is going in the right direction even if it is not the vision held by all.

Best to all
Jack

Dear Colleagues-

I noticed Helmut in his post [#7785] included the ‘few issues that were highlighted as still lacking in the Guidance, ‘for example ‘its overall objective, the target audience, key messages to convey, the level of detail given to specific considerations, and its structure/organization’.  He wrote in that post: ‘Concrete text proposals to address these and any other remaining gaps in the Guidance would be very useful and I invite you to make suggestions along these lines as well’. 

I am not sure that the participants in this forum understood from the opening message that concrete text proposals are invited on these issues and other remaining gaps. (Where were these issues highlighted?) However, I do think many comments in this forum (among the 25 participants or so) do address some of these issues, although perhaps not always with concrete text proposals.

On the topic of ‘its overall objective, target audience, key messages to convey’, I have noticed again in the current forum discussion, as there has been in the past, a concern that the Guidance will not be useful to novice risk assessors.  And I note in the revisions to the guidance, the addition of a few sentences in the ‘Background’ section of the roadmap that I believe are meant to address in some way this concern.

Lines 158-164:  Those inserted sentences, which were not in the tested version of the roadmap, read: 
‘The Roadmap may be useful as a reference for designing and planning risk assessment approaches.  It may also be useful for risk assessors when conducting or reviewing risk assessments and as a tool for training.  Based on its use, the Roadmap may also be useful for identifying knowledge gaps.  The Roadmap introduces basic concepts of risk assessment rather than providing detailed guidance for individual case-specific risk assessments.  In particular, the ‘elements for consideration’ listed in the Roadmap may need to be complemented by further information during an actual risk assessment.’

This addition gives the impression that the roadmap is trying to do many different things, but it is no longer is a ‘Roadmap’.  I believe many of us have had the impression that the ‘Roadmap’ is meant to provide detailed guidance for individual case-specific risk assessments, and this has created much of our angst because we truly do not see how it can be used in this way.

My concrete text proposal, therefore is to change the language in the last sentences above (Lines 162-164) to say: 
‘The Roadmap introduces basic concepts of risk assessment. The road map [ADD: does not provide] [DELETE: rather than providing] detailed guidance for individual case-specific risk assessments. 

We should also give some thought to changing the name of the ‘Roadmap’ to something else that does not imply that this is any sort of map.  Perhaps we could call it: ‘Basic concepts for risk assessment’.

These changes might help to deter the novice risk assessor from assuming they can somehow simply follow this Guidance and conduct a risk assessment.  My concern has always been that novice risk assessors will assume that the Guidance is somehow prescribed or required for parties to follow, even though it is stated as not.

Thanks,
Karen

Dear Helmut/AHTEG - Thank you for the opportunity to comment on the proposed revision of the Guidance.  I'm however inclined to think the suggestions require a more in-depth analysis given the importance of the document.  My comments therefore are general in nature:

- My impression is that the boxes provide some level of detail on specific topics/LMOs, thus contradicting the Background (in particular line 162-168). 

- There is no coherent flow between the main text and the boxes - a well elaborated Guidance should be understood by a novice to be applicable to all LMOs, on the contrary boxing specific LMOs gives an impression a different assessment method is employed in this example;

- In some instances, information in the main text is merely paraphrased in the boxes, whereas I would suggest the boxes are rather used to demonstrate how the RA principles are applied in practice;
 
- Food safety assessments are not within the remit of the CPB but the Codex Alimentarius Commission.

As such, most of the boxes as currently written do not add value to the original text.

Best regards

Kelebohile Lekoape

Greetings to all, and apologies in advance for this long intervention.

I have followed these boards over the last two weeks with great interest.  I would first like to thank the CBD Secretariat for the opportunity to comment on the current draft and changes to the text of the Roadmap.  I’d also like to acknowledge the efforts of the AHTEG and the chair.  There is no doubt that attempting to put together a document like this is an arduous task.

