Jay H. Lubin, Ph.D.
|Organization:||National Cancer InstituteDivision of Cancer Epidemiology & Genetics, Biostatistics Branch|
|Address:||Executive Plaza SouthRoom 8105|
Dr. Lubin received a Ph.D. in biostatistics from the University of Washington. He joined the NCI as a Staff Fellow in 1978. Dr. Lubin received the PHS Special Recognition Award in 1993, and was elected a Fellow of the American Statistical Association in 1994. He has served on numerous expert committees of the National Academy of Sciences and the National Council for Radiation Protection and Measurements. In 1999, Dr. Lubin was selected DCEG Mentor of the Year by the training fellows in the Division, and in 2005 was given the DCEG Exemplary Service Award.
Lung cancer is the leading cause of cancer mortality in the United States. After cigarette smoking, the inhalation of radioactive radon and its decay products in homes may be the most important risk factor for lung cancer. Epidemiologic studies of the carcinogenic effects of radon in mines and in homes are underway to characterize the exposure-response relationship and to evaluate other relevant factors. These studies are being linked to cellular and molecular approaches to aid in estimating the effects of low-level exposure to indoor radon. Other areas of research include studies to evaluate cancer risk associated with the use of commercial and non-commercial pesticides, exposure to chlorination byproducts and arsenic in drinking water and occupational exposures to arsenic, as well as an assessment of the long-term health effects of augmentation mammaplasty. The focus of methodologic areas of research is on design limitations that impede indoor radon studies, including exposure uncertainties, low total exposures with small expected relative risks and means for effectively evaluating gene-environment interactions. In addition, methodologic research is addressing the impact of uncertainties in radiation dosimetry in assessing the effects of head and neck irradiation on risk for cancers of the thyroid and modeling cancer risk and environmental exposures.
Previous studies of radon-exposed underground miners showed that exposure to radon decay products causes lung cancer. However, these studies were limited in size and were unable to assess risks for children or for long durations of exposure. Our study in Chinese tin miners, nearly one-third of whom were first exposed under age 13, revealed that radon was a lung carcinogen, but that children did not incur any extra risk as adults due to early-life exposures. We also found an inverse exposure-rate effect that showed for equal total exposure, long duration of exposure at a low exposure rate was more harmful than a short duration exposure at a high exposure rate. Collaborating with principal investigators of all the studies of radon-exposed underground miners, including 11 cohort studies, we pooled original data on 65,000 miners and over 2,700 lung cancer deaths. The exposure-response relationship was consistently linear in cumulative exposure, suggesting that radon progeny exposure at lower levels, such as in homes, would carry some risk. The effects of radon exposure diminished with time since last exposure, but there was no consistent relationship between risk and age at first exposure. The exposure-response trend for never-smokers was three-fold greater than for smokers. Among the miners, radon progeny exposure was responsible for an estimated 40% of all lung cancer deaths, 70% of lung cancer deaths in never-smokers, and 40% of lung cancer deaths in smokers. Biophysical models predict that at low total exposure, as occurs in homes, the inverse exposure-rate effect should not occur, since at low total exposure a target cell is unlikely to experience more than one radiation dose, and thus cannot "know" a reduction in exposure rate. Analysis of the pooled miner data demonstrated this diminution of the inverse exposure-rate effect. Aside from lung cancer, a detailed analysis of the underground miners found no other anatomic site at an increased risk of cancer from exposure to inhaled radon.
There are major uncertainties in estimates of radon effects in the general population due to differences between working in mines and living in houses. To reduce these uncertainties, case-control studies of lung cancer in women were undertaken in Shenyang, China, and in Missouri. Results were consistent with extrapolations from miners, but showed no clear-cut effect from radon. A third case-control study is ongoing in Gansu Province, China, where 40% of the houses have indoor radon concentrations above levels at which the U.S. Environmental Protection Agency (EPA) recommends remedial action. We also conducted a meta-analysis of eight published studies, which included a total of 4,263 lung cancer cases, and estimated a relative risk at the EPA action level of 1.14 with 95% confidence interval (1.0,1.3). The combined trend in the relative risk was significantly different from zero (p=0.03). This estimate was similar to extrapolations from miner studies, and to the relative risk computed directly from miners with low exposures comparable to those in homes. We concluded that the true risk from exposure to indoor radon for the general population is unlikely to be markedly greater than predicted from miners, and that the indoor radon studies provide general validity for using a miner-based model for risk estimation.
In other research, a detailed analysis was undertaken of a cohort 8,014 arsenic-exposed, copper smelter workers, with over 50 years of follow-up. For 428 deaths from respiratory cancer, there was a linear increase in the excess relative risk of respiratory cancer with increasing arsenic exposure. Since inhaled airborne arsenic increases systemic levels of arsenic, as measured by arsenic in urine, there is concern that inhaled arsenic may also increase risk of cancers of the skin, bladder, kidney, and liver, which have been linked to drinking arsenic-contaminated water. Among the smelter workers, there were no consistent excess risks observed for these sites, or for circulatory diseases or diabetes mellitus.
Methodologic studies have shown that errors in exposure assessment, missing exposure history data, and the mobility of the population can dramatically influence the power of indoor radon studies to detect an excess risk. This finding suggests that studies must be much larger than previously thought, or that data from them should be pooled to increase sample size. Also, our research showed that the type of model assumed for gene-environment interaction markedly influences the size of the study needed for evaluating this effect, while error in exposure assessment (or error in genotyping) reduces the power to detect interactions.