Charles Swanton of Cancer Research UK on How Tumors Branch Out

October 2012
Fast Breaking Paper Commentary, 2012
Charles Swanton

According to a recent analysis of Thomson Reuters Web of Science, the 2012 report “Intratumor heterogeneity and branched evolution revealed by multiregion sequencing,” (M. Gerlinger,  et al., New Engl J. Med., 366(10): 883-92, 8 March 2012) has been identified as a Fast Breaking Paper in Clinical Medicine. This designation is based on the paper’s increase in citations as tracked over two recent, successive bimonthly periods. The report, cited 64 times in Web of Science as of October 4, 2012, displayed a higher bimonthly citation increase than any other paper of comparable age and type in its field.

Corresponding author Charles Swanton is based at Cancer Research UK’s London Research Institute, and is also affiliated with University College London.


SW: Why do you think the paper is highly cited?

In the era of personalized medicine and targeted therapeutics, there has been much excitement. Despite this excitement and promise for improved survival outcomes, cancer patients experience treatment failure and drug resistance in advanced solid tumors on a daily basis. Furthermore, our ability to validate biomarkers has been limited, with very few predictive biomarkers of drug response implemented in routine oncological practice. These complexities seemed at odds with relatively simple models of linear cancer evolution and clonal sweeps that we are taught at degree-level cancer biology. This work provides evidence, which many clinicians have said they suspected but had no evidence for, of diversity at all genetic and transcriptomic levels leading to branched (rather than simple linear) evolutionary growth where primary and metastatic sites evolve independently. Regional separation of cancer subclones within individual primary tumors was also witnessed with up to two-thirds of somatic mutations not detectable in every tumor biopsy in the first two patients analyzed, indicating that one tumor biopsy may not provide a complete picture of the cancer’s evolutionary history or underlying genetic change. These data provide modest insight into the complexities of solid tumor management both for clinicians and cancer biologists, into the difficulties in validating biomarkers (where individual biopsies may differ from one another at the genetic level) and into the inevitability of acquired multi‑drug resistance driven by cancer diversity and subclonal selection of drug-resistant tumor cells in advanced cancers.

SW: Does it describe a new discovery or new synthesis of knowledge?

This paper uses new tools to resolve an old question. Intratumor heterogeneity has long been known to pathologists and cancer biologists. However, the true extent of intratumor genetic diversity is only just beginning to be untangled by next-generation sequencing technologies such as those used in our paper.

SW: Would you summarize the significance of your paper in layman's terms?

Cancer doctors rely increasingly upon the interpretation of single tumor biopsies, often taken months or years previously, to offer patients the correct treatment  (so-called personalized medicine). Doctors have an increasing array of therapies available that can block cancer growth, and as our understanding of tumors increases, so will the number of drugs available for individual tumor types. A question that has puzzled our laboratory is, how reliable is a single biopsy when it comes to deciding on a specific treatment at the right time in the disease course? We therefore set out to address how representative one biopsy is of the tumor's entire genetic change as a whole, and how cancers grow through time.

We sequenced all the coding genes in the human genome in multiple regions of the same tumor and found that the differences between biopsies outweighed similarities. Overall, the diversity in large renal cancers was greater than we expected. Results from this work contribute to a growing body of evidence that one biopsy, the mainstay of current cancer treatment stratification, may not always be representative of the tumor genetic landscape, and that tumor genomes are likely to change over time.

We anticipate that tumor diversity is likely to challenge our ability to improve cancer survival outcomes. We are addressing the clinical impact of tumor diversity by tracking tumor evolution over time to attempt to understand how tumors change and adapt to new environments and how underlying diversity may contribute to drug resistance and tumor metastasis.

On the basis of these results, we proposed a tree model of cancer growth, where the tumor "trunk" harbors the early genetic changes leading to tumor formation, which will be present in every cancer subclone and every tumor biopsy. In contrast, the tumor "branches" represent later genetic changes in the growth of the tumor that are not present in every cancer cell or every tumor region. The branches therefore represent the diversity within the tumor. On this basis we suggested that targeted drugs may work better if they are targeted against changes in the trunk of the tumor, present in every tumor cell. Knowledge of the tumor's evolutionary history, and the early abnormalities in the trunk, may provide a better understanding of how drugs can best be offered to patients.

However, the diversity present in the branches may also contribute to tumor growth at different sites and to treatment failure, indicating the potential need to target these changes too. Work from our laboratory had previously shown that the underlying diversity of a tumor (its branches) is associated with resistance of tumors to multiple different drugs. The results in this paper, showing extensive diversity in metastatic renal cancer combined with our previous findings that tumor diversity was associated with resistance to multiple anti-cancer drugs, began to offer an explanation for how metastatic tumors evade drugs so readily.

