A major need and critical collateral of what we are talking about here is that we all need to aim for big improvements in outcome. Incremental improvements are no longer good enough.
—Manfred Lehnert, MD
While the last 15 years have brought unprecedented advances in oncology drug development, the next 10 years promise to usher in even greater opportunities to realize the goal of precision medicine in the treatment of cancer, providing patients with more effective care and better outcomes. Reaching that goal will also entail overcoming unprecedented challenges, according to Manfred Lehnert, MD, Group Head, Oncology Clinical Research, Takeda Pharmaceuticals International, Inc.
The ASCO Post talked with Dr. Lehnert about those challenges and how they can be overcome in the new era of cancer care.
What will be the paradigm for oncology drug development in the future?
In the future—and that future is well underway—cancer will be predominantly classified by its genetic and molecular characteristics. As a result, there will be many hundreds if not thousands of distinct malignant entities, and common cancers, such as lung and breast, will largely cease to exist. This means that cancer will become a group of multiple orphan diseases, and this prospect is posing unprecedented challenges for the oncology community.
My view is that collaboration among all the stakeholders, including research foundations, academic institutions, regulatory agencies, payers, patients, patient advocacy groups, physicians, and the biopharmaceutical industry, is vital for making the major progress we require and expect in the treatment of cancer.
Takeda’s strong commitment to external collaboration includes models for more efficient ways to conduct clinical studies and develop drugs in the new era of precision medicine in oncology. One example of that is the MATCH study, which the National Cancer Institute is launching. The study will screen about 3,000 patients with advanced cancers to find approximately 1,000 cancers with genetic mutations. The ultimate goal will be to determine which new therapies, including ones not yet approved for use, are made available by the pharmaceutical industry through collaborative arrangements.
Another example is the WIN Consortium and the WINTHER project, through which an international consortium of members will use an algorithm-based bioinformatics tool to examine DNA and RNA data from dual biopsies of tumors and normal tissues and then provide relevant drug-target gene matches to improve treatments for patients in the trial. Drugs in the early stages of development will also be further investigated by this approach.
These are two examples of what I believe is a new paradigm for drug development in oncology. Enterprises such as these highlight a critical need for close cooperation among academia, nonprofit organizations, and across industry.
Intellectual Property Issues
In these types of collaborative efforts among pharmaceutical companies, who owns the patents on new medicines?
The fundamental thinking and attitude needs to be that resolving the issue of intellectual property is a must in order to make major progress fast and to enable efficient oncology drug development. But intellectual property interests are not limited to the pharmaceutical industry. This issue also comes up in partnerships with major academic institutions, and traditionally it requires case-by-case discussions and negotiations for resolution. Ultimately, there needs to be a master framework to resolve this issue, and there are positive early examples of efforts being made along those lines.
Another important challenge to overcome is the redundancy that exists in the system when it comes to oncology drug development. For many actionable molecular targets, a half-dozen compounds may be in clinical development—compounds that often have only marginal differences. This redundancy is a waste of resources and slows down progress.
New Research Strategies
How will drugs be investigated and tested for the specific molecular characteristics of each type of cancer in individual patients?
This is one of our major new challenges. We have categorized cancer into about 200 distinct malignant diseases, but in the future, that number will be in excess of a thousand, which will impact many things including research and clinical care.
For example, how will randomized phase III clinical trials be executed? How much time will it take to complete such studies? How will we identify small subsets of patients with the molecular aberration targeted by a drug, and how will those patients be able to access appropriate trials? This is a major logistical effort, costly to do, and, importantly, it means that a majority of patients who have undergone testing will be disappointed because they don’t carry that mutation and do not qualify for the study.
A major need and critical collateral of what we are talking about here is that we all need to aim for big improvements in outcome. Incremental improvements are no longer good enough. Big improvements are what cancer patients need and what they are waiting for.
At the same time, these new approaches will allow the conduct of smaller randomized studies. In certain situations, randomized studies may not be required at all, because the response rate and response durability will be so much better than historic data.
As we get smarter at matching the right drug to the right molecular characteristics identified in the right patients, these needs and goals will be realized.
Drug Development Costs
Will the process of individualizing treatment for patients add to the cost of drug development and approval?
That is difficult to predict; it can go either way. A large part of the high development cost of drug research is driven by failure in phase III trials, because this is where most of the money gets spent. So we need to make sure that we dramatically increase the probability of success when a drug enters phase III trials, and I believe that the new science of oncology will enable us to do that.
We need to perform rich scientific interrogation in early-phase development and apply an N-of-1 or single-subject approach to clinical trials. That is, we have to learn as much as we can about each patient, including the molecular makeup of the patient’s disease—ideally, before treatment and at the time of disease progression—and about the drug’s clinical and molecular effects.
Early on in the process, we need to filter out the more likely winner compounds from the more likely losers. Again, large treatment effects and outcome improvements should be the key criteria. This should enable many more drugs to succeed in pivotal-phase development.
While this strategy may help reduce cost, there will be added costs for molecular diagnostics, tissue processing, and compensation for study sites. There cannot be precision medicine without precision diagnostics, and the traditional business model of the diagnostic industry will need to change.
Forming close partnerships between the biopharmaceutical industry and diagnostic companies early on, when there is still considerable risk for failure, will be absolutely critical for success. We have already started seeing that change: Some diagnostic companies are much more willing to accept higher risk and to partner with the biopharmaceutical industry in codeveloping a drug and a diagnostic assay.
Over the next decade, will more winner-targeted therapies result in more cancer cures?
For advanced cancers, I think we will more likely see the transformation of acute life-threatening diseases to more chronic ones, where patients live a long, high-quality life. There also may be a small subset of certain cancers where cure is on the horizon. For example, recent observations with anti–PD-1 antibodies and CTLA-4 inhibitors in the treatment of advanced melanoma suggest that a proportion of patients may have durable responses that last for years. It remains to be seen whether this is a cure—at some point, the cancer cells might learn to evade the checkpoint blockade.
In many cases, there may be effective second, third, fourth, and fifth lines of treatment available, which collectively lead to long-term outcome benefit. This too has major implications for drug development.
One consequence is that survival may no longer be a practical primary endpoint for measuring clinical efficacy benefit, because it may simply take too long to get to that endpoint. Also, survival may get confounded by multiple lines of subsequent treatments.
So we need robust and clinically meaningful surrogate early endpoints, and they need to become acceptable to regulatory agencies as a basis for approval. In addition, payers will need to change their cost-effectiveness models and criteria because they are largely based on survival as an outcome measure. ■
Disclosure: Dr. Lehnert is Group Head, Oncology Clinical Research, Takeda Pharmaceuticals, Inc.