Characterizing Exposure of Veterans to Agent Orange and Other Herbicides Used in Vietnam: Scientific Considerations Regarding a Request for Proposals for Research (1997)

individual soldiers; and distinguishing the types of herbicides used. The section entitled ''Development of the Exposure Reconstruction Model'' in the Appendix (p. 46) discusses this in greater detail.

Biomarkers of exposure require the ability to measure a chemical in a biologic specimen—that is, urine, blood, tissue, organs, hair, and the like. If biomarkers of exposure are chosen as a dose index, such as serum TCDD, then the quantitative relations between the marker and herbicide exposure and its limitations must be discussed. For example, individual clearance rates of TCDD are variable, so low serum TCDD may not equate to low past exposure. There are also other sources of TCDD intake, such as diet, so current high TCDD levels may not necessarily indicate high past exposure. Finally, TCDD was a highly variable constituent of the herbicides used in Vietnam, and some herbicides did not contain TCDD; therefore, TCDD may not be a good marker for all types of herbicide exposure. Additional information on biomarkers is provided in the Appendix.

Validation of the exposure assessment approach developed will be a critical part of the proposal. The committee recognizes that retrospective exposure assessment of herbicide exposures among Vietnam War veterans may, at best, be semiquantitative or rank ordered. Because we have a generally poor understanding of the sources and magnitudes of exposures to herbicides that occurred in Vietnam more than 25 years ago, the predicted levels of exposure to individual veterans are likely to be imprecise. Thus, applicants should discuss the anticipated sources and magnitudes of uncertainty in their model predictions. Attention should also be paid to the role that quantitative or semiquantitative data might play in the validation of exposure predictions in a pilot study.

The best method for validation is not obvious and is left to the researchers to propose, critique, and defend. The limitations and uncertainties associated with the validation method—such as those discussed previously for serum TCDD levels—should be addressed.

Respondents are encouraged to consider other alternative approaches to the validation process.

The purpose of the research to be sponsored under the RFP is to develop exposure assessment models for use in epidemiologic studies of Vietnam veterans. Proposals of models for which the relevance to Vietnam veterans is unclear will not be further evaluated. The proposals will be assessed based on

five criteria: (1) context, (2) technical merit, (3) feasibility and validation, (4) personnel and experience, and (5) cost and timeline. A discussion of these criteria and the weighting they will be assigned. follows.

Proposals should be specific about how the intended assessment approach could be used in epidemiologic studies and should briefly describe the types of studies envisioned. Such descriptions should provide enough detail to allow an evaluation of their potential for successful completion. A strong proposal will describe feasible epidemiologic studies that have a high likelihood of detecting whether or not particular health effects (e.g., cancer) are related to veterans' exposure to herbicides in Vietnam.

The proposal should describe the following:

• How the proposed method may improve upon other available methods for Vietnam veterans and provide a better basis for exposure assessment.

• The degree to which the overall herbicide exposure in Vietnam of individuals or groups can be described by the proposed exposure assessment approach.

• The likelihood of discriminating levels of herbicide exposure among the individuals or groups and of identifying high-exposure subgroups or individuals.

• How critical sources of uncertainty will be identified and characterized.

• The characteristics of the groups or individuals studied that need to be controlled for in the analysis and, for the validation study, important confounding exposures and other factors.

• Explicit justification for any proposed biomarker (e.g., TCDD) studies.

• The weaknesses of the chosen approach and the means that will be used to minimize their impact, including a discussion of the critical sources of uncertainty and how they may affect the validation study being proposed.

The proposal should describe in detail the study designed to demonstrate the feasibility and validity of implementing the exposure assessment approach, as well as the accuracy, biases, and uncertainties of the approach. The proposal should also describe how this approach could be applied to achieve meaningful