Re: "The Next Generation of Large-Scale Epidemiologic Research: Implications for Training Cancer Epidemiologists"

I read with concern the recent commentary by Spitz et al. (1) in which they advocate training current and future cancer epi-demiologists in the complexities of the "-omics " (i.e., geno-mics, proteomics, metabolomics, and epigenomics), with the corollary of advanced data management and analysis training to handle the reams of " big data " generated by these technologies. Essentially, in their view, cancer epidemiologists should be trained in computer science and molecular biology so that they will contribute meaningfully to moving epidemiology away from being considered a " soft " science. Implicit in their commentary is the foundational belief that molecular approaches to investigating the causes of cancer offer the most promise for preventing cancer incidence and mortality. Arguably, the current paradigm for cancer research rests within the somatic mutation theory, which characterizes cancer as a cell-based, genetic, molecular disease (2). The promise of the relentless, well-funded search of genetic evidence was that it would first provide a cogent explanation of the disease and then the means to prevent cancer incidence and mortality. Instead, judging by candid admissions made by leading cancer researchers reflecting on the so-called war on cancer (3, 4), those promises remain mostly unfulfilled. If the goal of epidemiology is to study the causes and distributions of diseases in human populations so as to identify ways to prevent and control disease (5), then cancer epidemiology has had several notable successes, including the decrease in the number of cases and deaths from smoking-related cancers (6) and the precipitous drop in breast cancer incidence that was likely an unintended consequence of women fleeing from their hormone treatments (7). The body of research driven by the somatic mutation theory is largely unrelated to these few measurable successes; no big " wins " in our field have been scored from decades of focusing on genetic mutations. Unfortunately, many cancer rates are increasing for unknown reasons (8), and for most cancers, there are still no robust prevention strategies. Treatments based on somatic mutation theory have, so far, been fleeting and temporary (3). It is perhaps time for cancer epidemiologists to explore alternative theories of cancer causation. The tissue organization field theory (9), which posits that cancer is a tissue-based rather than a cell-based disease, has great potential for reframing the way we think about cancer causation and the means to treat and even prevent cancers. Sonnenschein and Soto (10) compellingly refute the …