Oncology R&D Challenges & Opportunities

By Cliff Mintz Ph.D., Life Science Leader magazine

Millennium Pharmaceuticals, Inc. is one of only a handful of biotechnology companies that was built almost exclusively on a genomics and bioinformatics technology platform. Despite modest beginnings in 1993, the company has transformed itself from a fledgling genomics startup into a fully integrated biopharmaceutical company. Its flagship product is VELCADE (bortezomib) for Injection, an FDA-approved treatment for multiple myeloma and relapsed mantle cell lymphoma. In May 2008, the company was acquired by Takeda Pharmaceutical Company, Japan’s largest pharmaceutical company, and currently operates as an independent subsidiary known as Millennium: The Takeda Oncology Company.

One of the scientists and executives who helped transform Millennium into a world-class oncology company is Nancy Simonian, M.D. Dr. Simonian is currently Millennium’s chief medical officer and has responsibility for the company’s clinical development programs, pharmacovigilance, regulatory affairs, and development project management. Prior to joining Millennium in 2001, Simonian was vice president of clinical research at Biogen, where she was responsible for the clinical development of AVONEX (Interferon beta-1a), Tysabri (natalizumab), and the company’s oncology programs. She earned a bachelor’s degree in biology from Princeton University and received her M.D. from the University of Pennsylvania, School of Medicine. She completed her medical residency at Massachusetts General Hospital (MGH) in neurology and was an assistant professor at Harvard Medical School and Massachusetts General Hospital before leaving academia for industry.

While much progress has been made in the treatment of cancer, it still remains a therapeutic area with immense unmet medical needs. Millennium’s expertise in genomics and bioinformatics, coupled with its focus in oncology, makes it a company to watch as we enter the world of personalized medicine to treat cancer. I was able to pry Simonian away from her busy schedule to discuss Millennium’s approach to drug development, the growing importance of personalized medicine in developing cancer-fighting drugs, and some of the hurdles that biopharmaceutical companies must overcome to garner new drug approval in an increasingly stringent regulatory environment.

What are some of the main challenges faced by pharmaceutical and biotechnology companies that are developing new cancer drugs and treatments?

In the not-so-distant past, cancer represented an enormous unmet medical need, mostly because there weren’t many effective drugs or treatment options for various cancers. Because of this, pharmaceutical and biotechnology companies started to invest heavily in research and development in oncology. Not surprisingly, there are a lot more oncology drugs available today than there were 15 years ago. And, in some areas, there are some pretty good drugs, and indeed some breakthrough drugs — which raise the hurdle for development of the next generation of cancer treatments. For example, 15 years ago the median survival for patients with metastatic colorectal cancer was six months. Today, with newer therapies, median survival times have improved to greater than 25 months. While this is great for patients with colorectal cancer, improved median survival times force companies to increase the number of patients and the follow-up time required to conduct human clinical trials to develop new colorectal cancer drugs. This raises development costs quite substantially. In the past, cancer drugs were approved after hundreds of patients were treated in clinical trials. Now, it is much more common to see thousands of patients in the initial regulatory filing. Ironically, in some ways, we are victims of the successes of new cancer treatments because we have raised the bar on what is deemed an acceptable and approvable new oncology agent.

Another consequence of our success is managing regulatory expectations and requirements for approval of new oncology drugs. Overall survival remains the gold standard clinical benefit endpoint for most cancers. Increasingly, regulatory agencies in the United States and Europe are focusing on the magnitude of clinical benefit, weighted against risks as well as other therapeutic options a patient may have. Statistical significance is not a guarantee for approval, and approvals based on Phase II studies are more challenging now than in the past. Paradigm-changing drugs like Gleevec, with cure rates approaching 95% in patients with chronic myeloid leukemia, tend to heighten regulatory expectations. The increased requirements to demonstrate clinically meaningful benefit and a well-characterized safety profile have substantially increased the cost and risks associated with developing new oncology drugs.

