Siddhartha Mukherjee Offers Insight on the Future of MPN Treatment
Oncologist, researcher and Pulitzer Prize-winning author Siddhartha Mukherjee, M.D., Ph.D, talked with CURE on the future of treating MPNs.
BY Allie Casey
PUBLISHED December 17, 2016
Winning the 2011 Pulitzer Prize for General Non-Fiction and named one of the 100 most influential books written in English since 1923 by Time Magazine, The Emperor of All Maladies: A Biography of Cancer is certainly an impressive book. However, its author is equally as impressive: Siddhartha Mukherjee, M.D., Ph.D. is not only a Pulitzer Prize winner, but also an oncologist and researcher He also served as the guest speaker at the award ceremony for the 2016 CURE Myeloproliferative Neoplasms (MPN) Heroes.
While there, CURE spent some time with Mukherjee to talk about MPN awareness, his newest book, The Gene: An Intimate History and the important role that “small data” in anti-cancer therapies.
There's some difficulty with diagnosing MPNs, as well as a lack of information with physicians. Do you feel that awareness has gotten better?
MPNs have been a somewhat mysterious disorder for a very long time. And for a while, it was so poorly understood that physicians had a hard time understanding what the diagnosis was, how to diagnosis it appropriately, what the subtype was and how to appropriately characterize that subtype.
Now, as the awareness has grown, primarily driven by patients and patient advocates, I think there's a growing awareness of what this family of disorders is, how it manifests and, perhaps most importantly, what the molecular lesions are, and what the genetics is for this family of diseases. I think all of this together – increased advocacy and better understanding of the molecular mechanism, and increasing physician awareness – has driven a change in how people understand and diagnose these diseases.
What are some of the challenges patients face when they're diagnosed with an MPN?
It's a relatively rare illness and often there's a lot of misinformation around this family of diseases. One of the challenges is to get the right information, to make sure that you connect with a physician who actually understands and knows how to treat this series of disorders, so you get the best and most optimal care.
There have recently been NCCN guidelines established for MPNs. Has that made a big difference in the community?
Absolutely. The presence of powerful, comprehensive guidelines really helps because rather than having ad-hoc ways of treating and understanding MPNs, there's a more systematic way to understand, classify and treat MPNs. That's helped a lot.
We need a lot more in these guidelines, but it's an evolving field. The molecular characterization of MPNs is constantly evolving, and we're still trying to understand what the various subtypes are and what the fundamental pathogenesis of these disorders is. That's going to change.
But absolutely, yes, having guidelines helps a lot. It standardizes the procedures for treatment.
Why did you set out to take on your new book, The Gene?
This new book really is a sort of prequel to the previous book on cancer, The Emperor of All Maladies. The Gene is a prequel partly because in the new book, I address the question of what it is that makes our cells function in normalcy, how do cells function, what is the code of codes that allows cells to work, our bodies to work and how. Can we understand that code and potentially manipulate that code to help diseases?
These two books are closely linked. One of the ways I was trying to understand them is that the deciphering of the genetic code had a powerful impact on our understanding of cancer, so I wanted to go backward and ask the question: What do we do? How do we know understand that code as we move forward into the future?
How far away do you think we are from reprogramming our own genetic code in order to side-step serious illnesses?
There are a host of new technologies available driven by a particular form of technology called CRISPR and Cas9, which allow us to make extremely precise, and with great fidelity, direct changes in the human genome. We don't know the exact level of precision; we don't know how to actually potentially use this in human embryonic cells, partly because there are strong limits and guidelines placed about the use of these technologies in humans, particularly when you start manipulating the human genome in a way that the change becomes permanent. And by permanent, I mean it becomes transmitted across multiple generations.
The analogy that I like to give is that you can think of the human genome as an encyclopedia. In fact, if it were actually printed like an encyclopedia, it would be about 70 full sets of the Encyclopedia Britannica. In principle what these genome-altering technologies allow us to do is to take one volume in one of these many sets and alter one word in that volume without touching anything else. We don't know if that is actually true, in human embryonic cells, we don't know what the potential is – are other words changed? We don't know what the limits of the technology are and can be. Obviously it's an evolving field and we're trying to find out how quickly it's going to change and is changing.
Do you really see Big Data as the boom for anti-cancer therapy?
I just want to clarify what Big Data is. Big Data just really means, as I understand it, very deep annotation of patients, their individual cancers, the genetic components of their cancers and how these cancers are behaving over time. The power of all of this, because of the nature of the genetic and other diversities in cancers, is to be able to look in the real world across multiple cancers to understand their biology, their pathology, what happens to patients, etc.
I think it's absolutely been transformative. We've found profound new insights about how cancers behave across large populations, profound new insights about how people respond to different medicines. Just to give you one example about which patients respond to one particular medicine, let's say Herceptin ( trastuzumab). And who doesn't respond, how do you find that population that doesn't respond, what is unique about that population? What differentiates that population from the population that does respond? How do you then make second-generation drugs to now make women who are not responding to the first generation of medicines, now respond to a second generation, etc, so these are very powerful things.
I also want to add that it's important to remember that data is data, as long as it's powerful, as long as it's deeply thought about. Big Data is not magic. It's just the accumulation of large amounts of very deeply annotated, typically human specimens which allow us to make these profound insights. I want to make a particular plug that while we do all of this, it's important to remember that small data is also very, very crucial. Many of the most revolutionary treatments came out of lab experiments, not involving tens of thousands of samples, but involving one sample, understanding one piece of evidence. While we understand the relevance and importance of Big Data, it's equally important to understand the relevance of small data, of the sort of foundational experiments. And not to be over-seduced by the value that Big Data is adding.
