The Good, the Bad, and the Ugly

A quick take-away from a landmark study of benzene exposure and cancer among petroleum workers: Defending trace benzene lawsuits will be more challenging.

• By Jeffrey R. Allen, Rachel L. Wright, Paul J. Johnson
• Sep 01, 2013

In the ever-expanding arena of so-called "trace benzene" litigation, there is no single scientific issue more hotly contested between plaintiffs and defendants than the question of what level of benzene exposure must be established in order to conclude that such exposure was a cause of a plaintiff's disease. Even with respect to myelodysplastic syndrome ("MDS") and acute myeloid leukemia ("AML"), two blood disorders whose association with at least high levels of cumulative lifetime benzene exposure has been generally accepted in the scientific community for years, plaintiff experts and defense experts typically have widely diverging opinions as to just how much benzene a person need be exposed in the course of his or her life before a causal link can be established. Indeed, the evidentiary crux of any trace benzene lawsuit centers upon the plaintiff's ability to present credible expert opinion testimony as to the extent and timing of the benzene exposure at issue, the source of the exposure (i.e., which defendant's products are the supposed culprits), and the association in the relevant scientific literature between the alleged exposure and the disease in question. The goal of defense counsel in these cases, of course, is to controvert such evidence with experts of their own. Now, into this already heated debate come the results of a benzene-related study years in the making, published in the Journal of the National Cancer Institute this past November. The landmark study, though a mixed bag for both sides of the bar, will undoubtedly have a significant impact on trace benzene suits and how we litigate them.

The work, "Myelodysplastic Syndrome and Benzene Exposure Among Petroleum Workers: An International Pooled Analysis,"¹ combines empirical data from three influential nested case control studies of petroleum workers previously conducted by the article's principal authors in Canada,² the United Kingdom,3 and Australia,⁴ respectively. The purpose of the pooled analysis was not only to reconcile epidemiological inconsistencies found within the three prior studies, but to introduce additional data gathered since the studies' publication dates, and to provide expanded epidemiological data regarding the relationship between benzene exposure and various diseases, some of which were unstudied or understudied in the prior works. The authors of the pooled analysis⁵ are widely considered to be some of the foremost researchers and experts in their fields, and are respected by plaintiff and defense experts alike. Given this, the report, though inconclusive in some respects, will inevitably become a part of the trace benzene dialogue, and as such, merits in-depth review and study by toxic tort practitioners. In this article we highlight some of the report's findings from the defense perspective -- the good, the bad, and the ugly.

The Good
For those involved in defending trace benzene suits, one of the most positive developments to come from the pooled analysis is that it failed to show any statistically significant relationship between low levels of benzene exposure and the development of AML.⁶ Indeed, the pooled analysis -- in a departure from previous research -- shows no statistically significant association between AML and benzene exposure at any level. The data also failed to demonstrate any evidence of a statistically significant relationship between benzene exposure and the development of chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), or myeloproliferative disease (MPD).

Another encouraging finding from the pooled analysis pertains to peak exposures and AML, CLL, CML, and MPD. Peak exposures have become an increasingly popular area of analysis and point of emphasis for plaintiffs' exposure assessment experts, since peak assessments create "bolus" exposure numbers significantly higher (and thus scarier to a jury) than a cumulative dose assessment examining exposures over years and decades. However, the pooled study failed to demonstrate any statistically significant effects from peak exposures regarding these diseases.⁷

In addition to these cumulative and peak exposure findings, the pooled analysis sheds further light on the question of whether the skin provides a meaningful route of benzene exposure. The inclusion of dermal exposure in the benzene exposure assessments of plaintiffs' experts has grown increasingly common in the past few years, with such experts developing ever more complex (though scientifically dubious) models to quantify exposure through skin contact. However, the analysis of the pooled study’s authors offers little to no support for the theory that dermal exposure to benzene increases the risk of developing disease. Despite examining dermal benzene exposures among workers diagnosed with AML, MDS, CLL, CML and MPD, the authors found no statistically significant increased risk of developing those diseases at any level of benzene exposure through the skin.⁸ Such findings are not necessarily surprising, but they are certainly significant, as relatively few studies have devoted specific analysis to this particular pathway of exposure.

