Wednesday, November 4, 2015

The Paradigm of Cancer Care in the Era of Genomics, Proteomics and Immuno-oncology

The word “paradigm shift” is used fairly loosely in many contexts, but in the field of Oncology there has truly been a paradigm shift in both the understanding of cancer biology as well therapeutics in the last decade and the pace of change has accelerated even more so in the last couple of years. Those outside the fields of oncology and related fields are unaware of many of these seismic shifts taking place, and this blog is an attempt at briefly updating the reader on what is happening in the field of oncology.

Let me start by illustrating by the example of a current patient of mine; she is a 56 year old patient, non smoker, with stage IV adenocarcinoma of the lung with brain metastasis, diagnosed over 2 years ago. If some one asked me how to treat a patient like this when I was doing my Hematology-Oncology fellowship over 12 years ago, the appropriate answer would have been “any doublet chemotherapy containing a platinum compound”, along with brain radiation. Basically, it didn’t matter what kind of mutations the tumor itself possessed as we not only knew little about mutations withing lung cancer, we also had no specific or targeted therapy for individual mutations even if we found them then. So all lung cancers were treated as a single group, and given a pretty similar type of chemotherapy regimen. Some of the patients responded to this, and those who did not and those who progressed, there were few, if any viable options apart from palliative care.

What has changed in over 12 years? So many things. For once, we have now so many different sub types of lung cancers, based not just on histology but also based on molecular and genetic profiling. My particular patient was found to have a lung adenocarcinoma with a so called EGFR exon 19 mutation, which is very sensitive to an oral kinase inhibitor called Afatnib. My patient initially underwent so called “Cyber knife Radiation” for her brain metastasis – which is a highly focused form of radiation targeting the metastasis while sparing normal brain. After this she was started on afatinib. This is a pill taken once daily while she continued to work. A PET scan done after 3 months of this therapy showed she was in a PET negative remission. She continued the same medication for another 18 months but unfortunately progressed with a new bone lesion. At the time of progression, a new test which can test mutational status of any circulating tumor cells became available and was performed, and this showed she had now circulating tumor cells positive for a mutation called T790M. This is a classic mutation which makes the EGFR clones resistant to drugs like afatinib. These mutant cells are actually a bit less robust than cells without the T790M mutation, but in the presence of drugs like Afatnib the cells with the T790M mutation have a specific survival advantage.

At this point she was enrolled in a clinical trial looking at a so called third generation EGFR kinase inhibitors which works around the T790M mutation, being developed by Astra Zeneca. She responded nicely for about 5 months on this yet to be approved drug; then around 1 month ago she progressed again. A repeat cell free DNA testing now showed she has developed cells expressing so called BRAF V600E mutation, which was not present in the original tumor or on first progression – one of her cancer cells have now mutated to be now the “fittest” clone. We have now obtained a BRAF inhibitor and a MEK inhibitor on a compassionate use protocol (as these drugs are not yet approved for lung cancer but only for melanoma). We and the patient and her family are all keeping our fingers crossed to see what would happen. My guess is she will respond for few months, and then her cancer cells will find a way around that pathway also and then progress, but hopefully we will find another mutation which we can target by then, or she could get conventional chemotherapy which is yet to receive nearly 2 years after diagnosis. And during these nearly 2 years of ongoing therapy, patient continued to work and enjoy a fair quality of life allowing her to pursue her work and hobbies. 

The above case illustrates how from randomly treating all lung cancer patients with a generic “platinum doublet” 10 years ago, cancer treatment now involves sub dividing the cancers in to very specific molecular sub types and treating with appropriate medications. The case also illustrates how dynamic is the process of cancer; we spoke about targets in the past as if they were fixed (like "estrogen receptor positive breast cancer"). We still do so, but is realizing these targets change and mutates and cancers evolve over time. While vast majority of cancers still do not have "targetable mutations " like the patent's case above, we are at the cusp of some exciting new fronts in cancer care. What follows is a very short summary of cancer biology as well as how this is changing how cancer is treated now a days, along with a brief overview of future potentials and challenges.  

Cancer as an Evolutionary Process:

What we have learned is that the best way to look at the cancer is through a Darwinian lens. This so called micro evolution – or evolution at cellular levels and happening at warp speed can explain a whole lot of once mysterious things about cancer biology. In the past we used to look at cancer as a “monoclonal process”, a set of uniform cells competing against normal cells. This appears to be a very simplistic way to look at cancer – cancer is unfortunately a way more complex disease process than that.

