Malnutrition and Cancer: The Sad State of the Science

The majority of patients with cancer suffer from nutritional deficits, and up to 85% of patients with cancer diagnoses experience some form of weight loss and/or malnutrition during their cancer treatment (DeWys, Begg, Lavin et al. 1980).

For some, the nutritional deficits can progress to cancer cachexia, a form of malnutrition characterized by loss of lean body mass, muscle wasting, and impaired immune, physical, and mental function (Fearon, Strasser, Anker et al. 2011). Poor nutritional status and weight loss often lead to poor outcomes (Marin, Laviano, and Pichard 2007, Andreyev, Norman, Oates et al. 1998).

Overnutrition, as a factor in carcinogenesis, has been a matter of concern for more than 80 years. Metabolically active fatty tissue has been the focus of many epidemiological studies. On the other hand, the effects of undernutrition have been studied experimentally, but in insects or rodent models. In these model systems, the studies show relatively consistent results in that caloric restriction inhibits the growth of tumors. The mathematical effect is observed even when the calorie-restricted animals ingest more fat than do the controls. It may be that in some model systems, caloric restriction reduces insulin levels and decreases oncogene expression. Energy restriction may also increase activity of antioxidant enzymes and lead to enhanced DNA repair. However, applicability to human patients in any cohort seems dubious at this point.

At cancer diagnosis, half of patients have some form of nutritional deficit (Halpern-Silveira, Susin, Borges et al. 2010). Nutritional status often declines further during treatment. According to some longstanding work, up to 20% of patients die from the effects of malnutrition rather than from the cancer itself (Ottery 1994).

In addition to malnutrition and weight loss, host and tumor-derived factors result in loss of lean body mass. These presentations contribute to decreased immune function, which only exacerbates tumor burden, increased rate of infections, increased skin breakdown, decreased healing, and increased mortality (Sauer 2013).

Expert nutrition groups have issued clinical guidelines for nutritional treatment of cancer patients. These guidelines state that patients should undergo nutrition screening and assessment and receive early nutrition intervention to improve outcomes. However, it appears that there has been an absence of innovative or clinically meaningful techniques (August and Huhmann 2009, Arends, Bodokyt, Bozzetti et al. 2006).

So where are we really? With unprecedented sophistication in molecular biology, investigational oncology, and computational power, clinical medicine in general and oncologic nutrition in particular have failed to achieve meaningful progress for patients.

A personal survey indicates virtually every major “comprehensive cancer center” has a page on cancer nutrition support. The American Cancer Society’s “Nutrition for People with Cancer” website page declares, “Nutrition is an important part of cancer treatment. Eating the right kinds of foods during and after treatment can help you feel better and stay stronger.” While the Internet material is, of course, intended for patients, its quixotic but superficial character speaks volumes.

Upon review of the website information and firsthand assessment of clinical reality, one may conclude that there is a strong conventional wisdom that the pathology of cancers can result in malnutrition, that cancer treatments often result in malnutrition, and that chronic undernutrition and overnutrition can have an impact in either accelerating or reducing cancer risk or tumor burden.

There is an emerging science about the effects of nutrition on epigenetic regulation. Epidemiology, “body fatness,” and migration studies, naturalistic examples of extreme exposures or shortages, have provided insights on how the food matrix alters the epigenetic state of genes; detail is now readily available on factors that may impact critical histone and DNA methylation and that of other chromosomal proteins, noncoding RNAs, including miRNAs (microRNAs) and long-noncoding RNAs (Sapienza and Issa 2016).

The collision and complexity of possible causes and effects of cancer and independent and dependent variables that impact this dreaded disease is dizzying. There is a significant refrain in the literature of substantial evidence about the importance of dietary and nutritional factors, and the applications of food science to make nutrition support even possible in cancer etiology and treatment. Unfortunately, specific and clinically relevant nutrition relationships are elusive and somewhat speculative. The published recommendations for nutrition interventions are either diluted, sophomoric, or simply nonexistent. Meanwhile, there are impressive inventories of the daunting methodological hurdles in explaining and, in effect, apologizing for this troubled landscape.

It is proposed that a radical two-part redesign of the infrastructure of the classic cancer research center—one which focuses on a dedicated coordination function between design of investigation, performance of bench and clinical research, and delivery of public health and individual bedside outcomes—be undertaken. In addition to a coordination and management script, embodied in an exclusively dedicated team, the additional key component is that of a powerful, collaborative “big data” processing architecture.

An exemplar of such an entity is the AvesTerra platform developed at Georgetown University. Integral features are data harmonization, transforming data, sharing diverse knowledge derived from that data, performing analyses on distributed knowledge at scale, and visualizing results of large-scale computations. As an enabler for scientific discovery, this framework provides a method for integrating evidence from many sources spanning multiple modalities and levels of abstraction. This framework was formulated to automate the fusion of disparate data types and reasoning methods to unify connection of heterogeneous forms of scientific information to accelerate the research process. The result of this process is the production of an analytic ecosystem that is highly reusable, agile, and resilient to the continuous evolution of data and data processing techniques with the end goal of achieving unprecedented levels of knowledge density and accompanying analytic yield. The primary architectural construct for achieving these objectives in AvesTerra is termed an Application Programming Interface.

This author contends that a preconception of the conduct of both clinical and bench research, in conjunction with state-of-the-art computational architecture, is the only way we can move off a plateau in terms of understanding and effectively treating cancers in at risk populations as well as among individual patients who are desperately in need.

Roger Clemens, DrPH, CFS, Contributing Editor
Adjunct Professor, Univ. of Southern California’s School of Pharmacy, Los Angeles, Calif.
[email protected]