One of the major research challenges in developing new cancer drugs is a lack of validated, predictive, animal models of diseases for certain cancers. While drug candidates may work well in mice, they don’t necessarily work great in people. In other words, it has been extremely difficult to predict whether or not something that works well in the lab will translate into a success in humans. This has resulted in a very high clinical attrition rate for new oncology agents — in the 90% range — from clinical testing to approval. We believe greater information on how our drugs are working in the early stages of clinical testing will help us decrease late-stage attrition. We focus on proof of mechanism and proof of activity biomarkers and use that data in our decision criteria for advancement and investment. Companies should use these predictive tools and any other available means to determine sooner rather than later the likelihood of success for a particular drug candidate.

Given the inherent complexities and risks associated with developing new oncology drugs, what distinguishes Millennium’s approach to drug development as compared with its competitors?

The company was founded as a genomics company and, not surprisingly, we leverage our expertise in genomics and molecular biology to understand underlying genetic and biological mechanisms to identify possible drug targets for certain cancers. We focus on biologic pathways that have multiple potential cancer targets, and we have a critical mass of scientists with deep expertise in these pathways. We utilize biologic assays in our preclinical models to establish pharmacokinetic, pharmacodynamic, and efficacy relationships, with a focus on utilizing similar assays in the clinic. We also have a state-of-the-art imaging center and are increasingly utilizing primary tumor in addition to our standard xenograft models, which we hope will better predict results in the clinic.

In the clinic, we rely on biomarkers and surrogate clinical markers to help improve our predictive power for drug efficacy. These biomarkers tell us whether we are inhibiting our target and impacting downstream cellular pathways. They help us in determining the optimal dose and schedule for Phase II. These approaches help to mitigate some of the inherent risks associated with oncology drug development. We also seek to identify patient subsets most likely to respond to our drugs. We start this work preclinically, and we believe a better understanding of which tumor types and subsets will be most sensitive to our drugs will mitigate risk and also potentially reduce cost. While Millennium isn’t the only company that uses these approaches, they were critical in developing and bringing VELCADE to market.

Lastly, we work in a highly integrated fashion across research, development, commercial, and business development. Biology, new clinical data, and the commercial landscape rapidly change in oncology. I believe a key to success in oncology drug development is to have a highly experienced and skilled team that is focused on the science that will drive the business.

What role do you think personalized medicine will play in future oncology drug development?

I think personalized medicine will play a very critical role in the development of new oncology drugs. That is, trying to figure out which patient populations are most likely to respond to your drug. This will help improve R&D productivity and effectiveness. Personalized medicine will aid in the development of drugs that have improved benefit-to-risk ratio, and this, in turn, will be important for commanding premium pricing. In other words, identifying the optimum patient population that is most likely to respond to a new drug candidate may lead to improved clinical success rates and ultimately better patient outcomes. For example, in the past six months or so, two companies developing drugs (one to treat lung cancer and the other melanoma) were able to identify patients with particular tumor genetic alterations who were high responders to their drug candidates. This understanding early in development is allowing these companies to go directly from Phase I into Phase III clinical studies (within the selected patient populations) with the drug candidates. I believe the more we can do this, the better it will be for cancer patients, because it will allow us to develop drugs more quickly and efficiently.

At Millennium, we recently initiated a Phase II clinical study in a subset of lymphoma patients who are genetically defined based on gene expression profiles. This subset of lymphoma patients was identified by academic researchers, and preclinical and clinical data pointed to a role for VELCADE. In this particular Phase II study, we are only enrolling patients with that genetic profile because we believe they are the ones most likely to benefit from our experimental treatment. Because of Millennium’s heritage and expertise in genomics and bioinformatics, every molecule that we move from discovery into development must have a biomarker strategy and plan associated with it. We fundamentally believe that this strategy will help us overcome many of the challenges that we routinely face in new oncology drug discovery and development.

Do you think that an increased reliance on personalized medicine and genetic profiling might hinder, rather than promote, drug discovery on genetically ill-defined but serious types of cancer?

I firmly believe understanding the genetic basis of tumors will help us in the development of safer and more targeted new drug therapies. But, I don’t think that this is the full answer for new oncology drug discovery.

