Status Report on the High Production Volume (HPV) Challenge Program, October 12, 2001SUMMARY
The Chemical Right-to-Know Initiative was launched in April, 1998, with the goal of obtaining screening level data for the approximately 2,800 chemicals manufactured or imported into the United States in quantities exceeding one million pounds per year. Manufacturers and importers were invited to participate in a voluntary "challenge" program to provide basic toxicity data on the high production volume chemicals they produce. A framework for the High Production Volume (HPV) Challenge Program was announced on October 9, 1998, by EPA, Environmental Defense, the American Chemistry Council and the American Petroleum Institute. The HPV Challenge Program is based on the Organization for Economic Cooperation and Developments Screening Information Data Set (OECD/SIDS), an internationally agreed upon set of tests for screening high production volume chemicals for human and environmental hazards. Companies were asked to indicate their commitment to supply the SIDS-level data on their chemicals by December, 1999. To date, 469 companies and 187 consortia have agreed to sponsor 2,155 chemicals. The balance is being addressed in the OECD HPV SIDS Program or as candidates for a rule under Section 4 of the Toxic Substances Control Act (TSCA). As part of the HPV Challenge Program commitment, sponsors submit to EPA a Test Plan and supporting robust summaries of existing data. The submitted information undergoes a 120-day public comment period to allow others to review and comment on the plan and to submit additional data that may be used to obviate the need for any proposed testing. 55 Test Plans covering 383 chemicals have been submitted by sponsors. An examination of a majority of the submitted Test Plans reveals that, sponsors are doing a good job following the various guidance documents EPA has prepared for the HPV Challenge Program; a significant amount of unpublished data is being provided to support the Test Plans; and the amount of new testing being proposed to address the SIDS endpoints is less than initially envisioned.
THE HPV CHALLENGE PROGRAM
A 1997 study by the Environmental Defense Fund (now Environmental Defense) showed that, for a sample of 100 HPV chemicals, there were relatively few US HPV chemicals which met minimum data requirements for health hazard screening. EPA undertook a similar study of approximately 2,800 chemicals manufactured or imported in the U.S. in quantities exceeding one million pounds per year. EPAs report found a similar dearth of publicly available data. These findings were subsequently supported by a study conducted by the Chemical Manufacturers Association (now the American Chemistry Council). As a result of these findings, EPA launched the HPV Challenge Program as part of the Chemical Right-to-Know Initiative on the eve of EarthDay, April 21, 1998. The goal was to encourage the manufacturers and importers of HPV chemicals to make publicly available existing data, or conduct testing if necessary, to address the screening level endpoints established by the OECD HPV SIDS Program. These include physicochemical properties, environmental fate data, acute ecological effects (chronic or terrestrial studies if appropriate), and a number of human health endpoints. A framework was established for the overall program through the collaboration of EPA, Environmental Defense, the American Chemistry Council and the American Petroleum Institute. The framework was announced on October 9, 1998.
Companies were asked to participate in the HPV Challenge Program by committing to supply the SIDS-level data by no later than 2005. As part of the commitment, companies were asked to specify a start year (1999 through 2003) in which they would supply a Test Plan and supporting robust summaries of existing data for public comment. Test Plans for categories of chemicals were requested to be submitted in one of the first two start years (not including 1999) so that in the event that the category hypotheses proposed in the Test Plan were not borne out, there would be sufficient time to conduct additional testing by the end of 2004, in order that a complete data set could be generated and all data made publicly available in robust summary form by 2005.
Concurrently with the US HPV Challenge Program, the International Council of Chemical Associations (ICCA) HPV Chemicals Initiative was created with the goal of handling HPV chemicals of international interest within the OECD HPV SIDS Program. Nearly 700 chemicals have been committed to through the ICCA HPV initiative. EPA has agreed to act as the sponsor country for over 200 ICCA chemicals to be evaluated in the OECD HPV SIDS Program. Confirmed ICCA commitments are recognized in the HPV Challenge Program, since the same set of screening data is made publicly available through the two programs.
To date, 2155 chemicals have been sponsored by 469 companies and 187 consortia. Non-sponsored chemicals will be considered for rulemaking under Section 4 of TSCA. Currently, a proposed Section 4 rule covers 37 chemicals. The 120-day comment period for the proposed rule closed on April 25, 2001, and a final rule is expected by spring 2002.
