AVBCC Expert Panel: Impact of Personalized Medicine on Future and Current Therapies

Wayne Kuznar

June 2014, Vol 5, No 5 - AVBCC 2014 4th Annual Conference


Los Angeles, CA—Increasing complexities in diagnostic science, the development of precision medicine, and the use of targeted agents require un­precedented levels of collaboration between pharmaceutical manufacturers, government agencies, and payers, said oncology experts during a panel discussion on personalized medicine at the Fourth Annual Conference of the Association for Value-Based Cancer Care.

As with the development and reimbursement of orphan drugs, all interested parties need to join forces, said George W. Sledge, Jr, MD, Chief, Medical Oncology, Stanford University Medical Center, CA. “The big challenge in the genomics era is ‘the number needed to study’ question,” he said.

Dr Sledge elaborated, “How do we collect rigorous data? It is statistically impossible to do phase 3 studies like we did in the past.” With personalized medicine, many cancers will fit the description of an orphan disease, and the US Food and Drug Administration approaches orphan disease very differently, he added.

When asked by panel moderator Michael A. Kolodziej, MD, National Medical Director, Oncology Solutions, Aetna, how he defines “success” in terms of personalized medicine, Dr Sledge answered, “I want care that is delivered to an individual in a timely fashion and in an effective way that maximally prolongs the life of the patient and minimizes the toxicity, without creating too much financial catastrophe, either for the patient or for the system.”

The era of precision medicine in oncology is not quite upon us, Dr Sledge said, noting the uncertainty involved in the treatment of a patient with multiple driver mutations of a cancer. That uncertainty extends to safely combining agents, leaving trial and error as an unsatisfying mechanism to do so.

Big Data and Genomics
A technical analysis of genomics tests as used in clinical practice is the duty of the oncologist, said Kevin B. Knopf, MD, MPH, Medical Oncologist, California Pacific Medical Center, San Francisco.

“Oncotype DX is useful sometimes, but I think I order it far less than other community oncologists do, because my research interests are in looking at the economic utility of these tests,” Dr Knopf said. “While everybody’s very excited about big data and genomics, right now we have a lot more noise than signal. I think that the community oncologists owe it to their patients to be skeptical about what tests are actually helpful to their patients and help them achieve good outcomes.”

Dr Knopf cited KRAS testing as an example of a genomics test with utility, because it can identify instances in which cetuximab would not be active, perhaps saving $10,000 monthly in unnecessary treatment costs.

Although there are thousands of potential genetic targets, only approximately 1 dozen have a targeted agent, said Christiane Langer, MD, Associate Group Medical Director in Oncology, Genentech, South San Francisco, CA.

“We need to understand the science and predict early which ones will be the drugs that are really going to make the difference,” Dr Langer said. “That would be a dream.” The most efficient way is to collaborate on clinical trials, and this collaboration should involve the biopharmaceutical manufacturers, medical societies, and the National Cancer Institute, among others. “Otherwise, it will take a long time,” said Dr Langer.

Jennifer Malin, MD, PhD, Medical Director, Oncology and Care Man­agement, WellPoint, said, “In the perfect world, we would have 1 test that could figure out the different targets, and for each patient, which drug or combination of drugs they should take.” Ideally, the combination would cost less than $800 monthly, Dr Malin said, unless it offers a cure, in which case it would have value at a much higher price tag.

The patient’s voice should not be ignored in the determination of value, said Kim Thiboldeaux, President and Chief Executive Officer, Cancer Support Community. “In the development of all these great tools and paradigms and solutions, in what meaningful way did you involve the patient voice in solutions? How would the patient define value? What would the patient define as important, in terms of what we are measuring in cancer care?” Ms Thiboldeaux asked.

Will Personalized Medicine Offer Cures?
Dr Kolodziej asked whether personalized medicine can offer a cure. “The answer is yes [for a cure], but it’s going to be a lot of one-offs rather than a global solution,” said Dr Sledge. “Big data attached to a lot of small science is going to solve some of the problems for some of the patients.”

The mechanisms of resistance to epidermal growth factor receptor–targeted therapy are different in lung cancer than they are in colorectal cancer and head and neck cancer, “so the idea that we’re going to have a one-size-fits-all, across all diseases, is a joke,” Dr Sledge said. “We already know it’s wrong.”

Louis Jacques, MD, Chief Clinical Officer and Senior Vice President, ADVI, a healthcare advisory services firm, wondered if choosing cancer treatments may someday resemble choosing antibiotics in infectious diseases.

“Is what we’re seeing, in terms not only of heterogeneity but essentially evolution of individual tumors, essentially pointing us to having super cancers that aren’t necessarily going to respond to much of anything, like superbugs in infectious diseases?” Dr Jacques asked.

Although the past 15 years have supplied adjuvant trastuzumab in breast cancer, rituximab added to the CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) regimen for the treatment of diffuse large-cell lymphoma, and imatinib in chronic myeloid leukemia as examples of agents that have delivered cures, Dr Knopf sees most of the reward from personalized medicine coming in the form of prolonged survival while maintaining quality of life.

Defining Value of Personalized Medicine
Dr Kolodziej asked, “If we accept as a premise that personalized medicine may not affect our ability to interpret a conventional cost-effectiveness analysis,…should we look at the information that we get from personalized medicine through a different lens?” How should we define the value of personalized medicine in oncology?

Dr Kolodziej pointed out that when using survival benefit, a traditional end point of cost-effectiveness, crizotinib at $5000 monthly does not meet the National Institute for Health and Care Excellence (NICE) definition of cost-effectiveness for the treatment of patients with non–small-cell lung cancer (NSCLC) whose tumors test positive for ALK gene rearrangement, despite being an effective agent for NSCLC.

Different criteria for cost-effectiveness should be applied to different treatment settings, said Dr Sledge. Trastuzumab, for example, improves median survival by 5 months in the metastatic setting, but it cures patients in the adjuvant setting, which may unequal different value. In the case of crizotinib, although resistance may eventually develop, it may represent the first step in an effective combination that may eventually cure the patient.

“To ignore these subtleties by talking about [cost-effectiveness] solely in the context of the first use of the drug, I think is an error,” Dr Sledge said.

One challenge in considering cost-effectiveness is further precision in selecting patients who will benefit from a given therapy, said Dr Malin. “Of 100 people who get trastuzu­mab as adjuvant therapy, 60 of them wouldn’t have had their disease recur anyway, and 20 still have their disease recur,” she said. “We’re giving 100 patients treatment to benefit 20.”