Publications : 2024

Vincent M, Fitch S, Bylsma L, Thompson C, Rogers S, Britt J, Wikoff D. Integration of toxicological and epidemiological information to evaluate biological plausibility and causality of associations between inhaled formaldehyde (FA) and lymphohematopoietic (LHP) cancers. Abstract 5157, Society of Toxicology Annual Meeting, Salt Lake City, UT, March 2024. 

Abstract

Background and Purpose: To date, available reviews of FA exposure and LHP cancers have not both (a) fully applied systematic review methodology (including critical appraisal via risk of bias (RoB)) and (b) considered the integration of epidemiological information with toxicological and mode of action (MOA) information and biological plausibility. Therefore, uncertainty remains regarding inconsistently observed associations in some studies, and whether there is support for a casual relationship. As such, a systematic review was conducted to investigate if there is a causal relationship between inhaled FA and LHP (including Hodgkin lymphoma, multiple myeloma, myeloid leukemia, monocytic leukemia, or lymphoid leukemia) cancers in humans. Methods: The methods and reporting used in this systematic review align with international systematic review guidance, including development of an a priori protocol and PECO question, systematic literature searching and screening using inclusion/exclusion criteria, critical appraisal of RoB, and structured synthesis and integration methods. Critical appraisal was conducted using NTP OHAT’s RoB Tool, refined to the outcome and exposures under evaluation. Exposure, outcome, and confounding domains were used as key domains. Evidence was synthesized qualitatively as well as quantitatively where applicable; meta-analyses were conducted to estimate pooled SMRs from the epidemiological studies. Confidence in the body of evidence was assessed with OHAT and GRADE, and causality via Bradford Hill with particular emphasis on biological plausibility (including consideration of toxicokinetics, MOA, endogenous capacities, and genotoxicity). Results: Four experimental rodent bioassays and 16 observational studies in humans were included. Experimental animal evidence consistently reported a lack of treatment-related LHP-related lesions. Increased LHP mortality was inconsistently observed in some occupational cohorts exposed to FA. Meta-analyses did not identify significant associations between FA inhalation (when measured as ever/never exposure) and LHP outcomes, with meta-SMRs ranging from 0.50 to 1.51, depending on LHP subtype. RoB analysis identified systemic limitations precluding confidence in the epidemiological evidence due to inadequate characterization of FA exposure and a failure to adequately adjust for confounders or effect modifiers, thus suggesting that effect estimates are likely to be impacted by systemic bias. Specifically, none of the identified epidemiological studies quantitatively measured exposure to FA during the study period, and attempts to quantify exposure relied upon historical recreations and understanding of job processes. Most studies did not account for smoking, a critical confounder, or occupational co-exposures. The confidence in the body of epidemiological literature was rated “Very Low”, primarily due to the high risk of bias in the observational evidence and inconsistency in the findings of association and exposure-response both within and across studies. Conversely, the confidence in the body of experimental animal literature was rated “High” due to the consistency of a lack of association and exposure-response gradient in toxicological evidence. No biologically plausible explanation linking inhalation of FA and LHPs was identified, supported primarily by the lack of systemic distribution and lack of in vivo genotoxicity. Conclusions: This systematic review places emphasis on the critical appraisal of epidemiological data and integrated consideration of experimental animal and mechanistic evidence streams to inform the potential for a causal relationship between FA exposure and LHP cancers. Integration of the evidence streams reduces confidence in the identification of a causal relationship, particularly when the lack of biological plausibility is considered along with the high likelihood that inconsistent findings in observational studies can be explained by residual bias. Therefore, the inconsistent associations between inhaled FA and LHP mortality reported in the observational epidemiological evidence were not considered causal when integrated with the totality of the epidemiological evidence, toxicological data, and lack of biological plausibility. Based on the findings of this review, LHP cancers do not appear to be suitable for use as a critical effect in FA risk assessments.