The future of cancer research

Abstract

New cancer research strategies have developed very rapidly over the past five years, including extensive DNA sequencing of tumor and normal cells; use of highly sensitive cancer cell detection methods; vaccine development and tumor-specific (designer) drugs. These developments have raised questions about where to concentrate efforts in the near future when establishing clinical trials, particularly important in an age of diminishing resources and during a period when competing strategies for cancer control are likely to overwhelm the opportunities for establishing large, effective clinical trials. In particular, it behooves the research community to be mindful of the inevitable, challenging obligation to responsibly choose between clinical trials that offer the credible hope of incremental advances vs. trials that are less traditional but may have revolutionary outcomes.

Keywords: :

• cancer and diet
• cancer and inflammation
• cancer prevention
• cancer screening
• cancer vaccines
• carcinogens
• designer drugs
• tumor promoters
• tumor-specific drugs

Prevention

Cancer usually requires mutations, and despite apparent exceptions related to epigenetic causes, significant cancer reduction has been achieved by carcinogen reduction. However, we now know that humans suffer tens of thousands of somatic cell mutations in a lifetime.^1,2 This raises the question, what would be the future opportunities for cancer prevention with additional strategies meant to significantly reduce mutation rates in somatic cells? Given DNA polymerase error rates, or other sources of mutations in somatic cells that cannot be practically avoided, what is the theoretical limit to reducing carcinogen-related mutation rates such that there will be a significant reduction in the potential of a deadly cancer?

Inherited, systemic mutations apparently only lead to cancer from relatively few cells, after a decade or more of life, often with less than 100% penetrance. A recent study has indicated that the risk of developing breast cancer by age 70 is 60% and 55% for BRCA1 and BRCA2 mutations, respectively.³ This leaves a relatively dramatic situation, which is given little attention, where every cell in the body has a cancer-predisposing mutation yet as many as 45% of carriers do not develop breast cancer.

The above data regarding mutagens, carcinogens, and inherited cancer mutations raise the question, to what extent is cancer facilitated by factors, other than carcinogen induced mutations, such as modifier genes, diet, inflammation, chemical tumor promoters, and even stochastic processes⁴? This question seems particularly timely in the case of persons with inherited mutations who never develop cancer. In the case of a systemic absence of one allele encoding a tumor suppressor protein, there is the expectation that cancer development requires the loss of the second tumor suppressor allele, as well as other mutations. Thus, in cases of persons with systemic mutations who do not develop cancer, are these cases the result of probabilistic events, simply the lack of occurrence of enough additional carcinogen exposure and resultant mutations in random cases? Or, are there other factors, such as modifier genes or non-carcinogen environmental factors that are key to cancer absence?

In the study referred to above,³ authors also noted an 87% incidence of contralateral breast cancer in the case of BRCA1 mutations, indicating that inherited modifier genes are only part of the answer to the question, what facilitates cancer development besides mutations? If this question can be answered for systemic mutations, the role of other factors in cancers attributed to somatic cell mutations would be ascertained more efficiently.

Screening

Given knowledge of cancer mutations, highly efficient DNA sequencing makes detection of circulating cancer cells, at extremely early stages, only a matter of time and money.⁵ It is already possible to detect some cancer cells so early that the value of treatment is unknown. This will be the case for all cancer cells. Thus, research challenges will be: what to screen for; what to treat; and who to treat, according to additional (mutational and non-mutational), potential risk factors referred to above?

Whole genome sequencing, or detection of small numbers of circulating cells, also offers the opportunity of “pre-screening” or post-screening for clinical trials, potentially allowing for more efficiency in determining which patient responds best. For example, a whole genome sequence, combined with the research literature, could rule out some patients and allow a productive focus on other patients. In a recent genome wide study,⁶ polymorphisms in the organic anion transporter gene, SLCO1B1, were associated with rates of methotrexate clearance, thus allowing future methotrexate clinical trials to segregate patients accordingly and the use of very simple technology, namely a PCR-based test, to detect the indicated polymorphisms.

How to Treat?

Along with watchful waiting, highly sensitive screening brings us to vaccines and tumor-specific drugs. Currently, there is wide acceptance of vaccine strategies if a cancer has a viral etiology,^7,8 with vaccines directed against the virus infection. With a low tumor burden, with reduced tumor-mediated inhibition of the immune system,⁹ and patients free of drugs that can prevent immune cell division, vaccines may be more successful for cancers that do not have an apparent viral etiology.

However, current data indicates that, at least in some cases, early tumor changes include an impact on the immune response. For example, loss of the retinoblastoma tumor suppressor protein is generally an early event and is accompanied by loss of major histocompatibility molecule inducibility.^4,10 Thus, it will be important to test preventative vaccines at very early stages.

Currently, there is a very wide gap in patient health when employing anti-tumor vaccines targeting tumor viruses vs. anti-tumor vaccines targeting well-developed cancers. This gap in experimenting with vaccines is largely traceable to a “do no harm” approach that understandably discourages experimenting with patients until there is a clear, severe risk to health. However, the opportunity to pre-screen patients, either for small numbers of tumor cells that can be genetically, indisputably identified as highly aggressive, or to screen for high risk patients based on modifier genes, will make it possible, for the first time, to justify targeting cancers, that do not have a viral etiology, with preventative anti-tumor vaccines, rather than therapeutic vaccines.

However, any vaccination strategy depends on knowledge of efficient immunogens, still a research challenge for cancer, particularly the earliest stage cancers, and for many infectious pathogens.¹¹ Also, the role of age and modifier genes, in vaccination success, needs to be understood.

The most reasonable, near future hope to reduce cancer deaths is rendering cancer chronic, with combinations of tumor-specific drugs, having few side effects, akin to the anti-AIDS drug cocktail. The key to meeting this goal is wider recognition of basic principles of cancer biology and natural selection. Application of one tumor-specific drug will select for small numbers of cancer cells with mutant, alternative signaling pathways,¹² allowing virtually the same cellular effector molecules to facilitate almost the same cancer.^4,12 The best chance for learning how to prevent selection for pre-existing cancer mutations is to plan clinical trials with simultaneous combinations of tumor-specific drugs lacking side-effects, exemplified by a case of combination of the BRAF inhibitor, Sorafenib, and the EGFR inhibitor, Cetuximab, in colon cancer.¹³ Some of the barriers, including intellectual property barriers, to developing clinical trials with such combinations have been recently discussed in detail by Humphrey et al.¹⁴ In some cases, outdated requirements by the FDA for pre-clinical studies have impeded the development of clinical trials with combinations of tumor-specific drugs, but recent efforts are intended to put more emphasis on basic scientific knowledge, such as the clear relevance of particular signaling pathways, hopefully reducing the experimental expenses before combination trials can be justified.¹⁵ Reducing the significance of animal models, where reasonably justifiable, may have the added benefit of success in humans that could not be predicted from animal results, due to significant differences in animal and human proteins, for example, differences in the binding affinities of mouse and human proteins for certain tumor-specific drugs.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Statement of Translational Relevance

This perspective provides a concise summary of recent basic and clinical research developments that will play a role in guiding clinician choices for participation in clinical trials.

10.4161/hv.27458