This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American
Please welcome this month's Scicurious Guest Writer, Karissa Milbury!
“This is why our drugs fail. Look at it. How do you treat that?” The professor, speaking to our graduate genetics technology class, was referring to a figure similar to this:
(Figure 1: A graph showing the array of mutations found in a single breast cancer, from a paper by Natrajan and colleagues[1].)
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
If that seems complicated to you, don’t worry – this is a genetic map of a single breast cancer, with every black line around the edge representing one mutation. When it comes to using data like this, scientists are overwhelmed by it, too.
The field of genetics has flourished with the publishing of the complete human genome in 2001, aided by the advent of fast, affordable sequencing technology. A completed genetic code of a healthy person allows us to compare against the genetics of cancer. With advanced analytical techniques, and decades of research into the characteristics of different forms of this disease, it seemed that it was finally time to pull out the answers from the code itself by looking for the mutations that cause or support the cancer’s growth – the differences between the cancer cell and a normal cell. But when the answers didn’t bubble up from our statistics and reams of data, it became clear that the questions left for us were far more complicated.
Life is messy: our distinctions between different species, different organisms, and different cells are largely arbitrary, because as much as we attempt to separate and define these categories, we always run into exceptions to our rules. As scientists, we seek to investigate the world through observation, classification, and prediction, but every distinction we make dissolves once we look closely enough: even the line between life and non-life is blurred. Even though this is an old problem in biology, cancer geneticists are only now running up against this as we attempt to decipher the specific changes to the genetic code responsible for driving cancer progression. To develop a specific drug for a specific cancer, we need to be able to tell the difference between a cancer cell and a healthy cell in a meaningful way. But it’s not so easy.
Grinding to a halt
Modern medicine has given us many tools for fighting disease, such as antibiotics, antivirals, and advanced diagnostic technology like magnetic resonance imaging (MRI). In the case of cancer, a set of diseases characterized by a population of cells multiplying out of control, most of our modern methods attempt to stop cellular replication. Surgery can remove many of the malignant cells, slowing down the cancer’s frenzied growth, while radiation therapy and chemotherapies such as taxol selectively kill cells that are dividing[2]. If diagnosed early, some cancers, such as certain types of melanoma or prostate cancer, are potentially curable[3,4]. Unfortunately, cancer is often diagnosed later, beyond this sensitive window. A true therapy for cancer must address these dangerous late-stage cases. So far, this has proven extremely difficult, due to the aggressive nature of advanced cancer.
Aside from the benefits we’ve seen from lifestyle shifts (such as reduced cigarette usage) and improved early detection of disease[5], there have been paltry gains in long-term survival from actual cancer therapeutics for most types of cancer[6]. While it’s true that this can be partially attributed to the increasing difficulty of acquiring drug approval[5], there’s another issue that can’t be dismissed as bureaucratic: the drugs we do approve can’t technically cure people, because we’re not able to accurately target a cancer with drugs like we do a virus or bacteria[7]. As well, if the disease returns, it is invariably resistant to the treatment that worked before, and further therapy is almost exclusively palliative[8]. Newer treatments also use sequencing and targeted molecular therapies to attack cancer cells specifically. Many researchers remain certain that each cancer will have a genetic Achilles heel, if only we can identify it. We should be able to find a mutation, or mutated pathway, or some druggable aberration specific to the tumor. I’m not convinced the answer will be that simple.
As an undergraduate professor of mine once said, “I know a thousand ways to cure cancer, it’s just that they all kill the patient, too.” I can kill cancer cells in a dish with bleach, but you obviously can’t give bleach as a treatment. For drugs that are approved, usually cancer cells aren’t the only cells suffering from the effects of treatment. Anyone that has undergone a brutal chemotherapy regimen knows this intimately, as the suffering of their normal cells result in hair loss, nausea, and a host of other side effects. As targeted as they attempt to be, often our treatments are still, first and foremost, poisons.
So what has gone wrong with our plans to target cancer abnormalities, and the ideas that have guided much of cancer research for several decades? The reason for the failure is the same as always: life is messy. A breast cancer is not simply a breast cancer, because it is a particular person’s breast cancer, and is operating from their altered code. Cancer cells can carry different markers, use different growth pathways, expand at different rates, and metastasize in different directions. Each cancer is as different as each patient, with its own code and characteristics[9], making them extremely different to treat with drugs that can only hit a certain range of targets.
But the problem is more complicated still. Following on several studies performed in the pre-genomic era[10,11], researchers began sequencing a number of biopsies from the same tumor, and demonstrated that different sections of a tumor can contain different genetic codes[12]. As the tumor cells divide, they are pressured by the limitation of space and nutrients, and are driven to compete amongst themselves to acquire adequate resources. This pushes the cells to evolve – like a population of microorganisms, the tumor diversifies as it divides. Some lineages will be sensitive to a particular drug; others won’t. Some lineages will be sensitive to radiation, while their dormant cousins remain unaffected. In every cancer, in every person, the cells will act differently. Some studies are now suggesting that chemotherapy may sometimes induce dormant tumor cells to become malignant[13]. All the while, each of these cells is experiencing a dramatically increased mutation rate – they are unstable and may die, but those that don’t die can evolve quickly.
