The following is an interview recently conducted by the PMWC Intl. team with Dr. Roger Perlmutter (President, Merck Research Laboratories). The discussion with Dr. Perlmutter focused on various aspects of personalized medicine ranging from key promises to current and future challenges: questions aimed at addressing how personalized medicine will influence the drug discovery process, how the pharmaceutical industry is adjusting and evolving over the next 5-10 years, what big hurdles there are to make personalized medicine a true reality, and – last but not least – how the ongoing pressure on cost control in healthcare vis-à-vis the continuing increase in drug development costs will ultimately shape the fate of personalized medicine and its acceptance by payors, providers, and patients.
Dr. Perlmutter is one of four individuals that will be honored at the upcoming PMWC 2016 Silicon Valley conference for his achievements in transforming two R&D organizations while developing groundbreaking new therapies. Besides Dr. Perlmutter, Dr. Laura Esserman (UCSF), Dr. Ralph Snyderman (Duke University of Medicine), and Dr. Irv Weissman (Stanford School of Medicine) will also be honored with an award at this very same conference. We at enlightenbio are particularly excited that all four honorees will enrich the conference not only with their presence but also by sharing their vision and insights in respective presentations throughout the conference.
Following is the interview conducted by PMWC Intl. with Dr. Roger Perlmutter.
What is your view on how personalized medicine will impact healthcare? What are some of the key promises and challenges we will or may encounter along the way?
Dr. Roger Perlmutter: Personalized medicine is already affecting healthcare practice in a very profound way. Physicians have practiced personalized medicine for as long as medicine has been practiced. None of us who actually engage in patient care put a great space between us and the individual circumstances of a patient – we don’t treat people according to a formulated cookbook. We do ask questions about the individual circumstances of the patient. What is changing is the fact that we now have very incisive tools that affect how we think about managing medical problems in individual patients. One of the most important examples, of course, is the availability of genetic data beyond the family history – actual gene sequence data which is relevant in many fields, but especially relevant in oncology. Over time, the availability of these kinds of information, like complete genome sequencing data from every individual (patient) as well as a variety of other measurements will affect every aspect of medical practice.
We originally developed a view that cancer was a cell-autonomous process, and the approach to treating cancer was to introduce drugs that affect a specific pathway that had been inappropriately activated or silenced in the cell that had become malignant. Over the last few years we have come to appreciate that tumorigenesis is quite a bit more subtle: we have learned the underlying process that includes not only the cell that has sustained mutations and that remains “malignant”, but in addition an immune response that – at least for a while – prevents that cell from expanding unreasonably. We have now the ability to influence different parts of this process, beyond just trying to identify a drug that targets a particular signaling pathway. We are of course still learning as we go, but that just reinforced the fundamental point: “as more information becomes available, it enables us to select an ideal therapy to balance benefits and risks”. More precisely, we will see patients treated more individually which is of course the essence of personalized medicine. That is exactly where oncology has been going over the last couple of decades.
How has personalized medicine already influenced the drug discovery process and how is the pharmaceutical industry adjusting? How do you see the field evolving over the next 5-10 years?
Dr. Roger Perlmutter: Personalized Medicine has been part of the drug discovery process for a very long time. Twenty years ago we thought about it mainly in the context of phamarcogenomics. Some individuals, compared to others, do not respond terribly well to a drug treatment, and this is often due to the fact that there are differences in the way a drug is metabolized. Heterogeneity in drug exposure among different patients has an effect on efficacy, but can also result in idiosyncratic adverse effects. The underlying basis could either be genetic or might reflect another aspect of physiology, or some part of pharmacotherapy. For example, people who take multiple drugs at the same time or add new drugs to existing drugs, have to be concerned about drug-drug interactions, as one of those drugs could affect the metabolism of the newly added drug or vice versa. These kinds of considerations have always been part of the thinking and decision-making process in drug discovery. A pharmaceutical chemist embarking upon selecting a potential new drug will always consider whether the drug candidates they have in front of them are susceptible to metabolic pathways that vary among individuals.
We already talked about the situation in oncology where drug discovery is already exquisitely focused on individual differences at the genetic level. It is not just a question of what the primary mutation might be in the signaling pathway, let’s say with respect to leukemia the fusion protein BCR-ABL, but it is also a question of the actual substitutions that occur in ABL that make the fusion protein then resistant to first-line kinase inhibitors like Gleevec. There has been an evolving process of looking for better and more active compounds that can be used in individuals with specific mutations. All of that has been going on for quite a while.
The industry has adjusted by adding more technical evaluations of possible therapeutic interventions, and they have done that fairly successfully. The industry had more difficulty in seeing the population benefits advocated by those who are devoted to personalized medicine. It has materialized in a few cases, but it is still more the exception than it is the rule. For example, the idea that one could do shorter and smaller clinical trials to gain registration less expensively was an early promise of personalized medicine, but these benefits remain elusive. The industry is clearly adjusting, but at the same time is wary, for the most general reason – and perhaps the most important reason – because it is an industry. If medicine becomes too personalized, when we pick a therapy specifically for the person in question, then it is a process that cannot easily be industrialized. It’s simply impossible, using current technologies, to address each patient with a new molecular entity based on the precise characteristics of that individual. Moreover, the regulatory pathway that would permit such an approach remains undefined.
For these reasons, the industry will adjust to a certain point, by taking advantage of advances in the understanding of the heterogeneity of a population as a way of selecting better drugs, but can push this approach only so far. Beyond a certain scale personalized medicine can’t be industrialized.
What are some of the biggest hurdles it takes to make personalized medicine a true reality for the majority of the population?
Dr. Roger Perlmutter: Well, the first question we have to ask is: “should personalized medicine become a reality for the majority of the population?” I believe we would all be perfectly comfortable if the development of new therapies took a non-personalized approach, provided that the resulting drugs were broadly effective and safe. Let’s take an example: we have extremely good drugs for lowering low-density lipoprotein-associated cholesterol. We know when we do that, we reduce cardiovascular risk and the benefit/risk proposition associated with those drugs – statins, the HMG coA-reductase inhibitors, is very favorable. Their adverse effects are very modest, when they occur they are idiosyncratic, and the benefits from therapy are enormous in terms of reducing cardiovascular risk in people who carry such risk. There is no need to build individual statins for each individual. Furthermore, it is fair to say that most people are healthy most of the time, and so in a sense, the right prescription for a lot of people will be a healthy diet, appropriate exercise, and the avoidance of adverse behaviors.
On the other hand when people are truly sick, it is important that the entire context of the illness is understood. The illness inevitably has a component of genetic susceptibility, but environmental factors also play a very big role. For this reason, for the majority of the population we have to develop drugs that have very big therapeutic indices, with very few adverse effects, and a large therapeutic impact.
How do you see the ongoing pressure on cost control in healthcare vis-à-vis the continuing increase in drug development costs shape the ultimate fate of personalized medicine and its acceptance by payors, providers, and patients?
Dr. Roger Perlmutter: Cost control increases pressure on our industry, which really just makes us better. The impact is clear: we must produce better medicines that have a meaningful impact – not on the signs and symptoms of the disease, but on the evolution of the disease process. I believe that society at large will be prepared to pay for the cost of great medicines, but society is not prepared to pay for medicines that provide little therapeutic value. We are all healthcare consumers and hence we all agree that healthcare dollars need to be spent wisely.
Cost pressure is therefore a constructive force, even though it will be painful in the near term. Companies that are especially inefficient and that do not provide real solutions to health problems, will be gradually eliminated from the commercial environment. The challenge will be to ensure that we offer adequate incentives to permit some biopharmaceutical companies – the best ones – to prosper.