Before I get to specific comments on the text, I would like to address the overall purpose of the online forum and what we have been asked to do in commenting only on the specific changes.  I am in full agreement with the posts of Boet Glandorf, Piet van der Meer, Karen Hokanson and others regarding the idea that we should comment only on revisions.  This suggests that the remainder of the text has somehow been agreed or finalized, and that these changes and additions can be substantially considered in isolation.  This is simply not realistic or productive.  The process of producing this guidance document has been hampered from the beginning by our seeming inability to allow for a discussion and agreement on what, exactly, the guidance is intended to accomplish and what “success” looks like.  Changes are made from one version to the next, sometimes with little or no explanation and frequently not in keeping with suggestions made in the online forum.  The process is confusing, even for members of the AHTEG.  This is not a new issue and it has been brought up many times both in these forums and in the face to face meetings.  With respect to the chair’s additional instructions on this matter, it is not possible to simply add a line or two of text to the preface of a 100 page document that has been created without an agreed and coherent vision and somehow “fix” the problem that there is no agreement among Parties, AHTEG participants and commenters on the forums as to what such a document should look like.  This is not a trivial issue that can be easily addressed without applying significant effort.  None of this is to imply that the effort should not be made, and I fully recognize that this input may not be welcome.

Regarding comments on the specific text, I am very inclined to agree with the comments of Yann Devos, Piet van der Meer, Karen Hokanson, Boet Glandorf and others.  My specific comments with line notations are contained below.  In general, it is clear from the draft that the additional text boxes are not contributing positively toward the stated goal of the document.  If there is substantial interest in these subject areas, then assembling a list of useful existing references would be a far better use of the groups time and energy than attempting to draft short sections of text that will invariably fail to fully address the issue and require a great deal of consideration to draft accurate and acceptable language.

Line 305-312:  This bullet is attempting to address the varying information needs for confined trials versus large scale introduction to the environment.  However, as stated it’s not accurate.  I’m not sure what the aversion to the mention of small scale releases and confined field trials is, but if we want to talk about confined field trials we should use the term – and I agree with others that there is a lot of confusion based on the differences in how assessments are done for these different activities.  It will be difficult, if not impossible, to provide any sort of detailed guidance that pretends to address both conditions.

Centres of origin and centres of genetic diversity:
This box is essentially outside the scope of the document.  With all due respect to the work of the authors, it doesn’t provide any useful guidance for conducting ERA.  Instead, it supposes to dictate to countries what their level of protection should be and what their standard of evidence must be in support of decision making.  Neither of these things is the remit of the AHTEG, nor the purpose of the guidance document.  The text is far out of line with the existing text in the document.

Lines 392-394 engage in a level of political discourse that is not appropriate for a guidance document on risk assessment coming out of this forum.  It is not appropriate (or desirable) for the AHTEG to be telling Parties what their protection goals are, and to suggest that somehow the existence of centres of origin “transcends” national protection goals seems like an intrusion on the sovereignty of governments to set their own protection goals.  The words “transcend” and “universal” do not appear anywhere in the CBD, or in the cited International Treaty on Plant Genetic Resources.  This guidance is not for the purpose of telling Parties or other governments what their protection goals should be, and it is certainly not the appropriate forum for expanding the scope of those agreements.

Lines 395 -397:  First, this is not guidance on all “human intervention.”  This is guidance for conducting ERA on LMOs.  Second, it is not up to the AHTEG to tell Parties what level of impact is acceptable when evaluating LMOs in the context of their sovereign decision making process.  It’s unclear, scientifically, what is meant by “stability of the original genotypes.”  So, beyond attempting to set policy for Parties, the text here introduces scientific confusion rather than clarity.

Lines 398-401:  Again, it is not the purpose of this guidance to identify to Parties what they should consider to be significant.

