FRANK MIRER, PhD, CIH, is a professor in the CUNY School of Public Health in New York. He can be reached at (212) 396-7782 or
Decades back, at a truck company’s labor/management health and safety conference, management’s toxicologist presented a talk that minimized (to the point of erasure) the potential of diesel engine emissions to cause lung cancer. From the back of the room, I told my union colleagues that if they bought management’s arguments for weakening emissions regulations and possibly sell more trucks, they would undermine our arguments for better ventilation to control exposure and symptoms at the drive-off end of the assembly line.
Lung cancer is not the only potential consequence of diesel exposure, and the quantitative assessments discussed in this article do not address other significant respiratory effects. But there is new information. First, there have been advances in risk assessment for diesel engine exhaust (DE), also known as diesel particulate matter (DPM). Second, the recently published data on the supposed lack of carcinogenicity of new-technology diesel exhaust has been overhyped by the Health Effects Institute. RISK ESTIMATES IN PEOPLE In 2014, a sextet of epidemiologists, including three members of an International Agency for Research on Cancer (IARC) working group and the author of a key study by the National Cancer Institute (NCI), published an exposure-response assessment for DPM in Environmental Health Perspectives. They estimated the unit risk for lung cancer for DPM in the occupational setting at 1 µg/m3 of elemental carbon as 1.7 per 1,000, therefore a “significant risk” at prevailing exposure levels. Further, they estimated that approximately 6 percent of all lung cancer deaths in the U.S. were due to DE. This scientific publication has not been addressed by any regulatory agency.
This new potency estimate followed the most recent determination of DPM’s carcinogenic (toxic) potential, IARC’s Monograph 105, published in 2013, which concluded that DPM is “known” to cause lung cancer and probably bladder cancer. The IARC review included the 2011 NCI-NIOSH Diesel Exhaust in Miners Study (DEMS), whose publication was delayed for years by management lawsuits and congressional intervention. There is still an ongoing effort at spit-balling DEMS (by you know who) and the dose-response assessment for DPM. My view is that the thirty studies previous to DEMS should have carried the day for “known” even if DEMS were null, which it wasn’t.
The IARC monograph noted that urban air pollution and airborne exposure to particulate were also “known” to be carcinogenic to humans, with DPM as a significant contributor. Also notable was the determination that gasoline engine emissions are “probably” carcinogenic to humans; this determination relied on laboratory studies since studies in people are not feasible because of overlapping exposures to diesel. IARC included no information about compressed natural gas (CNG) emissions because no one has attempted to gather health data.
EVOLUTION OF KNOWLEDGE OF DIESEL HAZARDS The carcinogenic potential of DPM was signaled by mutagenicity bioassays as far back as 1973. A finding of genotoxicity was thought to be persuasive for low-dose extrapolation. These findings spurred EPA concerns, and then a chronic bioassay. A chronic inhalation bioassay in rats was reported in 1983, showing dose response and 3,500 µg/m3 as an effect level for lung carcinogenesis—around fifty-fold more potent than tobacco smoke in similar systems. No authoritative quantitative risk assessment was published based on the laboratory data. The diesel bioassay was followed by a bioassay of carbon black, which also caused lung cancer with similar potency. Carbon black can be considered DPM without the polycyclic aromatic hydrocarbons previously thought to be the whole problem. Perhaps all carbon particles, including nanoparticles, should be feared equally? While the laboratory studies are sufficient evidence that carbon black is carcinogenic, it was classified as “possibly” carcinogenic because there were no studies of people considered adequate.
In 1998, a team of NIOSH scientists published a scientific paper with a quantitative risk estimate for miners, based on epidemiological studies. NIOSH did not adopt this estimate as official policy, but it was incorporated into the MSHA DPM standard for underground mining, first promulgated in 2001 and amended several times.
In 2003, EPA published an updated assessment of DPM that summarized the data in people: A persistent association of risk for lung cancer associated with DE exposure has been observed in more than 30 epidemiologic studies published in the literature over the past 40 years. The majority of the epidemiologic studies evaluate distinct populations of occupational groups…
Increased lung cancer risk has been observed in 8 out of 10 cohort studies, 5 of which were statistically significant. Increased lung cancer risk has also been observed in 10 of 12 case-control studies, 8 of which were statistically significant.
Overall, the increased lung cancer relative risks generally range from 1.2 to 1.5, although a few studies show overall relative risks as high as 2.6. Statistically significant increases in relative risk (RR), 1.33 to 1.47, are also shown in two independent meta-analyses of epidemiologic studies. EPA declined to calculate a slope factor for DPM. A 50 percent increase in lung cancer risk translates to a risk level for a fatal disease of approximately 3 per 1,000. Note that the quoted risks are cohort wide and do not implicate risk levels at particular exposure levels.
Regulation of diesel emissions proceeded at EPA and in the European Union, resulting in substantial reductions on emissions from new engines. Translation of a grams-per-mile PM limit to human exposure is beyond my technical knowledge; the vehicle emission limits appeared to be driven by feasibility without an accessible relationship to health data. HEALTH EFFECTS INSTITUTE In January 2015, the Health Effects Institute website trumpeted, “The HEI report on lifetime animal exposure to new-technology diesel engine exhaust finds no evidence of lung cancer.” I don’t challenge the “no evidence” conclusion, but the study was designed to have virtually no chance of finding evidence of anything about carcinogenic potential of new-technology DPM, despite elaborate and costly procedures. I’ve reviewed dozens of National Toxicology Program bioassays as a member of NTP’s Board of Scientific Counselors; I would characterize this study as “inadequate” because of dose selection. The highest exposure to DPM was about 5 µg/m3. In the 1983 study that supported sufficient evidence of carcinogenicity, 3,500 µg/m3 was an effect level. In addition, a different strain of rat was used than in previous studies, possibly a strain more resistant to particle carcinogenesis. To get any DPM effect, the new-technology particles would have to be hundreds of times more potent than the old technology particles, or the rat strain much less resistant.
DPM particle compositions vary between duty cycles on the same engine, and between engine type. The most prominent difference is the ratio of elemental to organic carbon, but particle size distribution would likely be important for health effects. It would be useful to compare potency of the particles generated by old and new technology, but this study can’t do it. Exposure levels were chosen based on nitrogen oxide (NOx) exposure levels, with the highest level less than the OSHA PEL. To be useful in hazard identification, a bioassay with a null result must include the maximum tolerated dose; it doesn’t appear the MTD was achieved for NOx and obviously not for DPM. The highest dose tested was a 50-fold dilution of the tailpipe emissions.
To me, this massive study has produced no information about the toxic potential of new diesel. A useful study would project how much particle exposure results at these emission rates, especially in the occupational environment. Meanwhile, getting the old diesels off the road, and also off the off-road (construction equipment, railroads, generators, mines), is a high priority. CNG EMISSIONS The hazard of emissions from CNG engines—a solution of choice for those who fear diesel—have not been evaluated sufficiently. While the mass of particles emitted by CNG is less than new-technology diesel, the number of particles is similar. CNG emissions are more in the nanoparticle range. Until there’s a well conducted chronic bioassay including the maximum tolerated dose for CNG particles, we are taking a leap into the dark. Money should be added to the NTP budget to conduct such a study. Otherwise, the transition will be from the devil we can measure to a devil we can’t.
The Little Engine That Could
Risk Assessments for Diesel Engine Exhaust BY FRANK MIRER