IMPLEMENTING THE PROGRAM
In order to assist sponsors in meeting their commitments to the HPV Challenge Program, the Agency prepared a number of guidance documents that were designed to maximize the use of existing information and various estimation techniques so that the level of any new testing proposed under the program would be minimized. This would reduce costs substantially and address animal welfare concerns by reducing the number of animals used in the HPV Challenge Program. The principal guidance documents are: Searching for Chemical Information and Data; The Use of Structure-Activity Relationships (SAR); Development of Chemical Categories; Developing Robust Summaries; Assessing Adequacy of Existing Data; and Fact Sheet on Animal Welfare. Other guidance documents include What to Test and Testing Closed SystemIntermediates. All guidance documents can be found at http://www.epa.gov/chemrtk/pubs/general/guidocs.htm. In addition, in a letter to program participants dated October 14, 1999, the Agency outlined 10 principles that sponsors were asked to observe as they meet their commitments. These 10 principles identified animal welfare considerations in the HPV Challenge Program and established a 120-day public comment period between posting and implementation of Test Plans. Follow-up letters were sent to participants in October, 2000, reiterating EPAs commitment to the principles outlined in the October, 14, 1999, letter.
Public review of submitted Test Plans is largely done electronically through the use of the HPV Challenge Program Website. Sponsors were advised to send Test Plan submissions to the Post Office Box established for the HPV Challenge Program or electronically via the Challenge Website. Submissions can also be sent directly by e-mail to email@example.com or firstname.lastname@example.org. After receipt by EPA, submissions are assigned an Administrative Record Number and placed in the TSCA Administrative Record, File Number AR201, established for the HPV Challenge Program. The submissions are then prepared for Website posting by converting all submissions to a uniform posting format. The first review step by EPA is to determine if a complete package is available for review. This step entails checking the Test Plan to determine whether and how each endpoint has been addressed. If the sponsor indicates that adequate data are available to address a particular endpoint, then a robust summary of those data must be provided. Any omissions are noted and the sponsor is directed to revise the submission. Once submissions are determined to be complete, they are prepared for posting on the HPV Challenge Website. Posting of the submission starts the 120-day public comment period. Concurrently, EPA technical staff review the submissions to develop EPAs comments. EPA and all other public comments are sent to the sponsors electronically for their consideration. All comments on Test Plans are also posted on the Challenge Website.
STATUS OF THE PROGRAM
To date, EPA has received 55 Test Plans covering 383 chemicals. The Test Plans cover 28 categories of chemicals and 27 single chemicals. Over 90 percent of the chemicals are addressed in category Test Plans. Fifty-one Test Plans have been posted on the HPV Challenge Program Website for public comment (http://www.epa.gov/chemrtk/pubs/summaries/viewsrch.htm). The balance is undergoing initial EPA review or on hold pending sponsor revisions. The 120-day public comment period has closed for 29 Test Plans. Public and EPA comments can be viewed under each Test Plan entry on the Website.
To gauge how well the HPV Challenge Program is being implemented and to determine the degree to which sponsors are using the HPV Challenge Program guidance, a number of analyses were conducted looking at data/study sources (that is, whether published or unpublished data were being submitted to support the data needs identified in the program), how the SIDS endpoints were being addressed, and whether sponsors were following the principles outlined in the October 14, 1999, letter to program participants.
As discussed earlier, the prime impetus for the HPV Challenge Program was the consistent finding in the series of reports that publicly available hazard information was lacking on a majority of the 2,800 US HPV chemicals. Following EPAs guidance, sponsors are to identify existing data on HPV chemicals and submit these data in the form of robust summaries. To determine the source of the existing data (that is, published or unpublished data), each Test Plan posted on the Website as of October 4, 2001, was examined (46 Test Plans). The Table below presents the results for the ecological and human health endpoints, and selected physicochemical and environmental fate endpoints (some endpoints are routinely determined with estimation techniques). The number of published and unpublished studies documented in the robust summaries prepared for each SIDS endpoint is listed in the columns below.
|ENDPOINT AREA||SPECIFIC ENDPOINT||PUBLISHED||UNPUBLISHED||TOTAL|
|Physicochemical Properties||Water Solubility||39||40||79|
As can be seen from the numbers of studies listed in the Table, a significant amount of unpublished data have been made public by the sponsors. This indicates that the sponsors, either individually or through the many consortia participating in the program, have made a concerted effort to bring forth existing data. This has had a significant effect on the amount of new testing that sponsors have proposed (see below). This is also in keeping with the Agencys experience implementing the testing provision of Section 4 of TSCA where industry has made every effort to bring forth existing data in response to proposed testing actions.