What are your thoughts on the Cancer Moonshot Initiative in terms of data-sharing and transparency, for example?
There's no doubt about the benefits of having engagement of larger groups of people with each other, scientists, physicians, physician-scientists, patients, advocates, people who work on information, the data-gatherers, people who crunch information, the data-miners, clinical trialists. Interacting together in a more transparent way clearly has a huge advantage and hopefully will be actualized in the next decade or so. Through the Moonshot project, but through other projects as well. I think there's no doubt that will be helpful but it's important to know that almost always these pieces lie atop very basic investigations into cell biology, cancer biology, genetics, etc. It's important not to forget that.
Just to give you an example, Gleevec (imatinib) – ultimately we understand it in a much deeper way today than we did 10-odd years ago, but it really grew out of very simple experiments, understanding what the particular lesion was in CML, particularly a genetic lesion in CML, understanding how that was potentially drugable. Very simple experiments, laboratory experiments, that demonstrated that that drug particularly would target the cells that carried the Bcr-Abl fusion and spare other cells. At that stage it wasn't driven by Big Data. Now we have much, much more data about it and it has helped us to understand things like who becomes resistant to Gleevec, but it was built atop a series of extremely small, boutique experiments you might say. It's important to combine those two things.
What are the benefits of a cancer registry for patients?
The importance of a registry is that you can collect, in one central bank, patient samples and potentially the data of a patient so that instead of lying in individual silos in separate laboratories as it were, it's all centralized and put into one central bank and everything that's assessed can be assessed from that central place. It's just the power of numbers. The more we learn about a particular disease, polycythemia vera, essential thrombocythaemia, the more we understand about that disease and just by power of numbers, the more we annotate around a registry, the more we understand. As opposed to sitting in individual silos, where there may be less of deeper understanding because you don't have enough samples.
Particularly with cancer now, it turns out we know now that even within one cancer there are subtypes and subtypes and subtypes, depending on the genetic diversity of cancer. No single site or individual can collect the breadth and depth that's required to really understand cancer or one particular type of cancer in molecular or genetic level. This is an incredibly important resource from registries.
What are some issues about when immunotherapy can be stopped?
It’s one of things that, as a community of scientists, we're trying to figure out. When to start, when to stop, who qualifies, who doesn't – by qualifies I mean who's likely to get the benefit. What subtypes of cancer are likely to get the benefit. In blood cancers, interestingly, we're beginning to understand how to deploy some immunotherapies. Others have actually started off in blood cancers.
These are such novel drugs, that probably five to 10 years will pass by to figure out how to stop. Just to give you some surprising phenomena, in some solid tumors, when immunotherapy is started, as the immune cells infiltrate into the solid tumor, you can actually get an expansion of the tumor. In fact, if you were using old criteria of trying to understand the progression of a tumor, you may say that the tumors were getting worse before they got better because the size expanded. I actually saw a case quite recently of a patient with metastatic melanoma who started immunotherapy and the first thing that happened is the tumor grew harder and bigger, palpably harder and bigger. In the old days, 10 years ago, we might have stopped therapy, saying this is making things worse.
The point here is that the very paradigms by which we judge responsiveness, or the decision to continue or discontinue therapy, have been overturned, or are being overturned and re-examined in the context of immunotherapy. You can't use what we had learned from cell-killing drugs in the traditional mechanisms of treatment. How to stop, when to stop, is an open question. I think we'll solve it, but it will require some kind of rethinking, in the same way, what do we think when and how to judge whether a patient is responding or not responding.
When patients are on therapies for a long time, it can often become quite expensive. Do you have any comments on the issues of financial toxicity?
Yes, the financial toxicity of many of these therapies is enormous. I think eventually we have to find a way, and this is so obvious to everyone, to deliver drugs that have powerful therapeutic benefits, at a cost which is reasonable for patients. Otherwise, we'll break not just an individual patient's resources, we'll break national resources.
Finding that sweet spot, where you have the appropriate incentive to develop these medicines, but also you have the capacity to deliver them to the patients who need them and in a way that doesn't cost them financial toxicity for the rest of their life, is very important. Also there are new mechanisms to try to figure out what the worth of a drug is, how much life extension, how much preservation of dignity, how much preservation of function, can a drug give. I think these efforts are thoughtful, they'll allow us to pick out medicines in sub-populations that are appropriately being useful.
I do think that science will help here. The science of cancer might be the lynchpin solution. Because if we can prevent giving a medicine to someone who is unlikely to respond in the first place, or stop giving the medicine to someone who is not responding based on biomarkers, based on deep molecular understanding of the cancer, then all of a sudden, you start getting more bang for the buck, as it were. So you're really treating patients who are responding, stopping it when patients don't respond so that you are appropriately matching therapy to the patient.
Just to give you a simple example, if you give tamoxifen to a patient with breast cancer who doesn’t have the over-expressions of genes that sensitizes them to tamoxifen, then you'd be wasting that amount of money. That's just one example of making sure you give the right medicine to the right people. And hopefully that will shift the balance toward appropriate benefit without the excessive financial toxicity.