The Bad
Undoubtedly, the most troubling findings from the pooled analysis are the authors' conclusions regarding low-level benzene exposure and the risk of MDS. According to the study, workers with cumulative benzene exposures in excess of 2.93 ppm years showed a statistically significant increased risk of developing MDS.⁹ In fact, the data demonstrated nearly a four-fold increased risk of developing MDS at such exposure levels.¹⁰ The threshold number of 2.93 ppm years is significantly lower than that found in previous studies of MDS workers¹¹ and ANLL/MDS combined groups¹² and dramatically lower than the 40-200 ppm years often cited by experts as being the threshold at which there is a statistically significant increased risk for developing AML¹³ (given that a certain percentage of AML cases have a preceding MDS diagnosis, the elevated risk of MDS at the relatively low benzene exposures shown in the pooled analysis has troubling implications for AML, as well).

Separate and apart from a lower cumulative benzene exposure threshold with respect to MDS, the pooled analysis also found a significant correlation between relatively low short-term peak exposures to benzene and a significantly increased risk of MDS.¹⁴ The authors report a five- to six-fold increased risk of MDS in workers who likely had weekly exposures of greater than 3 ppm benzene for as little as 15 minutes at a time.¹⁵ The peak exposures and the associated dangers identified by the study's authors could possibly call into question the efficacy of the current short-term exposure limit (STEL) for benzene promulgated by OSHA, which requires that workers be protected from exposures in excess of 5 ppm as averaged over any 15-minute period.¹⁶ If so, any defendant’s reliance on such a standard at trial would be precarious at best.

The Ugly
In addition to the obviously good and truly bad, the pooled analysis offers certain details that are less clear but point toward negative trends for defendants in trace benzene litigation.

One such finding pertains to the latency argument that exposures occurring 15 years or more prior to diagnosis are unlikely to play a role in disease causation. The authors note that, for MDS, AML, and CML, they found no strengthening of the association between benzene exposure and disease when they limited their analysis to only those exposures occurring within 2 to 15 years before diagnosis --which may suggest that exposures beyond 15 years prior to diagnosis are no less relevant to disease causation.¹⁷ If so, such results would argue for a much longer relevant exposure period for a given plaintiff than otherwise, thereby potentially expanding the number of exposure events (and, theoretically, the products and defendants) at issue, as well as possibly increasing the cumulative lifetime dose that could reasonably be said to have causal relevance in the case. That said, the findings in the pooled analysis suggesting a wider "etiological window" for benzene exposure would seem to fly in the face of prior well-regarded studies that found that remote exposures to benzene had little to no causal relationship with MDS and AML.¹⁸

Another potentially ugly finding relates to benzene exposure and CML. The current epidemiological literature shows little to no support for the proposition that benzene exposure increases the risk of developing CML,¹⁹ and, as noted above, this study's conclusions as to CML are consistent with that existing body of scientific research. There is, however, a curious finding: While the study reports no statistically significant increased risk of developing CML among those workers with a cumulative benzene exposure of less than .348 ppm years, nor among those workers with a cumulative benzene exposure of greater than 2.93 ppm years, there is a set of results that shows workers with a cumulative benzene exposure in the middle range of .348-2.93 ppm years not only have a statistically significant increased risk of developing CML, but in fact have a risk factor greater than three times that of unexposed workers.²⁰ Exactly what the plaintiffs' bar and their experts will make of such a finding remains to be seen, but it is a common tactic among plaintiffs' experts to cherry-pick such isolated (and often anomalous) results from epidemiological literature to bolster their opinion testimony in litigation. The defense bar should expect nothing different here.

The Future
Given the findings of the pooled analysis, what changes, if any, can trace benzene practitioners and their clients expect to see in the legal landscape?

