Following the completion of the U.S. Environmental Protection Agency (EPA) National Center for Environmental Assessment (NCEA) state-of-the-science workshop on formaldehyde inhalation last month, the American Chemistry Council’s Formaldehyde Panel invited the co-chairs of three workshop sessions to offer their thoughts on some of the key discussions that took place. We’ll be posting responses by each of the three experts in the coming days.
You can check out our first conversation with Dr. Lorenz Rhomberg here.
After hearing from Dr. Rhomberg, we spoke with Dr. James Collins, the current director of epidemiology at the Dow Chemical Company and an adjunct research professor at the University of Pittsburgh, School of Public Health, and at Saginaw Valley State University.
At the EPA workshop, Dr. Collins hosted a session entitled, “Review of the epidemiologic evidence of LHP cancers.”
We asked him a series of follow-up questions below:
ACC: What do you think were the three key issues discussed in your session of the workshop?
Dr. Collins: The three issues discussed in our session were:
1. What is the best way to examine the weight of the epidemiology studies to determine if formaldehyde is related to LHP risk?
2. How should lymphohematopoietic (LHP) cancers be best grouped for analysis in epidemiology studies?
3. How good are the exposure metrics for examining cumulative, peak, and average intensity in three major cohort epidemiology studies (NCI, NIOSH, British Industry Wide)?
ACC: Are there any uncertainties that you believe EPA still needs to address to reach a decision on formaldehyde’s association with leukemia?
Dr. Collins: Yes. To name a few:
1. Is there an excess in myeloid leukemia or any other of the LHPs in the three major cohort studies overall? This was not a subject of the conference although data was presented that indicated that there was not an excess across studies.
2. If there is an excess of myeloid leukemia or any other of the LHPs, what is the proper exposure metric for assessing potential association with formaldehyde? Cumulative exposure does not appear to be associated with increases of these cancers. I know of no other human carcinogen where cumulative exposure is not related to the increasing cancer risk. The unconventional measure of peak exposure employed in the NCI study must be explained and justified. The conventional measure of peak exposure employed in the NCI study showed no association with increased cancer risk. Further, even the NCI exposure assessor put no faith in the assessment of peak exposures.
3. What is to be made of the Zhang et al. results given that the mechanistic findings indicate that formaldehyde does not get to the bone marrow?
ACC: What steps are needed to integrate the information from the various sessions into a weight of evidence for the association between formaldehyde and leukemia?
Dr. Collins: I can offer a few suggestions:
1. EPA must adopt a conventional approach for summarizing the information for first determining if there is an excess in myeloid leukemia or any other of the LHPs. There was some discussion of “weighting the studies” by quality. This is difficult to do and is not the preferred approach in epidemiology. A more conventional approach would be a meta-analysis stratified by study type (cohort, case-control). Another approach would be to focus on the three major studies as was done in the Checkoway et al. review presented earlier in the conference.
2. The lack of leukemia findings compared to the U.S. population in the NCI study and the NIOSH studies is being attributed to the “healthy worker effect.” Thus some argue that only internal comparisons are valid. While there are limitations to the external comparisons, there are also limitations to internal comparisons. Typically, if there is no excess of a cancer in the study overall, it is difficult to argue that there is an effect of exposure using an internal analysis.
3. The Zhang et al. study should be redone eliminating the problems with that study. Without this being redone, speculation will still occur that some mechanism associated with formaldehyde could be causing bone marrow damage.
The EPA has indicated that it plans to hold further public forums to address additional issues that can inform the revised risk assessment. The ACC Formaldehyde Panel fully supports EPA’s efforts to increase transparency and continue the discussion of the science.