Nature, Nurture, and Bad Luck: Revisiting the Link Between Diet, Lifestyle, and Cancer

Predicting, with any degree of certainty, who will develop cancer is impossible today. But, as the second-leading cause of mortality in the United States, cancer is a major public health concern, as well as a potentially terrifying diagnosis for those who receive it. In 2016 alone, it is estimated that more than 1.6 million people in the United States will be diagnosed with a malignancy, and despite the many advances in cancer care in the last 2 decades, more than 35% of those will ultimately die from that diagnosis, if current trends hold true. As a result, policy makers, health care providers, and concerned consumers search for strategies to prevent cancer from occurring.

Ever since the US surgeon general’s first report on smoking in 1964, evidence has progressively accumulated favoring lifestyle and environmental factors as significant contributors to the development of cancer. The “healthy lifestyle” recommendations long made by naturopathic physicians, chiropractors, and enlightened medical doctors (and routinely dismissed by the conventional medical establishment as “food faddism,” “unscientific wishful thinking,” and “hucksterism”) very slowly, but steadily, began to be supported by rigorous scientific evidence.

In the last year, we have seen an interesting, at times passionate, and even a tad vitriolic scientific debate about the etiology of cancer. For at least 35 years, information from sources such as the American Cancer Society, the American Institute for Cancer Research, and the National Cancer Institute has consistently reported that somewhere between 30% and 65% of cancer in the United States was due to environmental, dietary, and lifestyle issues, with some estimates ranging as high as 90%.

This view was challenged in January 2015, in an article by Tomasetti and Vogelstein1 published in Science, which stated that a majority of the variation in cancer risk seen between different tissues “is due to ‘bad luck,’ that is, random mutations arising during DNA replication in normal, noncancerous stem cells.” This conclusion was picked up by news outlets across the world, with inflammatory titles such as “Most Cancer Caused by Bad Luck, Not Genes or Lifestyle,”2 “Two-thirds of Cancers Are Due to ‘Bad Luck,’”3 and “Cancer Is More Bad Luck Than Bad Behavior, Study Says.”4 As is too often the case, the popular press sensationalized these findings, which were a secondary analysis of previously reported data.

To their credit, the authors do point out that preventive measures could make a significant effect in certain cancers, such as lung cancer and smoking. And their analysis seems to point out that prevention measures for cancers such as pancreatic islet cell, small intestine, and duodenum would be unlikely to prove effective—an argument that seems self-evident, given the rare nature of these cancers.

Stochastic effects (ie, the accumulation of random mutations within specific cellular pathways) contribute significantly to cancer etiology, and it is well known that the accumulation of these mutations in the course of a lifetime contributes to the increased risk of developing cancer as we age. However, the emphasis placed by Tomasetti and Vogelstein1 on stochastic events alone paints a very incomplete picture and, as we know, simple correlation does not always translate to causality.

However, the analysis by Tomasetti and Vogelstein1 has many other flaws,5 especially when we attempt to apply the findings to clinical practice. Foremost is the exclusion of several common cancer types from their data set—breast cancer (28% of all cancer in women), prostate cancer (26% of all cancer in men), small cell lung cancer (15% of all lung cancers), and squamous cell lung cancer (25% of all lung cancers). This “selection bias” in the core data set significantly limits the clinical applicability of the analysis.

In general, risk factors for the development of cancer can be placed into 3 major groups: heritable, extrinsic, and nonheritable intrinsic. Heritable factors include those germ-line mutations that are passed from one generation to the next; a classic example are mutations in the BRCA1 and BRCA2 tumor suppressor genes, which result in a risk 3 to 7 times higher in women for developing breast and ovarian cancers (as well as increased risk in men for developing breast and prostate cancer and melanoma). These heritable risk factors are largely considered to be nonmodifiable, although they may still be subject to epigenetic influences. In general, these are thought to account for only 5% to 10% of cancer incidence.