Lines 405-409:  Although access to “adequate” baseline data is an easy concept to agree with, it is not particularly helpful to list it here.  Models to understand and simulate gene flow are not required, certainly not universally required, to identify consequences.  There are many ways to address introductions, including through risk management measures which may preclude the use of modeling.  If you are conducting a small scale field trial, for example, it may be far easier to simply remove the flowers than to produce a gene flow model. The idea that quantification of consequences is a crucial component of ERA is also not correct.  Most ERA for LMOs are qualitative – and have been demonstrably predictive and effective in informing decision making.  The need for quantification will depend on the protection goal, the tolerance for harm, and the needs of the decision maker for any specific case.

In short, this box needs to be removed from the text.

Line 455: should be “hypotheses” not “hypothesis.”
Lines 474-480:  These changes completely alter the meaning of this section of the document, and make the paragraph inappropriate for a discussion of comparators.  The original wording indicated that an LMO should be compared to a realistic alternative.  For example, Bt maize is not planted in place of pristine natural grasslands, it is planted in place of non-Bt maize.  So the comparative assessment should consider the realistic alternative.  As the paragraph is worded now, it says that risk assessment must reflect the standard management practices expected to be applied to the LMO, but no longer makes any mention of the standard management practices applied to the comparator.  The original text is better.

Lines 510-514:  Line 510-511 suggests that the technique used for making the LMO is somehow relevant to the comparative approach.  This is incorrect.  There is no reason to believe that future techniques will be any more or less likely to produce LMO’s that can be comparatively assessed.  I expect that what is intended here is to suggest that future LMO’s may not have a ready comparator in the environment.  While this sounds intuitive, it is also wrong. This is a misrepresentation of the comparative approach as well as the process of assessment for alien species (see Keese, Robold, Myers, Weisman, & Smith, 2013).  The point of the comparative approach is to focus the assessment on characteristics of the LMO that are different than what are seen in the environment.  This is perfectly applicable to assessing even alien species.  For example, when introducing a non LMO plant species into the environment you wouldn’t worry about characteristics that are shared broadly among existing plants in the environment, such as photosynthesis, cellulosic content etc.  You would focus on what is different about the plant that may lead to harms in the environment, and typically focus on characteristics that are known to cause environmental harms.  All of the questions you might ask regarding an alien species are useful only in so far as you can compare them to what is known about species already present in the environment.  The Australian post border weed risk assessment model cited above, and used to score introduced plants to determine if they pose a risk to the Australian environment, requires that each characteristic be scored in comparison to other species.  The idea that the only comparator is a near isogenic line already present in the environment is dispelled in lines 492-508.  Even the “approaches” being described in lines 481-491 are only considering the use of comparators for specific experiments or components of the evaluation. 

Keese, P. K., Robold, A. V, Myers, R. C., Weisman, S., & Smith, J. (2013). Applying a weed risk assessment approach to GM crops. Transgenic Research. http://doi.org/10.1007/s11248-013-9745-0

Types of effects box
Lines 584-603:  There is nothing incorrect about what is contained in these boxes.  Many different types of effects are described.  However, nothing in this box provides any guidance as to how these different types of effects are considered in ERA.  For example, none of these types of effects arise independent of the characteristics of the organism, and even so called delayed effects require a change in some first order interaction between the LMO and either another organism or component of the environment.  So the assessment for any and all of these effects follows the same path as the assessment for direct effects, just with consideration for the appropriate second and third order interactions that are involved in potential environmental harms.  The way these are listed here without any sort of context leaving the novice reader to guess as to what they are expected to do with this knowledge, is very likely to cause confusion and interfere with the function of the document.

It's also worth noting that we are citing a particular country’s law for these definitions.  I don’t think it is necessarily good practice to introduce national legislative definitions.  If the AHTEG will be citing national laws, there are many others that might be introduced as well.  I think it would be better to find other sources to cite definitions of effects.  Many are available.