What is meant by looking at cancer through Darwinian lens? Here is how: A body can be considered as an entire eco system –whose individual members are cells. These cells have ecological and classic Darwinian features like cell births, deaths, habitats, territorial limitations, and maintenance of population sizes. The one Darwinian rule that does not apply is however “natural selection” – there is no competition among somatic cells. The rules of somatic cells are instead self-sacrifice—as opposed to survival of the fittest. Ultimately, all somatic cells are committed to die; they dedicate their existence to support of the germ cells, which alone have a chance of survival and propagation. One could think of germ cells like the “queen bee” of an ant colony. There is no mystery in this, as the body is a clone, and the genome of the somatic cells is the same as that of the germ cells. By their self-sacrifice for the sake of the germ cells, the somatic cells help to propagate copies of their own genes.

To coordinate their highly cooperative behavior, cells send, receive, and interpret an elaborate set of signals that serve as social controls, telling each of them how to act, when to divide, when to die. As a result, each cell behaves in socially responsible manner, resting, dividing, differentiating, or dying as needed for the good of the organism. Molecular disturbances that upset this harmony mean trouble for a multi cellular society like our body. In a human body with more than 10^14 (or 100 Trillion) cells, billions of cells experience mutations every day, potentially disrupting the social controls. And an occasional mutation can give one cell a selective advantage, allowing it to divide more vigorously than its neighbors and to become a founder of a growing mutant clone, a “selfish” clone. Once cells learn to be “selfish” then the typical Darwinian “survival of fittest” rule applies within that clone– each subsequent cancer cell population gets “better” at resource utilization. And since mutational rates are higher among cancer cells, purely from random chance one of these cells would develop a special property (like more efficient anaerobic metabolism) - this cell then have a survival advantage over other normal and cancer cells. Note that by this time cancer cells are competing among themselves, and not against normal cells. Because cancer cells are way so advanced in their ability to divide and utilize resources, competing with normal cells is like Michael Jordan playing against a school kid. Their competition is mostly with other cancer cells.  Such repeated rounds of mutation, competition, and natural selection operating within the population of cancer cells cause matters to go from bad to worse, see the diagram below. 

For example: when a critical size is reached, the cancer cells may not get enough oxygen, then one of these cancer cells – who are all competing with each other like it is the great African Serengeti plains – would develop a mutation that allows it to utilize anaerobic pathways better, or attract blood vessels to grow in to it, or develop an ability to pump out chemotherapy drugs etc.  Several more cell divisions later another cell would find a way to get out of the resource poor primary site and goes and thrive in another, healthier and "less competitive" environment - what we call metastasis. See the diagram below to show how clonal evolution works:

This is why recurrent cancers are harder to treat and cure. Thus cancer is a disease in which individual mutant clones of cells begin by prospering at the expense of their neighbors, but soon their competition is among cancer cells themselves, each “out mutating” each other, eventually becoming such a remarkable , and almost immortal, dividing cells, who only dies in the end when they destroy the whole body. If those cells can be taken out and given perpetual nourishment, these cells can literally live forever (portrayed in the wonderful book “The Immortal Life of Henrietta Lacks). These advanced cancer cells have actually mastered what humans have been searching for all of history - immortality.

As noted a single mutation alone can rarely causes cancer. Genesis of a cancer typically requires that several independent, rare accidents occur in the lineage of one cell. If a single mutation were responsible, occurring with a fixed probability per year, the chance of developing cancer in any given year should be independent of age. In fact, for most types of cancer the incidence rises steeply with age—as would be expected if cancer is caused by a slow accumulation of numerous random mutations in a single line of cells. In fact if one lives long enough, it is almost a given that some type of cancer is bound to happen. A recent study from Sweden looked at whole genome sequencing of healthy adults, and found that over 12% of general population over age 65 had what we now call as ARCH (Age Related Clonal Hematopoiesis) in their blood – and they had 15 times risk for developing subsequent blood diseases like AML and MDS. (Interestingly those with ARCH in their blood also had higher cardio vascular and endocrine morbidity and mortality suggesting a possible common inflammatory pathway to both cancers and other common diseases)

Why are so many mutations needed for cancer? Because think of cell as a car; but instead of a single accelerator and a single break, cells have many many breaks and accelerators. A critical number of breaks have to be gone and a critical number of accelerators pressed before a cell truly becomes “cancerous”. (By the way those with familial cancer syndrome like Li Freumani already are born with some major breaks - like p53 - already gone, so it takes less mutation for them to get a cancer). Moreover, not all cancers are the same in terms of how "fast" they are. In fact we now routinely use a test called “Mammaprint” to decide which of the early stage breast cancer patients need treatment, and who does not. I explain to patients this test differentiates whether their cancer is an old car or a brand new Ferrari. And if they are lucky and if their cancer is a like an old car, we don’t give them chemo these days and treat them just with hormonal therapy.

