INTRODUCTION

There have been over 2000 reviews of field trials of genetically engineered organisms, most conducted during the last three years. It is therefore an appropriate time to review biotechnology risk assessment, its needs, accomplishments, and future needs.

RISK ASSESSMENT RESEARCH

The role of risk assessment advancing the development and commercialization of biotechnology is both difficult and simple to evaluate. Simple, because there is little doubt that biotechnology products, like all new products, will be evaluated at some governmental level before they are commercially produced. Difficult, because of the problems inherent in predicting the consequences of releases of genetically modified organisms into the environment (Witt, 1990; Sharples, Levin and Strauss, 1983; Tiedge et al., 1989). Thus, biotechnology risk assessments are among the more difficult cases to address.

Because biotechnology was commercialized initially in the US, this difficulty was recognized early on by the US agencies assigned the regulatory responsibility for evaluating new products for the marketplace. A US committee established a coordinated framework which laid out ground rules and a road map detailing how and by which agency such a review of new biotechnology products was to be performed (Office of Science and Technology Policy, OSTP, (1985). Each agency then had to decide on the information requirements for evaluation. A series of Interagency meetings resulted in the establishment of a Points to Consider Document which contains the basic information needs for biotechnology risk assessment (USEPA, 1990 ). As each agency began to conduct risk assessments, it became clear that while some of the required data and some of the procedures necessary to produce appropriate risk assessment data were available, much of the basic information and many essential techniques were not available as standard procedures or had not yet been developed. A similar chain of events occurred in Canada, Europe and Asia, as well.

In recognition of these shortcomings, government agencies responded by providing funds to support basic and applied research in the area of biotechnology risk assessment. Initial funding was on an ad hoc basis with funds provided for answering very specific problems with specific products, such as the ICEMINUS bacterium and its aerosol characteristics by USEPA in 1986 and SNOMAX in Canada in 1987. However, as it became clear that more products and a wider variety of products would soon be reality, funding increased and the number and variety of projects increased. In the US the USEPA initiated the first formal Biotechnology Risk Assessment Program (1983), followed by USDA (1992). Canada has been funding biotechnology risk assessment since 1983, and a European program was initiated in 1991 (Levin and Angle, 1994). EPA has allocated approximately 10 million dollars and USDA approximately 8.5 million to Biotechnology Risk Assessment. The objective of these funds was to develop protocols for generating improved data to assist risk assessors and risk managers in the decision making process of evaluating and approving the large variety of genetically engineered products and processes.

Thus, the process of risk assessment in biotechnology began as a scientific endeavor to produce an answer to the question: What is the probability of an adverse effect from this product. Adverse effect was defined by the agency involved (e.g. EPA, general ecological effects or USDA specific cattle and crop effects, Environment Canada, toxicity). Increasingly; though, it soon became apparent that the public perception was a very important component of risk assessment and management, and the annual risk assessment symposia attempted to bring together the scientists, risk assessors and the stakeholders (i.e. the public and the industry), with the objective of gathering information to guide funding agencies, ascertain public opinion about the perception of the credibility of the process and the acceptability of the findings and the products and identify future research needs and priorities.

The concept of a series of bilateral meetings between Canada and the United States began at joint Biotechnology Risk Assessment Symposium in 1994 (Levin and Angle 1994). 1996 marks the third year of this symposium. Participants at this symposium received a questionnaire seeking their opinion on science issues and general risk assessment and risk communication issues. The questionnaire was distributed and discussed at the last session of the conference. Although there were almost 250 attendees at the conference, only 100 attended the last session. 32 (32%) returned the questionnaire. The respondents were asked to identify their areas of specific research interest (i.e. microbes, plants, or animals) and respond to 16 items answering the question:

"How do you think that research work conducted to date in the following areas has contributed to clarification of the environmental/biotechnology risk assessment debate?"

Responses were limited to "poor", "well" "very well" or "not applicable". Because all respondents did not answer all questions (e.g. plant specialists felt uncomfortable answering questions about animals); it was necessary to use a ranking system to permit comparison of responses. For analytical purposes, points were assigned to each level of response (i.e. poor was set equal to one point, well at two and very well at three). The maximum number of points therefore was equal to 3 times the number of respondents in a given category. Thus, receiving only 33.3% of the possible points indicates that all respondents thought the answer was poor. Table one contains the percentages of all respondents to all questions. It should be noted that the group contained, in addition to researchers in the specific areas, representatives from industry and the general public.