• Expect more lawsuits. Despite limited and inconsistent findings, the study’s epidemiological data on MDS will likely lead to an expanded pool of potential plaintiffs, as plaintiffs' attorneys become emboldened by the results of the new report. Though plaintiffs with MDS may be rare,²¹ it seems certain that the plaintiffs' bar will attempt to shoehorn the more troubling findings of the pooled analysis into more lawsuits involving unrelated blood disorders.
• Expect more defendants. The pooled analysis, by suggesting a lower threshold at which there is a statistically significant increased risk for developing a disease due to benzene exposure, has potentially expanded the universe of products that could have conceivably contributed to a given plaintiff's illness. And if the study's findings relating to latency hold up under closer examination, the longer relevant exposure period is likely to put more defendants in the crosshairs. It will come as no surprise, then, if the ranks of toxic tort litigation welcome a new freshman class of defendants -- ones with products containing even lower trace benzene content than products previously at issue.
• Expect new products. With an expanded world of products, those defending trace benzene exposure lawsuits may see not only an increase in the volume of suits and enlargement of the roster of defendants, but also for their own clients an expanded line of products being identified and targeted in litigation. This means defense practitioners will need to look beyond the chemical content, instructions, and material safety data sheet language of their clients' usual "hot products" and take a more exhaustive inventory of the entire product line.
• Expect a lower threshold. It is a given that the plaintiffs' bar and their experts will argue for a lower exposure threshold for MDS as a result of the pooled analysis. However, the study will undoubtedly also be used in support of a lower exposure threshold for all hematopoietic diseases. As noted above, this will most certainly be the case for AML. The science of MDS and AML makes increasingly clear that MDS is comprised of different morphological subtypes, and that of those, only certain types of MDS are likely to progress to AML.²² And, according to some researchers, the subtype of MDS most closely associated with benzene exposure is not among the subtypes of MDS most likely to develop into AML.²³ However, the fact remains that anywhere from one-tenth to one-third of MDS cases develop into AML,²⁴ and as such, any study which lowers the exposure threshold for MDS will be used by plaintiffs to bolster an argument to lower the exposure threshold for AML. Similar "biological plausibility" theories will almost certainly be espoused for other hematopoietic diseases as plaintiffs' experts use the results of the pooled analysis to form a muddied picture of low-dose benzene exposure, MDS, and all other diseases of the blood and blood-forming organs.
• Expect Daubert challenges to be more challenging. Just as low-level exposure lawsuits may gain enhanced viability from the results of the pooled analysis, plaintiffs' experts may gain enhanced credibility for their conclusions and a larger gray area within which to maneuver. As a result, excluding plaintiffs' experts prior to trial may become an increasingly difficult task to accomplish, thereby prolonging and increasing the expense of trace benzene lawsuits.

Final Thoughts
Whether the results of the 2012 pooled analysis constitute a permanent game-changer or a temporary fluke is unknowable at this point. But for the immediate future, defendants should expect that the findings --positive, negative, and perplexing -- will change the way trace benzene cases are pursued, prepared and defended, likely at greater cost and risk than ever before.