Just to add some subtlety to this story – mutations alone can’t cause cancer – often many mutations are needed unless it is one of the “driver” mutations like 9-22 translocations in CML. An estimated 10^16 cell (100,00 Trillion) cell divisions take place in a normal human body in the course of a lifetime. (By the way these are truly cosmic numbers. For example an average human have 1000 times the number of stars in the Milky Way galaxy!)  Even in an environment that is free of mutagens, mutations will occur spontaneously at an estimated rate of about 1 in a million mutations per gene per cell divisions.—a value set by fundamental limitations on the accuracy of DNA replication and repair. Doing the math one can see that in a lifetime, every single gene is likely to have undergone mutation on about 10^10 (or 10 billion) separate occasions in any individual human being. Among the resulting mutant cells one might expect that there would be many that have disturbances in genes that regulate cell division and that consequently disobey the normal restrictions on cell proliferation. From this point of view, the problem of cancer seems to be not why it occurs but why it occurs so infrequently. I routinely get asked by patients – often who have done all the right things from their diet to exercise who gets cancer – why they got cancer. (Without elaborating the numbers what I tell them is this: That it is a miracle of our immune system that we don’t get a new cancer every day!) 

What is a Driver Mutations versus “Passenger” mutations:

Countless studies have shown that sequential acquisition of mutations results in gains in evolutionary fitness. Furthermore, tumor initiator clones (also often referred to as cancer stem cells) have been identified in a subset of cancers and highlight the potential for a genetically “simple” tumor cell progenitor to propagate disease relapse. There is perhaps no disease with greater evidence of this than CML.

The introduction of imatinib, a small molecule inhibitor of ABL family kinases including the BCR-ABL fusion gene, revolutionized the way that CML is managed and dramatically improved outcomes for these patients. An important factor contributing to the unusual success of imatinib is that it targets the initiating event in the clonal evolution of CML. This means that all daughter cells that evolve following this initial event (ie, every cell in the clonal pool) also carry the BCR-ABL trans location and are susceptible to the effects of imatinib. . The terms “driver mutation” and “passenger mutation” were coined to discriminate between (1) those mutations that play an active role in disease pathogenesis (ie, driver mutations) and (2) those mutations that do not contribute to disease pathogenesis but undergo clonal expansion alongside one that does (ie, passenger mutations). Also note that not all driver mutations are created equal but rather are acquired in an ordered hierarchy. That is, some driver mutations occur as early events during clonal evolution and play a role in disease genesis (early drivers), whereas others occur as later events during clonal evolution and play a role in disease progression (late drivers/accelerators). Early driver mutations that have a role in disease genesis, such as the BCR-ABL trans location, will therefore be present in every tumor cell, whereas late driver mutations may only be present within a subset of tumor cells (ie, in a subclone).

To further complicate understanding and measurement of the clonal origin of mutations, each driver mutation will confer a variable boost in evolutionary fitness, which will cause them to overtake less-fit clones at different rates. This means that some driver mutations, despite occurring as late events in disease evolution, may appear to be present in the majority of tumor cells because they provide a significant boost to clonal fitness.

Currently, we do have “actionable” mutations (mutations that are matched to targeted therapies) for a number of cancer types, but as of this writing we know of many more “mutations” than there are “actionable mutations”. However, the future of precision medicine is one in which we will have a much wider array of actionable mutations matched to suggested therapeutics or clinical trials. Understanding the hierarchical order in which somatic mutations are acquired in each cancer will become an important consideration in ranking therapeutic targets for drug development, but this is a complex scientific undertaking, to say the least, and would require significant computing and clinical resources. A future format of molecular genetic results should incorporate a measure of not just the presence or absence of a mutation, but also the clonal representation or allelic frequency for each mutation, so that oncologists can be more informed about the biology of the tumor they are treating.While cancer is primarily a disease of the genes and result from acquired mutations within somatic cells, there are additional layers of complexity involved in actual cancer. We have to consider many host factors affecting the development and propagation of cancer (like familial predisposition, immunologic factors and metabolic factors). Add to this fact that cancer cells are intimately linked to the host in multiple ways, briefly as shown in the diagram below; unlike in experimental models, cancer in vivo has complex interactions between various host factors as represented below:

The initiating event in CML is acquisition of the t(9;22)(q34;q11) translocation, which creates a fusion between the BCR and ABL1 genes. Secondary genetic alterations, such as mutations of TP53, RB1, and CDKN2A, can be acquired after the BCR-ABL translocation and may play a role in progression of CML from an early chronic phase to a more aggressive blast phase.

