Cancer risk: Accuracy of literature

See allHide authors and affiliations

Science  13 Feb 2015:
Vol. 347, Issue 6223, pp. 729
DOI: 10.1126/science.aaa6212

We read with interest the recent cancer etiology Report “Variation in cancer risk among tissues can be explained by the number of stem cell divisions” (2 January, p. 78), in which C. Tomasetti and B. Vogelstein claim that most cancer risk can be explained by chance mutations. However, the selection criteria used for cancer types included in the study are not robust. First, the authors report using an “extensive literature search” to identify eligible tissue types. There is no evidence that a systematic literature review was conducted. Second, the assessment of literature quality and subsequent inclusion criteria is not clear. According to the authors, “Other cancer types could not be assessed, largely because details about the normal stem cells maintaining the tissue in homeostasis have not yet been agreed upon or accurately quantified.” There have been volumes written about the necessity of systematic literature reviews and subsequent appraisal as a critical component of obtaining accurate and unbiased research results (1).

The method used by Tomasetti and Vogelstein leads to the exclusion of breast and prostate cancer, together accounting for ∼25% of all newly diagnosed cancers (2). No doubt other cancer types are excluded as well. Breast and prostate cancer have been closely studied, in many cases to a much greater extent than those cancers that the authors select. Lack of agreement regarding accurate quantification of these cell types should be addressed by sensitivity analysis rather than exclusion. Large bodies of literature will invariably contain disagreement between authors. This is hardly justification for exclusion.


Navigate This Article