Directly related to the above discussion is an analysis of how sponsors have proposed to address the SIDS endpoints. The following analysis looked at the first 46 Test Plans posted on the HPV Challenge Website to determine how the health and environmental effects endpoints were addressed in the Test Plan. Three methods to meet the minimum data requirements for each SIDS endpoint were proposed. Data needs were met by (1) the use of existing scientifically adequate data, (2) the use of an estimation technique (SAR, "read-across" in categories) or by providing a rationale for no testing, or (3) proposing new testing. For human health effects, five endpoints (acute, repeat dose, reproductive, developmental, and genetic toxicity) were considered. The analysis was done with and without the API Petroleum Gases Test Plan. It was decided to also generate results with and without this Test Plan, because it covers a very large category of 161 chemicals, nearly all of which had no data submitted, it was critical that these data did not skew the results of the analysis. For environmental effects, three acute toxicity endpoints were examined (acute toxicity to fish, daphnia, and algae). Again, the analysis was done with and without the API Test Plan. The 46 Test Plans (including the API Test Plan) address 325 chemicals in 23 categories and 23 single chemicals. Each calculation below is based on the total number of endpoints that could be addressed for each chemical multiplied by the total number of chemicals. This formula was employed since it was impossible to identify all scientifically adequate existing studies prior to the start of the HPV Challenge Program.
|1. All Chemicals considering 5 endpoints
325 X 5 = 1625 Total Endpoints
|2. All chemicals considering 5 endpoints minus the API Test Plan (161 chemicals)
164 X 5 = 820 Total Endpoints
|Estimation/No testing||348||R:9 C:3|
|1. All chemicals considering 3 endpoints
325 X 3 = 975 Total Endpoints
|2. All chemicals considering 3 endpoints minus the API Test Plan (161 chemicals)
164 X 3 = 492
As can be seen by the above analysis, sponsors have made maximum use of the guidance concerning the use of SAR and category proposals, and in combination with the significant amount of unpublished data made available through the robust summaries, only a minimal amount of testing has been proposed. Overall, for health and environmental effects, approximately six percent of the endpoints are proposed to be addressed with new testing. Even after removing the API Test Plan from the calculations, the overall amount of proposed testing is less than 10 percent. The exact mix and number of new tests may change as sponsors consider EPA and public comments (comment periods on 17 Test Plans have not closed); however, there is no reason to believe that the overall conclusions highlighted through analyses of these 46 Test Plans will change significantly. Comments could lead to some additional tests being performed, but could also result in fewer tests. For instance, the American Petroleum Institute originally proposed to conduct five acute inhalation tests in the Petroleum Gases Test Plan. However, the revised Test Plan, taking into consideration EPA and public comments, removed those five tests from the category Test Plan.
Conformance with October 14, 1999 Letter to Program Participants
On October 14, 1999, EPA provided additional guidance to manufacturers and importers participating in the HPV Challenge Program. The letter outlined 10 principles that sponsors were asked to observe in Test Plans submitted under the HPV Challenge Program. In some cases, sponsors may choose to deviate from these principles because of testing needs not associated with the HPV Challenge Program if a rationale is provided. In essence, the principles address the development of a well-considered Test Plan for HPV chemicals that accounts for all scientifically sound information, and uses categories of chemicals and SAR where appropriate. By developing the Test Plans in this way, unnecessary testing can be avoided while the need for the development of scientifically defensible health and environmental effects information on HPV chemicals will be fulfilled.
To see how well sponsors were following the 10 principles, the first 40 Test Plans posted on the HPV Challenge Program Website were examined (see Appendix). It should be noted that some of the principles include subjective judgements. For instance, the first principle requests that sponsors conduct a "thoughtful, qualitative analysis rather than use a rote checklist approach." Since sponsors are not required to justify testing under the HPV Challenge Program (the OECD HPV SIDS Program also does not support such a requirement, since it has already been established that the SIDS test battery represents the minimal set of data needs for screening HPV chemicals) the absence of text discussing the testing decisions from a number of the single chemical Test Plans does not necessarily indicate that the sponsors did not do a "thoughtful analysis." Also, it is difficult, considering only the Test Plan rationales, to judge the thought, time, and effort that went into developing the Test Plans for single chemicals.
In examining the 40 Test Plans it was determined that, overall, sponsors were following the guidance and were making extensive use of categories and SAR/estimation techniques which reduced the need for additional testing. While EPA may not have agreed with all aspects of some of the category proposals, it was clear that sponsors have devoted considerable time and thought in constructing the category Test Plans. Some of the single chemical Test Plans had considerable discussion of the proposals, while others merely provided a matrix of available data and proposed testing. Again, the HPV Challenge Program does not require justification for proposed testing.
