In the journal The Oncologist, researchers report that the median survival for a group of Medicare patients on the drug sorafenib was three months, which was significantly lower than the median survival of nearly 11 months for patients treated with the drug during a phase III clinical trial. As the drug comes with significant side effects and a cost to patients and insurers of more than $10,000 a month, researchers are questioning the value of the drug for all patients.
“No drug that results in a three-month survival can be thought to be offering a meaningful life expectancy,” said Hanna K. Sanoff, MD, MPH, a UNC Lineberger member and an associate professor and section chief of the UNC School of Medicine Gastrointestinal Medical Oncology Program.
“This doesn’t mean that we shouldn’t prescribe it, but we should be mindful that the broader population of liver cancer patients is sicker than the patients in the landmark trial. Our patients deserve to know that the promise of nearly a year of life may not be their reality.”
The U.S. Food and Drug Administration approved sorafenib — known commercially as Nexavar – for the treatment of advanced hepatocellular carcinoma in 2007. In a phase III clinical trial, patients with advanced liver cancer had a median survival of 10.7 months on the drug, which was 2.8 months more than patients who didn’t get the treatment. But patients in that study also were in good physical condition and their level of cirrhosis, which nearly universally accompanies liver cancer, was well controlled.
“This drug was tested in a clinical trial with patients with mild cirrhosis who were pretty fit,” Sanoff said. “Because of concurrent cirrhosis, it may be that the gap between the trial population and the average liver cancer patient may be greater than in some other cancers.”
In the new study, the researchers analyzed survival data for a group of patients insured by Medicare, a government health insurance program for people aged 65 years and older or with disabilities, and who were diagnosed between 2008 and 2011. Of the 27 percent of 1,532 patients given sorafenib, median survival from the first prescription was three months, which was not statistically longer than survival of untreated patients.
“No drug that results in a three-month survival can be thought to be offering a meaningful life expectancy.”
They concluded that lower survival in the Medicare population was likely due to a generally sicker population. Further analysis suggested that patients in the study with earlier stage disease might be more likely to benefit from taking the drug.
The researchers pointed to issues of cost – both financial and in terms of side effects – as factors that patients and doctors should consider when deciding on a course of treatment.
“We need to question who we prescribe this to, not only because of the cost of the drugs from a side effect perspective, but also the actual financial cost,” Sanoff said.
In previous studies, researchers have found that the median monthly price for the drug across all available Medicare part D plans in 2014 was $10,811 per month, said study co-author Stacie Dusetzina, PhD, a UNC Lineberger member and assistant professor in the UNC Eshelman School of Pharmacy and UNC Gillings School of Global Public Health.
That price tag can mean thousands of dollars in out-of-pocket costs for patients, Dusetzina said, as most plans require cost sharing of at least 25 percent when filling the drug’s prescription. Even for patients who have reached the catastrophic spending level in Medicare Part D – when the amount they are expected to pay out-of-pocket decreases – they would still pay $540 per prescription fill each month.
“This is obviously going to present financial challenges for many patients,” Dusetzina said.
“This underscores the fact that establishing effectiveness of therapies outside of trial settings is complicated but important, if we want to really understand the value of cancer therapies. Translating the benefits of treatments into a ‘real world’ setting isn’t always easy.”
Patient data was drawn from the National Cancer Institute’s Surveillance, Epidemiology and End Results Program Medicare linkage.
This work was supported by grants from the National Cancer Institute, NIH Building Interdisciplinary Research Careers in Women’s Health K12 Program, and the North Carolina Translational and Clinical Sciences Institute. Additional support was provided by the Integrated Cancer Information and Surveillance System (ICISS), and from the UNC Lineberger Comprehensive Cancer Center with funding provided by the University Cancer Research Fund.
Conflicts of interest: Sanoff has received research grant funding from Bayer and Novartis. Another co-author Bert H. O’Neil of the Indiana University Simon Cancer Center has consulted for Bayer.
In addition to Dusetzina, O’Neil and Sanoff, other authors include YunKyung Chang, PhD, of UNC Lineberger; and Jennifer L. Lund, PhD, of UNC Lineberger and the UNC Gillings School of Global Public Health Department of Epidemiology.