References
1. Schnatter AR, Glass DC, Tang G, et al. Myelodysplastic Syndrome and Benzene Exposure Among Petroleum Workers: An International Pooled Analysis. J Natl Cancer Inst. 2012 Nov 21; 104(22): 1724-1737.
2. Schnatter AR, Armstrong TW, Nicolich MJ, et al. Lymphohaematopoietic malignancies and quantitative estimates of exposure to benzene in Canadian petroleum distribution workers. Occup Environ Med. 1996; 53(11): 773-781.
3. Rushton L, Romaniuk H. A case-control study to investigate the risk of leukaemia associated with exposure to benzene in petroleum marketing and distribution workers in the United Kingdom. Occup Environ Med. 1997; 54(3): 152-166.
4. Glass DC, Gray CN, Jolley DJ, et al. Leukemia risk associated with low-level benzene exposure.  Epidemiology. 2003; 14(5): 569-577.
5. The authors of the pooled analysis include A. Robert Schnatter, Occupational and Public Health Division, ExxonMobil Biomedical Sciences, Inc.; Deborah C. Glass, Department of Epidemiology and Preventative Medicine, Monash Centre for Occupational and Environmental Health, Monash University, Melbourne, Australia; Gong Tang, Department of Biostatistics, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania; Richard D. Irons, Health Sciences Center, University of Colorado, Aurora, Colorado; and Lesley Rushton, Department of Public Health and Biostatistics, Faculty of Medicine, Imperial College of London, London, United Kingdom.
6. See, e.g., Schnatter 2012 at 1728.
7. Id. at 1730.
8. Id. at 1731.
9. Id. at 1727-28.
10. Id. (OR = 4.33 [95% CI=1.31 to 14.3]).
11. See Irons RD, Gross RA, Le A, et al. Integrating WHO 2001-2008 criteria for the diagnosis of myelodysplastic syndrome (MDS): a case-case analysis of benzene exposure. Chem Biol Interact. 2010; 184 (1-2): 30-38 and Lv L, Lin G, Gao X, et al. Case-control study of risk factors of myelodysplastic syndromes according to World Health Organization classification in a Chinese population. Am J Hematol. 2011; 86(2):163-169 (examining levels of benzene exposure in excess of 20 ppm).
12. Hayes RB, Yin SN, Dosemeci M, et al. Benzene and the dose-related incidence of hematologic neoplasms in China. J. Natl Cancer Inst. 1997; 89(14): 1065-1071 (finding a statistically significant increased risk of developing MDS/ANLL at exposures in excess of 40 ppm).
See, e.g., Constantini AS, Quinn N, Consonni D, et al. Exposure to benzene and risk of leukemia among shoe factory workers. Scand J Work Environ Health. 2003; 29: 51-59; Hayes 1997; Rinsky RA, Smith AB, Hornung R, et al. Benzene and leukemia – an epidemiological risk assessment.  New England J of Med. 1987; 316: 1044-1049.
14. Schnatter 2012 at 1734 (Table 3).
15. Schnatter 2012 at 1726-27, and 1734 (Table 3)
16. See 29 C.F.R. 1910.1028(c)(2)
17. Schnatter 2012 at 1727
18. See, e.g., Hayes 1997; Wong, O, Harris F, et al. A hospital-based case-control study of acute myeloid leukemia in Shanghai: Analysis of environmental and occupational risk factors by subtypes of the WHO classification. Chemico-Biological Interactions. 2010; 184: 112-128.
19. See, e.g. Lamm SH, Engel A, Joshi KP, et al. Chronic myelogenous leukemia and benzene exposure: A systematic review and meta-analysis of the case-control literature. Chemico-Biological Interactions. 2009; 182: 93-97; but see, e.g., Huebner WW, Wojcik NC, Rosamilia K, Jorgensen G, Milano CA. Mortality updates (1970-1997) of two refinery/petrochemical plant cohorts at Baton Rouge, Louisiana, and Baytown, Texas. J Occup Environ Med. 2004; 46(12): 1229-1245.
20. Schnatter 2012 at 1727-28 (OR = 5.04 [95% CI=1.45 to 17.5]).
21. The incidence of myelodysplastic syndrome (MDS) in the United States is estimated at 4.5 cases for every 100,000 people. This averages out to approximately 12,000 new cases of MDS each year, though this number seems to be increasing as the average age of the population has increased. American Cancer Society. What are the key statistics about myelodysplastic syndrome?, available at http://www.cancer.org/cancer/myelodysplasticsyndrome/detailedguide/myelodysplastic-syndromes-key-statistics (last visited Feb. 6, 2013). It is worth noting that the incidence of AML in the United States is only slightly greater, with approximately 14,590 new cases of acute myeloid leukemia estimated to be diagnosed in 2013. American Cancer Society. What are the key statistics about acute myeloid leukemia?, available at http://www.cancer.org/cancer/leukemia-acutemyeloidaml/detailedguide/leukemia-acute-myeloid-myelogenous-key-statistics (last visited Feb. 6, 2013).
22. See Irons 2010.
23. See id.
24. Schnatter 2012 at 1735; Shukron, O, Vainstein, V, et al. Analyzing transformation of myelodysplastic syndrome to secondary acute myeloid leukemia using a large patient database. Am J Hematol 2012; 87(9): 853-860 (citing that one-third of MDS cases progress to AML).

This article originally appeared in the September 2013 issue of Occupational Health & Safety.