However, a number of instances were noted where it appeared that sponsors did not follow the guidance. One sponsor proposed terrestrial toxicity testing (earthworm and plants) because land application was a known use of the material. This is entirely consistent with OECD SIDS guidance in this area. Dermal toxicity testing was proposed in five test plans. In three test plans the sponsor was proposing testing because of non-U.S. testing requirements, one sponsor decided not to do the dermal test in response to comments, and the other is under review (the proposed dermal irritation study is outside the scope of the HPV Challenge Program). Similarly, in-vivo genetic toxicity testing was proposed in a few Test Plans. It was subsequently determined that in three Test Plans, the testing was proposed to meet non-U.S. testing needs, and in another the sponsor - citing positive in-vitro genetic toxicity tests and the need for responsibleproduct stewardship - has decided to conduct the study. Testing of single chemicals in advance of the request to delay such testing until November, 2001, was noted in a few Test Plans. For three Test Plans (the same three discussed above), this was being done for non-U.S. testing requirements (the sponsor subsequently agreed to delay the testing) and in two Test Plans it was reported in public comments that testing had already been conducted.
It appears that when all the HPV Challenge guidance is taken into consideration, sponsors are making good faith efforts to follow that guidance. The conclusion is that following the guidance has resulted in the need to conduct fewer tests than originally envisioned. Sponsors have provided a considerable amount of unpublished data to support their Test Plans, which in combination with the extensive use of categories, this has resulted in about six percent of the health and ecological effects endpoints being addressed with new testing. There have been some deviations from the principles articulated in the October 14, 1999 letter to program participants; however, in many instances this was because the sponsor needed to address testing needs beyond the scope of the HPV Challenge Program.
LESSONS LEARNED/NEXT STEPS
To date, the overall experience in implementing the HPV Challenge Program is positive. A few recommendations to improve the processing and review of Test Plans are noted below. Occasionally, Test Plans are sent directly to EPA officials involved in the HPV Challenge Program. To speed the processing for Website posting and entry into the Administrative Record, all submissions should be sent to the e-mail addresses cited above or by following the directions on the HPV Challenge Website for electronic submissions. If submitting paper copies to the HPV Challenge Post Office Box, a disk with electronic files should be included with the submission. This will permit the preparation of more accurate electronic files for posting. Paper copies alone are scanned, which is time consuming and increases the potential for error.
In the Test Plans themselves, it would be helpful for submitters to explain the rationale for any testing that is proposed beyond the SIDS endpoints. A number of times sponsors have included in their Test Plans, tests they need to perform to meet non-U.S. data requirements. Oftentimes, the goal is to avoid duplicative testing; however, the lack of an explanation leads to needless speculation about why the tests were proposed and the generation of comments requesting explanations for the proposed tests. In category Test Plans, greater attention should be placed on how the category hypothesis addresses the main endpoints (human health, environmental effects, environmental fate, and physicochemical properties). A number of times it appeared that the category was based mainly on human health endpoints and the other endpoint areas were not as thoroughly discussed or supported. EPA recommends that basic physicochemical properties for discrete substances be measured instead of estimated, since these values are used in other estimation models (e.g., transport and distribution). Using estimated values multiplies uncertainty in the results from the subsequent modeling. Finally, it isrecommended that sponsors include the results of literature searches conducted in addition to the robust summaries, i.e., lists of those studies reviewed but not considered key studies (see the end of Section 5 in the Assessing Adequacy of Existing Data guidance document).
EPA is receiving Test Plans and robust summaries in a number of formats. An MS Access Data Input Tool and database were created to assist sponsors in submitting their robust summaries to EPA. Over 250 CD-ROMs have been distributed to sponsors. However, only a few sets of robust summaries have been submitted so far in this format. In any event, all robust summaries received, regardless of the format in which the data are being submitted, are input to the Access database so that a searchable database will be available. Although all the Test Plans and robust summaries submitted to date (and which have successfully completed the initial review) are now available on the HPV Challenge Website, there is no search capability. The MS Access database will serve as an interim tool until an Oracle-based system is available in late 2002.
Although this report represents only a "snapshot" in time, it appears that the HPV Challenge Program is being implemented in a cost-efficient, judicious manner. Sponsors are providing a large amount of unpublished data and are following the various guidance documents to a significant degree. However, more attention should be devoted in the category Test Plans to documenting how all the major endpoints (health, ecological, fate, and physicochemical properties) are addressed by the category rationale. Sponsors are using category approaches, SAR, and other estimation techniques to reduce costs and the need for new testing. The net result is that new testing is being proposed for about six percent of the health and ecological endpoints, due in large part to the amount of test data, particularly unpublished data, brought forward. The analysis of the degree to which sponsors were following the principles listed in the October 14, 1999, letter to program participants did not reveal any significant trends, except that sponsors were not adequately explaining why they were proposing tests that were beyond the base set of screening tests and/or were not consistent with EPAs guidance.