RNAi box
Lines 643-665: In general (as many others have already stated), there is nothing listed in this box that can’t be applied to “traditional” LMOs that express proteins to produce novel phenotypes.  This is unstated, probably because it highlights the lack of a need for this box.  Again, there is no context given for any of the different pieces of information listed here, and there are no explanations of terms.  Specifically, epigenetic inheritance, which again serves more to add confusion than to provide any clarity or insight into the risk assessment process related to this topic.

LM crops and the use of herbicides box
Lines 726-759: This is an area where the world has accumulated significant experience with both risk assessment of herbicide tolerant LM crops, and with the regulation and use of herbicides.  None of this experience is drawn from or even mentioned in the box.  It is nowhere mentioned that herbicides are typically regulated under a different authority than LMOs and that those regulations normally address use patterns.  It is not mentioned that herbicides may be used in the absence of LMOs, that “tank mixes” applying multiple herbicides are very common and that any and all possible synergisms from using multiple herbicides can and do occur independent of LMOs.  Herbicide use patterns change often, and the idea that an LMO risk assessment is the best place to assess the impacts of herbicide use is a dubious one at best.  However, if we are going to discuss this, we need to have a realistic discussion that presents the reader with some understanding of the actual relationship between HT LMOs and herbicide use.  This box fails to accomplish that. 

Risk assessment of LMOs and human health
Lines 792-805:  As others have mentioned, this box is a discussion of food safety (and a woefully inadequate one at that).  It doesn’t belong anywhere in this guidance document for ERA.  A few people have suggested that incidental human ingestion is a significant environmental issue – however this is not a component for many assessments of environmental impact.  And, even if it were, a reference to food safety assessment guidelines would be far better than trying to replicate those guidelines here.  Lines 795-799 are very problematic, misrepresenting the process of collecting experimental evidence to determine toxicity.  It’s unclear what we are assessing here, whole plant toxicity?  Novel protein toxicity?  And why are we assessing this?  Human health issues created through environmental exposure are treated much differently and certainly don’t rely on nutritional composition analysis.

Lines 944-950:  The change from “moreover” to “on the other hand” is incorrect.  This sentence is in agreement to the previous sentence not in contrary to it.  The introduction of the threshold of acceptability is also unnecessary.  It doesn’t make the text better than it was before.  It’s worth noting that this paragraph is accurate, and in direct contradiction with the box on centres of origin.

The purpose of a guidance document is to provide clarity on how to conduct a risk assessment.  This should not be confused with having a brief mention of every conceivable aspect of every conceivable situation that may be subject to assessment.  The inclusion of the text boxes seems, at this point, to provide nothing of value to the document while introducing incomplete and occasionally inaccurate descriptions of certain specific issues.  Instead of trying to wordsmith these boxes, it seems like time devoted to addressing some of the more fundamental issues remaining in the document would be better spent.

Dear Colleagues,

I do too truly wish to acknowledge the hard work of the AHTEG and the Chair, and the Secretariat over these many years.  I am in awe of your ability to listen so patiently to these deeply divided views, and I know you have made a great attempt to take all sides into consideration.

Contrary to Jack’s comment in his last post [#7824 ], I do not believe that the results of the testing indicate that we are headed in the right direction.  I believe the testing process was flawed, and the results have been somehow misrepresented. 

This troubles me so much that I have taken a close hard look at the numbers that were assigned to the ‘level of agreement’ from the testing questionnaire.  Many of us expressed concern in the online forum when these results were first presented because too much stock should not be put into a survey of this sort.  One point I have noticed is that the numbers can be interpreted in very different ways to support a very different position. 

The Secretariat in the results that are posted on BCH, has averaged the ratings according to ‘Developed country Parties’, Developing country Parties’, ‘Non- Party governments’, and ‘ Other nongovernment organizations’.  The graphic presentation of these groupings show that ‘Parties’ on average rated the Guidance much higher than ‘Nonparties’, and ‘Developing country Parties’ rated the Guidance highest of all.  This has been taken as an indication that ‘Developing country parties’ who should be a critical target audience for this Guidance, find the Guidance most useful and practical.

