Latest Chernobyl Cancer Study Contains Numerous Problems

Dec 14, 2012 | Fukushima Accident Updates

A published research paper entitled “Radiation and the Risk of Chronic Lymphocytic and other Leukemias among Chernobyl Cleanup Workers”(1) (here-in, The Report) has stirred international interest due to its declaring there is a significantly elevated incidence of leukemia with Chernobyl cleanup workers, especially Chronic Lymphocytic Leukemia, due to low level radiation (LLR) exposure. Cyber-colleague Rod Adams has detailed The Report’s unconventional conclusion quite well in his November 9 posting(2), but made no judgment as to its veracity. Recently, the Ukrainian Radiation Protection Society revealed problems which taints The Report’s authenticity, including; (a) errors in the estimated values of relative risk, (b) incorrect “fitting” of dose distribution, and (c) large errors in hidden dose calculations based on a questionable methodology.(3) However, this writer has found additional problems in The Report that also bear public exposure.

To provide background: adult leukemia has been extensively studied and documented for many decades. The most common form is called Chronic Lymphocytic Leukemia (CLL), which constitutes about 45% of all adult leukemia cases and is the best-understood of them all. (NIOSH, CDC, NCI, ACS) While the incidence of leukemia in general was elevated with Hiroshima/Nagasaki bomb survivors, there was no increase in CLL. As a result, all scientific organizations rejected the notion of CLL being caused by LLR until the mid-1990s when a few speculative reports were published saying CLL/LLR link might be possible. NIOSH held a public debate on the conjectural papers in 1994, the panelists of which agreed that a CLL/LLR connection was speculative and inconclusive. Another wave of hypothetical reports on the possibility of a CLL/LLR connection began to re-emerge in 2004, resulting in NIOSH/CDC studies which once again found there was nothing conclusive to suggest an actual relationship.

Returning to the topic-at-hand, The Report says “…risks associated with protracted exposures, and associations between radiation and chronic lymphocytic leukemia are not clear.” The Report’s researchers refer to several papers published between 2005 and 2009 as being supportive of this claim. It turns out that only one of the cited references is in any way supportive. None of the rest can be correctly identified as correlative. In particular, The Report references elevated leukemia with LLR exposures to nuclear industry workers. I investigated their posted references of support on nuclear worker data [Cardis (2007), Muirhead (2009), Linet (2009), Jacob (2009), Krestinia (2010), and Richardson (2009)], and found the only one remotely corroborative is Richardson. Richardson contains but a partial sentence which says “…there is not a persuasive basis for the conclusion that CLL is a non-radiogenic form of cancer,”, but it does not say there is evidence for a CLL/LLR connection, either. On the other hand, Muirhead said there was no statistical basis for a CLL/LLR relationship and two others don’t mention CLL at all (Linet; Jacob). Cardis includes CLL in the types cancer studied, but only lung cancer, multiple myeloma and “ill-defined and secondary cancers” were found to have a statistically supportive relationship with LLR…not CLL. All referred papers did say that more study on LLR and leukemia-in-general is needed, but that’s a ubiquitous phrase common to nearly all modern science papers regardless of topic. To make matters worse, The Report also references a 2010 UNSCEAR report on low level radiation exposure and leukemia as being supportive of their “unclear” CLL/LLR claim. UNSCEAR’s report does just the opposite. UNSCEAR 2010 concludes that high-exposure Chernobyl data does not differ from Hiroshima/Nagasaki data which showed no CLL/LLR connection. In addition, UNSCEAR says the low doses experienced by the vast majority of Chernobyl clean-up workers were equivalent to those naturally experienced in China and India which show no increased risk of leukemia. To put it bluntly, UNSCEAR 2010 makes no statement that even remotely supports The Report’s claim of a connection between CLL and LLR. Thus, I must conclude that The Report’s listing of supportive references with respect to its claim of a CLL/LLR relationship being “not clear” is materially incorrect.