We also have to consider that not all genetic changes will lead to protein production, so will need to interrogate cancer cells at not only genomic level but also at the proteomic levels. But once we start looking at cancer in such a comprehensive way, oncology care wil change from one that relies primarily on trial-and-error treatment strategies based on the anatomy of the tumor to one that is more precisely based on the tumor’s molecular, proteomic, metabolic level and appropriate for the particular host, thus enabling many cancers to be turned into manageable chronic disease, and providing patients with long-term high quality of life.

Dr. Patrick Soon-Shiong, MD is one of the leaders of this new push for a paradigm shift in how to address cancer given all these current understanding in to the biology of cancer. He is the inventor of nab-paclitaxel (Abraxane), the first U.S. Food and Drug Administration (FDA)-approved nanotechnology-based chemotherapeutic agent. He is also the Founder and CEO of NantWorks and its subsidiary NantHealth, a cloud-based biomolecular medicine and bioinformatics company that uses high-frequency, high-throughput tumor genome sequencing to analyze the DNA, RNA, and protein levels of an individual patient’s cancer cells. Proud to say the our practice is one of the select few from the US South Central region who is collaborating with Dr. Soon-Shiong’s company in developing a global consortium of cancer providers and patients to advance this field of precision medicine.

GPS for the Oncologist:

How do you navigate this “maze” of molecular genetics, protoemics and other data to come up with the best suited treatment for a particular patient? I think Dr. Soon-Shiong is a true visionary in this regard, as his aptly named GPS (genome/proteome sequencing) involves a next-generation sequencing technology to analyze genomic (DNA) and transcriptomic (RNA) sequencing data. This identify variants between somatic and germ line DNA. What was interesting, though not unexpected was that, after analyzing a cohort of 3,784 patients (on 19 anatomic tumor types), it was found that genetic mutations in gene panels do not always result in protein expression. Thus even the informed clinician who is up to date with the latest molecular genetic data could miss or over read these reports – as what truly matters is downstream protein expression and not just DNA alterations alone. And this signatures change with time. Luckily however this entire GPS Cancer can be run now on circulating tumor cells, avoiding the need for repeat biopsies, capturing cancer as it evolves, rather than considering cancer as a static process as was the orthodox - and flawed - view in the past.

Advancing the Next Paradigm of Cancer Care

As detailed above, cancer is no longer considered as a single clonal disease nor static in it's genetic make up. Cancer cells also show enormous inter- and intra patient tumor heterogeneity and cancer progression is driven not just by one genetic mutation, but in many instances driven by tens and even hundreds and perhaps thousands of mutations, rearrangements, and structural changes in the genome, dynamically changing across time and space.

What is the downside of this approach of interrogating cancer at a much deeper level? Basically from being a “common disease” cancer becomes a set of rare diseases. Each patient’s mutational, proteomic, immunologic signature is going to be unique, and hence treatment should be tailored to that individual patient. Even large institutions will not have enough patients of a “specific GPS signatures” so Dr.  Dr. Patrick Soon-Shiong is spearheading a large collaborative “omics” network—a muscularly sophisticated network of oncologists to share outcome data and create an “adaptive learning system.”  This will require an infrastructure for sharing of outcomes in real time as well as an infrastructure to receive an in-depth whole-genome, RNA, and proteome sequence analysis in a timely manner to take advantage of real-time knowledge that may better inform a clinical decision.

Cancer care in the not distant future:

How do we effectively attack this multi clonal disease that changes its gene expression over time and space? The best way is to explore ubiquitous pathways driving proliferation and metabolism of the cancer cell, to attack both the stem and metastatic cancer cells, to recognize that the biology and evolution of these two cell types differ, and to use multiple drugs focused on multiple points of attack, targeting the cell’s nucleus, DNA, cell signal pathways, and metabolism and micro environment all simultaneously while also enhancing host immune system. The field of immunotherapy for cancer is worth it’s own blog, but briefly this has been one of the most exciting areas in all of oncology in the year 2015, with several new medications approved across a variety of cancer types where the treatment work by stimulating patient’s own immune cells or by enhancing the anti tumor immunogenecity.

Is there going to be a “magic pill” for cancer:

If you are still asking this question after reading the above blog, I would strongly recommend reading it again! Cancer is an extremely heterogeneous disease, so unlikely we will have a “magic pill” that will work across cancer types. However by “intelligently and selectively” blocking the various metabolic and survival pathways of cancer cells, by going after the cancer stem cells as well as the metastatic cells, and by enhancing the patient’s immune system we have a chance of changing the paradigm of how we manage cancer patients. Eventually, by more deeply understanding the biology of the cancer stem cell, we will provide long-lasting remission and get closer to a functional cure for cancer.

1 comment:

  1. Dear Doctor,

    Is there anyway I could get your email Id. I wanted to discuss something details about Carcinoma Nasopharynx. I would appreciate if you could email me at

    Hope to hear from you soon Dr.