Causation of Cancer: Bad Genes, Bad Lifestyle—or Simply Bad Luck?

February 2015

“Why me?” is a direct or implied question from many patients diagnosed with cancer, but there is rarely a simple answer. At some point in any discussion with healthcare professionals or during the patient’s own researches, the words “gene” and “environment” are likely to crop up. Both can cause confusion. All cancer mechanisms are related to changes in a cell’s genetic make-up, but malignant disease is not usually inherited. “Environment” is not much better. Does this refer to carcinogens, such as those to which someone might be exposed at work or naturally (radon), or to more personal factors where there is more choice (smoking, diet)?

Small wonder then that attempts to quantify the relative contributions of genes and environment are so troublesome. Estimates sometimes appear to have been achieved by crude subtraction; take away cancers known to be associated with inherited genes, and that leaves 90-95% due to environmental factors. A paper on cancer genetics in the Jan.2 issue of Science has ensured that media coverage of matters medical got off to a controversial start in 2015.

“We have concluded that during normal life, random heritable changes may occur by accident in particular stem cells and that when certain changes all happen in a single stem cell the result may be proliferation into recognizable cancer.” This was a conclusion drawn, not from the Science paper, but 40 years earlier, from experiments in mice exposed to the skin carcinogen benzpyrene (R. Peto , et al., Brit. J. Cancer, 32[4]: 411-26, 1975).

The role of the stem cell has a distinguished record in the annals of research into the mechanisms of cancer, and a significant further contribution comes in the recent Science article from Cristian Tomasetti and Bert Vogelstein, who report a correlation between stem cell division rates and cancer incidence for 31 tissue types (see the accompanying table, paper #1). That is by no means every tissue prone to malignant change, of course; importantly, neither prostate nor non-familial breast cancer were included because stem-cell data were insufficient. The more frequently a stem cell divides the more likely it is that that tissue will be associated with a greater malignancy rate.

Published collaborations between Bert Vogelstein and Cristian Tomasetti*

1 C. Tomasetti, B. Vogelstein, “Variation in cancer risk among tissues can be explained by the number of stem cell divisions,” Science,347(6217): 78-81, 2015.
2 C. Tomasetti, et al., “Only three driver gene mutations are required for the development of lung and colorectal cancers,” Proc. Nat.l Acad. Sci. USA,112(1): 118-23, 2015.
3 C. Tomasetti, B. Vogelstein, G. Parmigiani, “Half or more of the somatic mutations in cancers of self-renewing tissues originate prior to tumor initiation,” Proc. Nat.l Acad. Sci. USA, 110(6): 1999-2004, 2013.
*and others
Source: Thomson Reuters  Web of Science

The correlation is 0.81 (in medicine, that is strong) and implies that around two-thirds of cancer-causing mutations are random events or “bad luck,” a recurring phrase in media coverage of this publication. Has that coverage—prompting what Science termed a “backlash”—overstated or badly sensationalized this research?  Andrew Maynard from the University of Michigan’s Risk Science Center, for one, having looked at the paper and the Johns Hopkins University press release as well as the resulting press coverage, thinks not.

Tomasetti is quoted in The Economist (Jan. 10, 2015) as saying that, “We have not shown that two-thirds of cancer cases are about bad luck”; what he and Vogelstein wrote was that a majority two-thirds of the variation in cancer risk among tissues was “due to ‘bad luck.’” Tomasetti does go on to say that “Cancer is a combination of bad luck, environment and genes,” an opinion with which few would quarrel.

The Science paper next classifies the 31 tumor types into two groups, R(eplicative) and D(eterministic), an exercise that appears to strike a blow at prospects for primary prevention of some tumors. There were 9 type D and 22 type R, the division being achieved by a modelling exercise based on a tissue’s ERS (extra risk score). The paper’s final sentence is unambiguous: “For R-tumors primary prevention measures are not likely to be very effective, and secondary prevention should be the major focus.”

Vogelstein will be a familiar name to ScienceWatch. He is one of the world’s most-cited scientists, with more than 500 papers captured by Web of Science since 1981 and, over the last decade, average citation rates of 151 for all fields and 267 for molecular biology and genetics (Web of Science Essential Science Indicators). Tomasetti brings mathematical talent to bear on biomedical problems. Clinicians may find his more formula-laden articles challenging (e.g, , “On the probability of random genetic mutations for various types of tumor growth,” Bull. Math. Biol., 74[6]: 1379-95, 2012, and “Stochastic modelling of multiple random genetic mutations under the cancer stem cell hypothesis,” Math. Pop. Studies, 19[4]: 200-13, 2012).

These two scientists—one highly cited, the other much less (so far)—have appeared as co-authors three times to date (see table). If paper #1, as Tomasetti and Vogelstein observe, “could have important public health implications” (i.e, for preventing cancer), #3 has relevance to the conduct of genome-wide studies of disease. As the authors note, there should be more focus on younger patients to reduce “the ‘noise’ from passenger mutations occurring in normal tissues as individuals age.”


Mr. David W. Sharp, M.A. (Cambridge), formerly deputy editor of The Lancet, is a freelance writer in Minchinhampton, UK.

The data and citation records included in this report are from Thomson Reuters Web of ScienceTM. Web of ScienceTM is a registered trademark of Thomson Reuters. All rights reserved.