Another problem concerns The Report making no mention of a comparison between the frequencies of leukemia in male Chernobyl workers versus its frequency within the non-irradiated male public at large. NCI and NIOSH say the natural occurrence is about 1,500 total leukemia cases per 100,000 individuals over their lifetime (~75 years). The frequency in men is double that of women, so 1,000 leukemias can be reasonably expected among 100,000 men over a typical lifetime. This equates to an incidence in men of about 13.5 leukemias per year in a population of 100,000. With the 20 years of data used in The Report, covering over 110,000 male Chernobyl workers (the “cohort”), there should have been at least 300 total cases of leukemia…but there were only 162 among the Chernobyl worker cohort! Further, the frequency of CLL for men is established at roughly 6.5 per 100,000 per year. Out of 110,000 Chernobyl clean-up workers, roughly 140 should have contracted CLL over the 20 years following the accident in the Ukraine. However, The Report documents only 79 CLLs. Total numbers of any disease in a large body of cases will necessarily fluctuate some 15-20% over time. However, the extreme degree of fluctuation here strongly suggests that there is no statistical possibility which could show a leukemia/LLR or CLL/LLR relationship exists with respect to the Chernobyl workers. Why does The Report fail to make a comparison between typical non-irradiated leukemia statistics and those gleaned from Chernobyl worker records at all? It appears The Report looks at Chernobyl in isolation from all necessarily-related statistics: a critical omission that can only have been intentional.

Two key lessons should be learned here. First, references of support for any paper making a sensational claim that flies in the face of the historical record should be closely examined for correctness before a Journal publishes it. Second, drawing any conclusion on cancer incidence without comparing to the world’s characteristic, cohort-specific rate of cancer incidence is presumptuous, at best.

Abbreviations:

  1. NIOSH – National Institute of Occupational Health and safety
  2. CDC – Center for Disease Control
  3. NCI – National Cancer Institute
  4. ACS – American Cancer Society
  5. UNSCEAR – United Nations Scientific Committee on the Effects of Atomic Radiation
  6. RADRUE – Realistic Analytical Dose Reconstruction with Uncertainty Estimation
  7. ERR/Gy – Excess Relative Risk per Gray (sievert equivalent)

References:

  1. “Radiation and the Risk of Chronic Lymphocytic and other Leukemias among Chernobyl Cleanup Workers”; Environmental Health Perspectives; November 8, 2012. http://dx.doi.org/10.1289/ehp.1204996
  2. Adams, Rod; “Out of 110,645 Chernobyl clean-up workers, 19 might have contracted radiation related leukemia”; Atomic Insights; November 9, 2012. http://atomicinsights.com/2012/11/even-in-science-there-is-room-for-an-english-major-who-can-parse-words.html
  3. “Do leukemia effects at low doses look convincing?”; Ukrainian Radiation Protection Society; December 9, 2012. http://urps-notices.blogspot.com/2012/12/do-look-leukemia-effects-at-low-doses-convincing.html?goback=%2Egde_117546_member_195635885
  4. Cardis, et el; “The 15-Country Collaborative Study of Cancer Risk among Radiation Workers in the Nuclear Industry: Estimates of Radiation-Related Cancer Risks”; Official Journal of the Radiation Research Society; April, 2007. http://www.rrjournal.org/doi/abs/10.1667/RR0553.1
  5. Muirhead, et al; “Mortality and cancer incidence following occupational radiation exposure: third analysis of the National Registry for Radiation Workers”; British Journal of Cancer; January 13, 2009. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2634664/
  6. Linet, et al; “Children’s Exposure to Diagnostic Medical Radiation and Cancer Risk: Epidemiologic and Dosimetric Considerations”; US National Library of Medicine; December 16, 2010. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2814780/
  7. Richardson, et al; “Ionizing Radiation and Chronic Lymphocytic Leukemia”; Environmental Health Perspectives; January, 2005. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1253701/
  8. Jacob, et al; “Is cancer risk of radiation workers larger than expected?”; US National Library of Medicine: December, 2009. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2776242/
  9. “SEER Fact Sheets: Chronic Lymphocytic Leukemia”; National Cancer Institute; http://seer.cancer.gov/statfacts/html/clyl.html
  10. “Leukemia – Chronic Lymphocytic”; American Cancer Society. http://www.cancer.org/cancer/leukemia-chroniclymphocyticcll/detailedguide/index
  11. “RADRUE Method for Reconstruction of External Photon Doses to Chernobyl Liquidators in Epidemiological Studies”; US National Library of Medicine; October, 2010. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2930607/