Extrinsic risk factors are nearly all related to diet, lifestyle, and environmental exposures, and they are generally modifiable to a large extent. These include tobacco exposure, obesity, sedentary lifestyle (an independent risk factor from obesity), diet (high fat intake, low fiber intake, etc), and heavy alcohol consumption. These extrinsic risk factors have long been at the core of cancer prevention programs.

The nonheritable intrinsic risk factors are those mutations that are not passed from generation to generation but that arise as acquired genetic changes occurring in somatic tissues by random chance during DNA replication. It is these nonmodifiable risk factors that Tomasetti and Vogelstein1 propose contribute to a majority of cancer diagnoses.

However, in January 2016, another study taking a different look at the question, came to alternative conclusion. Song Wu et al,6 publishing in Nature, conclude that “cancer risk is heavily influenced by extrinsic factors. These results are important for strategizing cancer prevention, research and public health.” To arrive at these conclusions, the authors utilized 4 independent, model-driven approaches:

1. Extrinsic Risks by Cell Tissue Turnover. In this analysis by Wu et al,6 which included both the breast and prostate cancer types not present in the original analysis,1 the authors demonstrate that “irrespective of whether a subpopulation or all dividing cells contribute to cancer, these results indicate that intrinsic factors do not play a major causal role.”

2. Epidemiological Evidence. Extensive epidemiological data demonstrate that the incidence of many cancers is significantly influenced by environmental factors. Perhaps most compelling are data showing that immigrants moving from a country with a low cancer incidence to one with a higher cancer incidence will acquire the higher risk of their new country.7

3. Analysis of Mutational Signatures. Recent mutational analysis has identified approximately 30 different “fingerprints” left on cancer genomes, differing by whether they were intrinsic or extrinsic in nature. They found that “the majority of cancers have large proportions of extrinsic mutations.”

4. Modeling Theoretical Lifetime Intrinsic Risk. Using complex mathematical modeling, the authors demonstrate that the theoretical lifetime cancer risk due to intrinsic DNA mutations alone is inadequate to account for the observed incidence of cancer.

When taken together, these 4 model-driven approaches all arrive at a consistent estimate of the degree of contribution that extrinsic factors have on cancer incidence: 70% to 90%. It must be kept in mind that this analysis, like that done by Tomasetti and Vogelstein,1 is a theoretical analysis of cellular data, not a real-world observation of actual people. However, it should give us further justification to diligently work to motivate our patients to make those important, but difficult-to-implement diet and lifestyle changes that will reduce their risk of developing cancer.

What Do I Tell My Patients?

The good news is that what most of us have been telling our patients to do still appears to be valid. Despite the “bad luck” nature of random gene mutations, an impressive body of evidence remains suggesting significant, simple steps that patients can take to reduce their risk of developing cancer. Eating a plant-based diet, maintaining a lean body weight, and avoiding tobacco exposure still form the basis of a rational cancer prevention plan. In addition, growing research is showing that a number of environmental toxins passively increase cancer risk, ranging from vinyl chloride exposure (apparently the only cause of hepatic angiosarcoma)8 to arsenic in water significantly increasing risk of bladder (2.7-fold)9 and lung (4.0-fold)10 cancers. Perhaps best of all, these same recommendations are likely to reduce the risk of heart disease and type 2 diabetes as well.

Simple Recommendations to Reduce Cancer Risk

1. Avoid tobacco exposure in any form.

2. Eat a healthy, plant-based diet. Focus on portion control; avoiding processed foods and limiting red meat; consuming at least 2.5 cups of fresh vegetables and fruits daily; and eating organic foods as much as possible.

3. Maintain a healthy weight. Typically target a body mass index (BMI) in the range of 18.5 to 24.9 kg/m². Further, limit high fat and high sugar foods.

4. Be physically active. Engage in 150 minutes of moderate intensity exercise per week, and limit sedentary activity: “Sitting is the new smoking.”

5. Limit alcohol intake. Consume no more than 2 drinks per day for men, and 1 drink per day for women.

6. Avoid exposure to environmental toxins and carcinogens, such as asbestos, radon, bisphenol A (BPA), vinyl chloride, and arsenic.