SW: How did you become involved in this research, and how would you describe the particular challenges, setbacks, and successes you've encountered along the way?

I work as a cancer clinician and a scientist. Any success we have had is entirely attributable to my laboratory staff, outstanding scientific mentorship, and UK research funding. Mentorship from senior Cancer Research UK scientists, who have always supported my attempts to combine a career in medicine and science, has proven invaluable over the last decade. The research infrastructure and funding support from Cancer Research UK at the London Research Institute, the Medical Research Council and University College London, which have always actively encouraged clinician-scientists to develop independence, has had an immeasurable impact on our work. Trying to combine two careers has been challenging but immensely rewarding over the last 15 years, and it has taken patience and understanding from my laboratory supervisors during my training that clinical duties and medical-examination hurdles always had to come first.

I was fortunate enough to be given the opportunity to study on the University College London Medical Schools MD PhD program in 1995 at the Imperial Cancer Research Fund Laboratories. This allowed me to combine my interests in basic science with clinical medicine from an early stage in my career. The mentorship and guidance from my PhD supervisor Nic Jones at the time were unforgettable and laid the foundations for my love of science and its power to address clinically relevant questions. This became much more focused during my cancer clinical training, where we began to address the fundamental clinical question with no scientific explanation, of why metastatic solid tumors evolve resistance to all available drugs over time. It is this simple question that has preoccupied my time as a post-doctoral fellow and over the last five years since independence.

There have been setbacks. Not least of all, the early struggle for funding to return to science after junior doctor training. Following two worrying grant rejections, I was awarded a Cancer Research UK Clinician Scientist Fellowship grant in 2004 which enabled me to combine cancer training with a postdoctoral position in Julian Downward's laboratory where we were involved in the early days of RNA interference screening to attempt to identify mechanisms of cancer drug resistance. Similar to Nic Jones, Julian was very supportive of trying to combine two careers. In retrospect I can see how important both successful funding and finding the right supervisors were to my future career. Had either one of these not been right I would not be running a laboratory now, as the pyramidal career structure in science is severe, cutting short many early careers.

My postdoctoral fellowship was followed by another rejection for funding to gain independence, which was probably the most severe research setback. However, both the Medical Research Council and Cancer Research UK offered funding through their Senior Clinical Fellowship programs to set up my independent laboratory at the London Research Institute in 2008; this has given us the freedom to grow and pursue our own interests in cancer medicine.

Without a doubt, this is an enormously privileged job, being surrounded by great scientific minds who keep this work going. Almost every day science delivers a new answer, positive or negative, to a clinically relevant question and nature in general, which one day we hope will result in practical and positive changes to the way we treat patients.

SW: Where do you see your research leading in the future?

We anticipate that tumor diversity is likely to challenge our ability to improve cancer survival outcomes. We are addressing the clinical impact of tumor diversity by tracking tumor evolution over time to attempt to understand how tumors change and adapt to new environments and how underlying diversity may contribute to drug resistance and tumor metastasis. Analogous to the tumor tree, we are testing whether cancer treatments prune some tumor branches and enrich for others, enabling tumors to grow and branch (and by implication diversify) further. Our early insight into this process suggests that some of these changes may be predictable and potentially targetable.

As a result of this knowledge, our laboratory’s main focus is to address how tumors initiate their branched development by identifying cancer genes that serve to protect the human genome from diversity or accelerate it.

SW: Do you foresee any broader social implications or impact for this research?

Progress in science is entirely unpredictable; history tells us to expect the unexpected. The desire for broader social implications and societal benefit, balanced against the clinical experience of cancer treatment failure, maintains the laboratory’s focus on the questions we hope are relevant to patient outcome. We are aiming to identify abnormal genetic events in a tumor that lead to diversity. These events may be detectable in tumors at an early stage of disease and forecast poorer survival outcome if surgically resected tumors are left untreated. If we can identify processes in the cancer cell that initiate differences between cancer cells, it may be possible to develop therapeutic approaches to limit tumor diversity and in so doing, arrest the adaptive changes that may contribute to treatment failure and tumor relapse following primary surgery.

Charles Swanton
London Research Institute
Cancer Research UK
London, England

cancer, biopsy, tumor genomics, tumor diversity, personalized medicine

The data and citation records included in this report are from Thomson Reuters Web of ScienceTM. Web of ScienceTM is a registered trademark of Thomson Reuters. All rights reserved.