I felt so concerned that this is a skewed representation of the results, that I went back to those numbers and grouped them in a different way.  My thinking is that the true distinction in these ratings has not to do with developed vs. developing countries as it does with experienced vs. inexperienced risk assessors.  To test this notion, I assigned to each party and non-party who tested the Guidance a number that represents the number of crops approved by that country.  This may be a crude measure of experience, but surely those countries that have approved more crops have more risk assessments on the record.  What I found is that the average rating of the Roadmap by countries that have no approvals on record was 4.25.  The average rating of countries that had 1 or more crop approved was 3.2.  And, 13 out of 20 of those countries that have approved 1 or more crop are Developing countries. 

This leads me to believe that the countries that tested the Guidance and gave it the highest ratings (and the least amount of comments) also have the least experience with risk assessment.  And therefore their rating of the Guidance as practical and useful was not based on their experience, but based on the fact that they do not know better. 

I heard a good analogy for this from another AHTEG member: You can give someone a toy shovel and tell them to dig a large hole in the ground, and if you ask them afterward if the tool you gave them was useful and practical, they might say yes.  But if you give them a real shovel and tell them to dig the same hole, they will realize that the toy shovel was not really very practical or useful.

Now I have truly strayed from concrete text proposals.  So I guess this is where I should quit.

Thanks to everyone for listening to my thoughts.

Kind regards,
Karen

Dear Helmut and colleagues

The mandate that the AHTEG received was to address the comments received during the testing.  The AHTEG and online forum has address comments that were more specific to a particular section of the document and even not all of them, but there are two other groups of comments classified as 'Overall evaluation of the Guidance' (Category B) and 'Suggestions for substantive changes without a specific location in the Guidance' (Category D) that have not been addressed. Many of those comments are much in line with the general comments that Yann made [#7780] and that should be address.  Also as Sol stated [#7817] all the comments that have partially been taken or not accepted should have a justification why not accepted or modified, I have not seen an adequate table of that.

I agree with posts #7780, #7783 and many others that follow of similar opinion that the text boxes as written are not useful.  Where they have definitions of terms, those should be moved to the 'Use of Terms' section.

The box related to human health is unclear what it should contain.  Considerations on food are best addressed by Codex so it suffice to provide a reference. In terms of other aspects, there would have to be an agreement of kind of topics are relevant under the Protocol.

Herbicide related topics are addressed by other regulations and their experts, this is not the place for that.

Dear Chair, Dear Secretariat
I am deeply concerned with the post [#7827]. I believe that the language and metaphor used are far from being appropriate and are in fact insulting to developing parties. I refrain myself from directly replying to it but will impatiently wait for your reaction to that issue.

Warm regards,
O.A.El-Kawy

Dear All,

Thanks for the opportunity to provide some comments in this latest discussion.  I will keep my comments brief. I appreciate the many thoughtful comments, and I would like to briefly support the comments of Yann Devos, Karen Hokanson, Ayako Yoshio, Boet Glandorf, Hector Quemada, Paul Keese, Sol Ortiz, Piet van der Meer, and Andrew Roberts.

Regarding the overall process for development and drafting of the guidance, it is clear that the process is part of the problem why the document grows ever larger in size yet ever weaker in its clarity to help novice risk assessors to do an environmental risk assessment consistent with Annex III of the Cartagena Protocol on Biosafety.

Although the chair opened this session by calling for concrete text proposals on only the changes made, on 2 May he acknowledged that the Guidance document draft still lacked clear statements of "its overall objectives, the target audience, key messages to convey, the level of detail given to specific risk assessment considerations, and its structure/organisation."  It is incredible to consider the amount of time spent on this document when the process has not yet settled on such key aspects. 

- David Heron