Chaotic data
Dr. Ashutosh Jogalekar, in his blog The Curious Wavefunction, wrote in 2011 about how random events can have profound impacts on the way our world exists today[14]. Because of the importance of these random events, he explains, it would be impossible to predict chemical structures from a knowledge of physics, and similarly, it would be impossible to predict biology from a knowledge of chemistry. This is the essence of evolution: selection acts on random events, and even under similar circumstances, evolution can play out differently because it feeds on randomness. In cancer genetics, we take our knowledge of nucleotides (the building blocks of DNA) and proteins, and our knowledge of their chemical interactions, and use this to explain the actions of the cancer. We quickly run into a diagnostic wall: between these two levels of analysis, selection is acting upon random code changes, which are inherently impossible to predict.
One method for addressing chaos of this sort is to develop a statistical plan. We want to be able to take the information we know, look at the probabilities that certain changes will occur, and use statistics to determine what the cancer will likely do next. This challenge seems insurmountable when you look at the variables we must contend with: rapid evolution, unchecked growth, subtle migration, and so on. Changes that occur in the genome during cancer initiation and progression involve massive genetic rearrangements, damage, and mutation. This makes it difficult to distinguish between causes and consequences of the cancer. It would be far easier if each gene, or even each chromosome, carried out its business without interaction with the rest of the cell. But not only can changes in one area of the genome have profound direct and indirect effects on the expression of genes elsewhere, we now have evidence that cancer cells can send these activity-altering signals to other cells, both tumor-derived and normal[15,16]. This further affects our attempts to analyze the genetic structure of tumors; no longer can we treat a lineage of tumor cells independently. Instead, we must accept the possibility that signals originating in one cell can have effects in others. They are not only diverse: this diversity constantly changes, and the information in one cell is communicated to others in ways we are only just beginning to understand[17].
The image below demonstrates how the composition of a tumor can change over time; imagine the green cell on top as being the first cancerous cell, and every new color indicates a fundamental change in the code as the cell divides down through the figure. If you imagine all of those cells having direct effects on some of their neighbors, you can start to appreciate the difficulty of untangling information about the cancer’s growth from the genetic codes of the resulting tumor.
(Figure 2: Five simplified models for cancer cell lineage growth, adapted from Navin & Hicks (2010)[18].)
Moving forward
How can we develop a cancer therapy sophisticated enough to address such a profoundly complex disease? Currently, we cannot predict cancer progression this way any more accurately than a meteorologist can predict the weather – we have some ideas, we know trends, and we have mountains of data, but otherwise it’s all conjecture. Our predictions are almost always wrong if we try to predict too much.
The answer is not treating cancer like an infection that can be directly targeted or removed. Research in areas beyond traditional targeted molecular drugs, where we have begun treating cancer analysis with a multifaceted approach, is beginning to show promise. Immunology has offered tantalizing hints at what is possible when you take advantage our natural, adaptive defense mechanisms: our immune systems are very good at systematically attacking invaders, and enhancing these cells with genetic technology has shown increased survival for patients with certain types of cancer[19-21]. Scientists are also looking at ways to use viruses to chase down cancer cells specifically without harming healthy cells nearby[22]. Bioinformaticians are moving biological analysis to a new era with algorithms designed to recreate and track the flow of genetic information. This has opened our eyes to signatures of cancer in the genetic code that we never knew were there, but may be useful in diagnosing early-stage disease[23]. These complex analyses are needed if we are to dig down to the complicated roots of cancer, far beyond even the genetic code itself, to decipher the forces driving cancer progression.
I am not the first to write about the changing face of cancer genetics, and I expect to see much more discussion on this subject as we continue pushing the boundaries of cancer therapeutics[24,25]. To face the challenges of cancer we must look to other fields, especially computer science, chemistry, and physics, see what other scientists and mathematicians are doing to work with data this complex. Hopefully this will allow us to make a broad interdisciplinary move to creatively re-think our tactics. I've heard it said that the age of scientific genius is over, because the discoveries of today require massive computational efforts and the answers are not intuitive – but maybe we've lost sight of the fact that this flavor of genius is bold and imaginative, not conservative and shy. Our intuition about cancer is changing, and we must change as a field to keep up with it, however intimidating that may be. If that means learning more chemistry and math, then it’s time cancer researchers hit the books.
References:
[1]: Natrajan, R. et al. 2012. A whole-genome massively parallel sequencing analysis of BRCA1 mutant oestrogen receptor-negative and -positive breast cancers. J Pathol. 227:29-41.
[2]: Arnal, I. & Wade, R. H. 1995. How Does Taxol Stabilize Microtubules? Curr Biol. 5(8):900-908. doi:10.1016/S0960-9822(95)00180-1
[3]: Carter, H. B. et al. 1999. Influence of Age and Prostate-Specific Antigen on the Change of Curable Prostate Cancer Among Men with Nonpalpable Disease. Urology. 53(1):126-130.
