Avastin for Colon Cancer: A Conflicted View

In the long-running war to hold down health care spending, controlling cancer chemotherapy costs looms as the next great battleground. Pharmaceutical companies have been slapping price tags on their latest cancer drugs from $50,000 to $160,000 a year, even though most of them have limited ability to extend the lives of people in the final stages of this devastating illness.

In the long-running war to hold down health care spending, controlling cancer chemotherapy costs looms as the next great battleground. Pharmaceutical companies have been slapping price tags on their latest cancer drugs from $50,000 to $160,000 a year, even though most of them have limited ability to extend the lives of people in the final stages of this devastating illness.

A new study that appeared today in the Journal of Clinical Oncology documented the small bore efficacy of one of these high-priced drugs – Avastin – in the Medicare population. Yet the study’s implications were immediately dismissed in an accompanying editorial written by a clinician at a leading academic medical center. That clinician also happened to be on the payroll of the company that makes the drug – Roche/Genentech.

The dust-up over using Avastin for advanced colorectal cancer is only a minor skirmish in the escalating war against rising health care costs. But it reveals the height of the institutional hurdles that must be scaled before the cash-strapped Medicare program even gets a chance to make sound scientific judgments about the effectiveness and cost-effectiveness of the latest therapies.

The study, by researchers at the Dana-Farber Cancer Institute in Boston, looked at the outcomes from adding Avastin (generic name bevacizumab) to chemotherapy regimens for patients with late stage colon cancer. The study was based on a retrospective review of more than 2,500 Medicare patients who were diagnosed with the disease between 2002 and 2007.

You may recall that Avastin last year made headlines when the Food and Drug Administration withdrew its preliminary approval for advanced metastatic breast cancer because the most recent clinical trials showed it did not extend life. The original approval for colon cancer was in 2004, and it was based on trials that showed median survival from the onset of therapy rose from 15.6 months to 20.3 months.

The good news in the latest study is that the drug was indeed beneficial for Medicare beneficiaries, who are older on average than the patients in the original trial. So there’s no reason for the FDA to withdraw its approval.

But lives among the over-65 crowd with the disease were extended for a shorter period of time than those in the original trial, and it was only among those who also received one particular chemotherapy regimen in addition to Avastin. Moreover, exposure to Avastin, which inhibits blood vessel formation, increased the risk of vascular perforations and strokes.

“Elderly patients with colorectal cancer can be counseled that including bevacizumab (Avastin) in first-line therapy regimens for metastatic colorectal cancer seems to be no more than marginally effective,” the authors concluded.

The Journal of Clinical Oncology (JCO) goes out to 25,000 subscribers – nearly every oncologist in the country. All things being equal, this study could have a major impact on the practice of medicine, since the majority of the 50,000 patients who die each year from colon cancer are eligible for Medicare, a program that is under tremendous pressure to cut costs. That, were it to occur, would have a major impact on Genentech’s bottom line. The company generates an estimated $6 billion a year from the drug, most of it from treating Medicare patients with colon cancer.

Yet whom did JCO ask to write the accompanying editorial interpreting this new piece of evidence? Herbert Hurwitz of Duke University Medical Center offered a host of caveats in his interpretation of the study. Using survival as the sole endpoint may overlook the psychic well-being associated with delaying tumor progression, even if the patient ultimately doesn’t live any longer, he wrote. Registry studies (in this case, outcomes generated from Medicare records) are not as scientifically robust as the gold standard of clinical research – the carefully controlled randomized clinical trial.

Hurwitz also complained that many of the trials that combined Avastin with chemotherapy were done outside the U.S. “where access to additional anticancer treatment and some supportive care is often more limited – factors that tend to reduce compliance and benefit.” But then he turned around and noted that those same trials showed “the clinical benefit observed in older patients has been consistently similar to the clinical benefit seen in younger patients.”

Clearly, this wasn’t an editorial that wanted to acknowledge the limited benefits of this drug. It seemed to be looking for every possible reason to continue its widespread and escalating use.

That’s why at the very end of the editorial, it was shocking to read this disclosure: Dr. Hurwitz was a consultant to Roche/Genentech, which manufactures and markets the drug. He had also received honoraria and research funding from the firm, as well as others.

I called the editorial offices of JCO to inquire about its policy covering writers of editorials that interpret studies that appear in its pages. Are physicians with financial ties to specific drug firms allowed to write editorials that will impact the sale of that company’s drugs? Many journals have specifically forbidden scientists with conflicts of interest from writing editorials.

“We don’t have a hard and fast policy,” said a spokeswoman for the American Society of Clinical Oncology, which publishes JCO.

The New York Times today reported that the Centers for Medicare and Medicaid Services has published its preliminary rule for the section of the Affordable Care Act that requires pharmaceutical, medical device and other health care companies to divulge the names of physicians that it has on its payroll as consultants, speakers, or outside researchers. Disclosure of conflicts of interest is a fine thing, and the law is long overdue. Even JCO disclosed Hurwitz’ ties to Roche/Genentech.

But that doesn’t do a thing to end the pervasive influence that conflicts of interest have over the practice of medicine, or limit the myriad ways those conflicts drive up costs. As this latest dust-up over Avastin for colon cancer shows, that influence begins with how the science is interpreted by the medical journals, which in turn is reflected in the continuing medical education that practicing physicians receive and the content of clinical practice guidelines, which dictate which therapies be reimbursed by Medicare and other insurers. And that, of course, is the ultimate bottom line.