Table 1. Results of survey*

	Percent for	Percent for	Percent for
Identification	52.2	48.7	76.9
Survival	54.6	46.5	76.0
Gene Expression	53.8	48.3	56.0
Competition	43.4	36.2	69.5
Gene transfer	44.2	42.7	82.1
Gene stability	45.8	40.2	59.1
Fate	41.7	43.8	59.2
Effects	44.7	36.7	83.3
Microbial Ecology
Taxonomy	45.7	36.4	81.8
Monitoring	41.6	48.7	62.9
Risk Assessment	55.7	44.6	93.7
Risk	32.1	33.0	71.4
Information	34.2	17.4	56.1

*>66% = very well, 66%-33% = well, <33 = poor

SUMMARY OF FINDINGS

It is clear that the majority of respondents felt that the contribution of risk assessment research to risk assessment falls in the well or very well done categories. The overall average for microbial and plant specialists were almost identical (42.2 and 40.1, respectively). However, these two groups also felt that risk communication and information dissemination was poorly handled (32.1 and 20.2 respectively). Microbial specialists felt that greater emphasis on research in the general area of microbial ecology is required. This area received an average score of 28.7 (poor). Animal specialists, however, were much more complimentary (average score: 71.4), with the lowest scores in the areas of gene expression, stability and fate still in the well category (58.2).

One general theme was expressed by many respondents: The use of molecular techniques, either to monitor the fate of organisms, the dispersal of transgenes or to monitor community structure has significantly contributed to environmental risk assessment of biotechnology products. Data suggesting that gene flow occurs naturally in nature was also cited as a significant factor in the risk assessment process. This may reflect the general uneasiness reflecting the value of laboratory data for predicting what really happens in nature.

Two areas, risk communication and information dissemination, received poor rankings from both microbial and plant specialists. However, since most of the 2000 filed trials conducted have involved plants or microbes, this may reflect experience related to field trials and encountering public resistance. The surprisingly high scores for animal specialists is perhaps due to the fact there have been few trials and therefore limited experience. Further, animals are kept in enclosures and are not perceived to pose the same ecological or public health threats as released plants or microbes. In general, animal specialists felt that accomplishments to date were very positive about risk assessment (risk assessment scored 93.7, highest in any category) and risk communication efforts.

A number of comments were received relative to the overall relationship between risk assessment research and the conduct of risk assessments. Some researchers and representatives of the public and private sectors felt that there should be a stronger link between the research and the assessment. One suggestion was that each research project should state how the project is related to assessment. This would link funded research to risk assessor's needs. Risk assessors comments indicated that while most of the projects produced useful information or techniques, there is a need for closer coordination which would permit focussing research directly on current problems resulting from new product development. The risk assessors also felt that the symposium offered them an opportunity to hear scientists discussing their work, other scientist's opinions about techniques and findings and permitted them to gain an overview of what was really possible or not possible in terms of data acquisition or predictive ability.

It is not surprising that there is a general consensus that additional risk assessment funding is needed. Biotechnology is a 500 million-dollar industry: risk assessment research funding currently amounts to less than ½ of 1%. In addition, the need for better risk communication and information dissemination and the need for more public input into the decision making process were stressed as critical by attendees at the symposium. This need was identified in 1990 by the OSTP (OSTP, 1990) and was recently highlighted in a report to the US Congress (Presidential Commission, 1997).

REFERENCES

Levin, M.A. and J.S. Angle. 1994. Biotechnology Risk Assessment: USEPA/USDA/Environment Canada/AAAF Canada. 3-11. In: Proceedings of biotechnology risk assessment symposium, June 22-24 1994. Levin, M.A., C. Grim, J.S. Angle eds. University of Maryland Biotechnology Institute, Baltimore, Md.

Levin, M.A. and H. Strauss. 1993. Overview of Risk Assessment. 35-49. In: Levin M.A. and Strauss H. (eds.). Risk assessment in genetic engineering 1-17. New York: McGraw Hill.

Office of Technology and Science Policy. 1985. Coordinated Framework for the regulation of biotechnology; establishment of the biotechnology science coordinating committee. FR 50: 47174-47195. 14 Nov.

Office of Technology Assessment. 1987. Public perception of Biotechnology. Washington, DC.

Presidential/Congressional Commission on Risk Assessment. 1997. Washington, DC.

Sharples, F.E. 1991. Ecological Aspects of Hazard Identification. 99-122. In: Levin, M.A. and Strauss, H. (eds.). Risk Assessment in Genetic Engineering. New York: McGraw Hill.

Tiedge, J.M., R.K. Colwell, Y.L. Grossman, R.E. Hodson, R.E. Lenski, R.N. Mack and P.J. Regal. 1989. The Planned Introduction of Genetically Engineered Organisms: Ecological Considerations and Recommendations. Ecology 70(2):298-315.

USEPA. 1990. Points To Consider in the Preparation and Submission of TSCA Premanufacture notices for Microorganisms. Program Dev. Branch, Washington, DC: USEPA.

Witt, S.C. 1990. Biotechnology, Microbes and the environment. Center for Science Inf. San Francisco, CA.