For example, the discovery platform at Millennium is focused on what we call “targeting the infrastructure of the cancer cells,” that is, identifying cellular pathways that are selectively upregulated in cancer cells due to stress. We think this approach can work in a wide variety of cancer settings and across different genetic backgrounds — since the stress response occurs through common and conserved pathways. I believe the understanding of the genetic basis of tumors, the pathways involved, in addition to the response of the cancer cell to stress, all will likely lead to a greater understanding of the common cellular pathways that drive cancer and ultimately result in the development of better, safer, and more targeted treatments for multiple oncology indications.

What are your thoughts on the use of biomarkers rather than traditional endpoints like disease progression or median survival time in oncology clinical trials?

Regulators will require strong clinical data sets that support the relationship between the biomarker and a clinical outcome before allowing the marker to be substituted for medically proven clinical endpoints in pivotal clinical trials. I want to point out that in oncology there are already accepted clinical surrogate markers that may be substituted for clinical benefit outcome like progression-free survival or response rates. Having said that, there must be strong clinical data linking a new biomarker back to disease progression, survival, or long-term patient outcomes before it can be substituted for those endpoints. In nonpivotal trials, utilizing imaging and other biomarkers for decision making is acceptable. Even for internal decision making, the key is still to know how a change in a given market is likely to correlate with clinical benefit.

You previously mentioned that improvements in most cancer treatments are incremental. How much incremental change do you think should be required to win regulatory approval for a new oncology drug?

It is hard to directly answer that question because the answer is very dependent on the type of cancer that we are trying to treat. There is probably a range, say anywhere from 20% to 30% up to 50%, of improvement over the standard of care for approval by regulatory agencies. Most cancer drugs that have been approved are in that range. But, you have to remember that the observed overall improvements may not apply to all patients. In many cases, the drug may benefit some patients more than others. That is why I think identifying patients who we think will benefit most from a new drug is vitally important to the success of future oncology drug development.

Do you think that industry’s approach to developing new drugs to treat cancer is changing?

Since most new cancer drugs are targeted to only one or a few pathways, and because many cancer cells have abnormalities in multiple pathways, treating patients with combinations of the newer drugs (or adding one or more of them to older treatment regimens) will likely offer better long-term survival rates and improved clinical outcomes for cancer patients. We have certainly seen the value of combinations with VELCADE. So, we think we are moving in the right direction.

As our understanding of the various pathways that appear to be driving cancer continues to improve, and the number of approved, novel cancer drugs increases, we are beginning to see more rational combinations of these drugs to determine whether or not we can obtain better patient outcomes. We and other companies are exploring this approach in a major way, even to the point of thinking about evaluating two unapproved treatments if we think the combination may improve patient outcomes. To that end, Millennium might consider entering into partnerships or alliances with other companies to test the combination treatment concept. The notion that a single company can do it on its own may no longer be viable. In the future, you will likely see more companies collaborating and working together to develop safer and better cancer treatments.

There is a growing perception among industry executives that the regulatory environment for approval of new molecular entities is becoming increasingly stringent and more difficult to navigate. What are your thoughts?

I think the FDA has done an excellent job in weighing the benefits and risks of new oncology agents before rendering decisions on their approvability. Overall, at the regulatory agencies there is an increasing emphasis on safety and risk management. I think the uncertainties around regulatory agency positions and requirements surrounding safety and risk are what many companies find most vexing and challenging these days. Companies are now factoring increased requirements for safety monitoring and studies into their portfolio decisions. The reality is that all of the drugs we develop have some risk associated with them, and some of these risks will not be identified until many thousands of patients have been exposed. To be successful, companies must develop better ways to monitor and manage these risks both during development as well as postapproval. We need to enhance the ways in which we quantify and communicate risks to physicians who prescribe the drugs and to the patients who use them.

What therapeutic areas or unmet medical needs deserve more attention?

According to the World Health Organization, cancer is going to surpass cardiovascular disease as the leading cause of death by this year, and we still don’t have cures for most types of cancer. Rates of cancer occurrence will continue to grow and remain a high unmet medical need in the future.  

Will the life sciences industry of the future look much different than it does today?

I think we’re going to see a very different life sciences industry in the future. I believe there will be more collaboration between pharma companies and between pharma and biotech and academia. I think we are at a crossroads in the realization that no one company has all the infrastructure, technology, and biology platforms to discover and develop the next generation of drugs. The future will be in how companies can most effectively collaborate to drive value for their businesses.