Washington, DC—Precision medicine is the optimal approach to the future of cancer care, but unless a new pricing model for innovative treatments is implemented, the cost of care will be the greatest barrier to effective treatment, said Maurie Markman, MD, President, Cancer Treatment Centers of America, and Clinical Professor of Medicine, Drexel University, Philadelphia, at the Sixth Annual Conference of the Association for Value-Based Cancer Care.
Considering that sequencing the entire genome of a tumor and the corresponding germline of an individual patient with cancer now costs <$5000, the “future” is already in the present, said Dr Markman. One of the challenges to taking full advantage of this capability is optimally converting the massive quantity of raw data generated by genomic sequencing into information of genuine value in the management of the individual patient, he said.
“The idea is…that the standard of care eventually for an individual patient with cancer would be to find what is driving that cancer forward, with the idea of being more precise,” said Dr Markman. “Not only are you going to be more effective with your therapy and reduce side effects, but also get the right drug to the right person at the right time,…ultimately at a cost we can afford.”
Molecular Matching Improves Treatment Outcomes
Precision medicine is not a new concept in oncology, but its application has often resulted in an oversimplification of complex biology. Dr Markman recounted early studies of agents targeting epidermal growth factor receptor (EGFR) overexpression as treatment for ovarian cancer. Several trials were considered failures at the time they were concluded, but, in fact, they revealed personalized medicine for the handful of patients who benefited from the treatment.
For example, in a study of gefitinib (Iressa) for the treatment of ovarian cancer (Schilder RJ, et al. Clin Cancer Res. 2005;11:5539-5548), the 1 patient who responded (among the 27 evaluable patients) had a tumor with a mutation in the catalytic domain of EGFR, the same mutation that was responsive to gefitinib in patients with lung cancer.
Another study (Bookman, MA, et al. J Clin Oncol. 2003;21:283-290) that was considered a negative trial at the time evaluated the anti-HER2 antibody trastuzumab (Herceptin) in patients with ovarian cancer and overexpression of HER2. The study included 54 patients and had a response rate of just 7.3%—1 patient had a complete response and 2 patients had partial responses.
Dr Markman argued that this study demonstrated the susceptibility of tumors with high HER2 expression to the study drug, reinterpreting this trial as an example of precision medicine rather than a failed study. Of the 54 patients who received treatment, only 14 had tumors with high HER2 expression (3+).
“If you have the target, if the abnormality is present, isn’t that where we want to increasingly go from the perspective of the patients we treat?” he asked.
With the availability of multiple gene panels at a reasonable price, molecular matching of treatments is now possible, and the evidence shows that molecular matching favorably affects outcomes, Dr Markman said. As reported by investigators at M.D. Anderson Cancer Center (Tsimberidou AM, et al. Clin Cancer Res. 2012;18:6373-6383), therapies that matched a patient’s molecular aberration were associated with a 27% response rate and a median overall survival (OS) of 13.4 months compared with a 5% response rate and a median OS of 9 months among patients with no matched treatment.
To show an improvement in OS, randomized clinical trials require a large number of participants, which are time- and cost-prohibitive. Alternatives include the so-called basket trials, in which treatment is assigned based on identified molecular abnormalities rather than on the site of origin of the tumor or its morphology. Several such trials are in progress, including the National Cancer Institute–Molecular Analysis for Therapy Choice (ie, NCI-MATCH) trial and the American Society of Clinical Oncology (ASCO)’s Targeted Agent and Profiling Utilization Registry (TAPUR) trial. Another option is the use of “N of 1” experiences, assuming an available mechanism for drug payment.
Disseminating the results of precision medicine in oncology is another challenge, because the peer-reviewed publication process is slow. A public database of “N of 1” experiences is planned as a component of ASCO’s big data initiative, CancerLinQ.
The Problem with Overall Survival As An End Point
Randomized clinical trials with an end point of OS are problematic in showing the true value of a treatment, because of the difficulty in isolating the impact of a particular treatment on a predefined relevant outcome. For these trials to have meaning, the control and experimental populations need to be relatively homogeneous except for the treatment being examined. As such, clinical trial populations rarely represent the real world of patients with cancer who tend to be older and have multiple comorbidities, said Dr Markman.
In addition, OS data from clinical trials of the same treatment regimen conducted years apart will not reflect improvements in subsequent lines of therapy over time that also improve OS. For this reason, progression-free survival represents a more meaningful end point and has recently been shown to correlate strongly with OS.
Finally, the potential impact of germline variants on toxicity and efficacy is rarely considered, because “we don’t have any way, in this country, to use this information at a clinical level today,” Dr Markman said.
“We have to find out who will benefit, versus who will not, with the drugs we have available,” he said. “Again, that’s precision cancer medicine.”
Cost of Treatment the Greatest Barrier
Dr Markman expressed strong concerns about the cost of cancer care, which he argued may put optimal care out of reach for many patients. He pointed to immunotherapy and the escalating costs as combinations of immunotherapies may prove to be most effective in treating cancer.
“We will very soon…start seeing combination immunotherapy regimens. Let me assure you, the $1 million a year price tag is on the horizon. It’s not $1 million a year for a year. It’s $1 million a year for many years per patient,” he said, adding that a 20% copayment or deductible on that kind of price tag is not realistic for patients.
“Immunotherapy will bankrupt the United States healthcare system, unless we change,” Dr Markman predicted. “Our society, individual patients and families, simply cannot continue to pay the charges. Make it very clear, we’re not talking about cost. That’s an arbitrary number about what the market will bear. It doesn’t seem to have a limit yet.”
Identification of biomarkers that will propel a move toward precision medicine to define the patients who will benefit from expensive therapies is the change that must happen to improve the affordability of cancer care.