Cancer risk: tumors excluded.

Abstract

In their Report “Variation in cancer risk among tissues can be explained by the number of stem cell divisions” (2 January, p. [78][1]), C. Tomasetti and B. Vogelstein discuss an interesting correlation (0.804) between estimated lifetime stem cell division number in 31 tissue types and corresponding cancer incidence rates in the United States. However, their assertion that only 35% of cancer risk variation is due to environmental or genetic factors is problematic. The correlation analysis excluded many cancers (such as stomach, breast, prostate, cervix, kidney, endometrium, bladder, and lymphoma) that are common in the United States or worldwide, so no statement about overall cancer rate variation that is “explained” by stem cell divisions can be made. Furthermore, the correlation was anchored by five data points for osteosarcoma and included tumor subtypes having genetic (colorectal) and environmental influences (lung), but stem cell division rates were not estimated separately for organ subtypes. There are strong time trends in cancer incidence rates and large incidence-rate variations internationally for nearly all cancer types [for example, the rate of squamous esophagus cancer among men with the highest incidence (Jiashan County in China and African Americans in South Carolina) is more than 100 times the rate among men with the lowest incidence (Algeria) ([ 1 ][2])]. If international rates were added to Figure 1, a much smaller fraction of incidence rate variation would be explained by stem cell divisions. Moreover, as the authors note, “The total number of stem cells in an organ and their proliferation rate may of course be influenced by genetic and environmental factors,” so that stem cell division numbers could serve, substantially, as a mediator of genetic and environmental influences, rather than a distinct etiologic factor. Finally, high values of the authors' extra risk score (ERS) are described as arising when “there is high cancer risk relative to the number of stem cell divisions,” but ERS is calculated not as the ratio, but as the product, of cancer incidence rates and stem cell division number. Hence the resulting classification into D and R tumors does not seem interpretable and, regardless, could aim only to identify tumors that have etiologic mechanisms other than stem cell division number. 1. [↵][3] , Cancer Incidence in Five Continents, Volume IX (International Agency for Research on Cancer Scientific Publications, No. 160, Lyon, France, 2007). [1]: /lookup/doi/10.1126/science.1260825 [2]: #ref-1 [3]: #xref-ref-1-1 View reference 1 in text