[4]: Jaimes, N. & Marghoob, A. A. 2012. An update on risk factors, prognosis and management of melanoma patients. G Ital Dermatol Venereol. 147(1):1-19.
[5]: de Bono, J. S. & Ashworth, A. 2010. Translating cancer research into targeted therapeutics. Nature. 467:543-549. doi:10.1038/nature09339
[6]: Siegel, R. et al. 2012. Cancer Statistics, 2012. Ca Cancer J Clin. 62:10-29. doi:10.3322/caac.20138.
[7]: Blackwell, T. 2013. “Is the war on cancer an ‘utter failure’?: A sobering look at how billions in research money is spent”. National Post, News. http://news.nationalpost.com/2013/03/15/war-on-cancer/. Retrieved May 17, 2013.
[8]: Steeg, P. S. 2006. Tumor metastasis: mechanistic insights and clinical challenges. Nat Med. 12(8):895-904. doi:10.1038/nm1469
[9]: Nik-Zainal, S. et al. 2012. Mutational Processes Molding the Genomes of 21 Breast Cancers. Cell. 149:979-993.
[10]: Heppner, G. H. et al. 1983. Tumor heterogeneity: biological implications and therapeutic consequences. Cancer Metast Rev. 2:5-23.
[11]: Nicolson, G. L. 1984. Tumor progression, oncogenes and the evolution of metastatic phenotypic diversity. Clin Expl Metastasis. 2(2):85-105.
[12]: Marusyk, A. & Polyak, K. 2010. Tumor heterogeneity: Causes and consequences. Biochim Biophys Acta. 1805:105-117.
[13]: Kreso, A. et al. 2013. Variable Clonal Repopulation Dynamics Influence Chemotherapy Response in Colorectal Cancer. Science. 339(543):543-548.
[14]: Jogalekar, Ashutosh. “Why biology (and chemistry) is not physics”. The Curious Wavefunction. Published on Mon, Aug 29, 2011. http://wavefunction.fieldofscience.com/.
[15]: Camussi, G. et al. 2011. Exosome/microvesicle-mediated epigenetic reprogramming of cells. Am J Cancer Res. 1(1):98-110.
[16]: Deregibus, M.C. et al. 2007. Endothelial progenitor cell-derived microvesicles activate an angiogenic program in endothelial cells by a horizontal transfer of mRNA. Blood. 110:2440-2448. doi:10.1182/blood-2007-03-078709
[17]: Ghose, T. 2011. “Cancer's Escape Routes”. The Scientist, News & Opinion. http://www.the-scientist.com/?articles.view/articleNo/31454/title/Cancer-s-Escape-Routes/. Retrieved May 17, 2013.
[18]: Navin, N. E. & Hicks, J. 2010. Tracing the tumor lineage. Mol Oncol. 4(3):267-283. doi:10.1016/j.molonc.2010.04.010.
[19]: Rosenberg, S. A. et al. 2008. Adoptive cell transfer: a clinical path to effective cancer immunotherapy. Nature Rev. 8:299-308.
[20]: Vincent, K. et al. 2011. Next-generation leukemia immunotherapy. Blood. 118:2951-2959.
[21]: Coghlan, A. 2012. "Souped-up immune cells force leukaemia into remission". New Scientist, Health, News. http://www.newscientist.com/article/dn22613-soupedup-immune-cells-force-leukaemia-into-remission.html. Retrieved May 17, 2013.
[22]: Comoli,P. et al. 2005. Cell Therapy of Stage IV Nasopharyngeal Carcinoma With Autologous Epstein-Barr Virus–Targeted Cytotoxic T Lymphocytes. J Clin Oncol. 23(35):8942-8949.
[23]: Kim, Y.-A. et al. 2011. Identifying Causal Genes and Dysregulated Pathways in Complex Diseases. PLoS Comp Biol. 7(3): e1001095. doi:10.1371/journal.pcbi.1001095 [24]: Fessenden, M. 2013. "News from the Front in War on Cancer--Mission Not Accomplished". Scientific American, Health, News. https://www.scientificamerican.com/article.cfm?id=news-from-cancer-war. Retrieved May 18, 2013. [25]: Chi, K. R. 2010. "The New Face of Cancer". The Scientist Magazine. http://www.the-scientist.com/?articles.view/articleNo/29013/title/The-New-Face-of-Cancer/. Retrieved May 16, 2013.
Karissa Milbury is a graduate student in the Genome Science & Technology program at the University of British Columbia, in Vancouver, BC, working with Dr. Julian Lum. She received her BSc from Dalhousie University in Halifax, NS. Her scientific interests include genetics, bioinformatics, and immunology, and her research involves using genetic analyses to harness the power of the immune system in order to develop promising new cancer therapies. Beyond the lab, she invests herself in science communication, biking, and earl grey tea. You can visit her blog, Point Mutations, or follow her on Twitter (@Point_Mutation).