Here is a current story that illustrates some of the major conundrums of modern drug development and health care. For those who are familiar with this little tale-in-progress, my guess is that you are, like me, always up for another recounting.
The story concerns the drug Avastin, an anticancer drug often cited as the evidence that costs are getting out of hand. Approved for the treatment of colorectal and lung cancers, this agent – a monoclonal antibody that blocks vascular endothelial growth factor – can cost $100,000 per treatment (Fun fact: Global sales for Avastin in 2009 totaled $5.9 billion). The extension of survival time can be as small as a few weeks.
In March 2007, Avastin (generic name, bevacizumab) was approved by the FDA for the treatment of metastatic breast cancer. (New England Journal of Medicine study PDF) That approval was contingent on further data to be provided by the manufacturer, Genentech, confirming the overall survival benefit for patients given this drug.
Overall survival (OS) is, of course, considered the gold standard measurement of benefit. Does a patient live longer as a result of this medication? But measuring OS can be tricky. If a drug might be helpful, shouldn’t it be approved as soon as possible so that patients in need don’t have to wait? In that case, surrogate measures should be taken into account. And they are. Progression-free survival — the amount of time that a patient’s cancer is at a standstill — is often used as an early indicator of benefit. Another confounding factor in measuring OS is the fact that patients receiving an experimental drug have often been through the ringer, exposed to many different compounds. Isolating the exact effect of the exact drug can be difficult. Also, if a drug helps a patient live better, even if not for longer, then it might be worth approving, especially for people suffering from painful advanced cancers.
In 2010, those follow-up studies were completed and submitted to the FDA. The FDA’s Oncologic Drugs Advisory Committee (ODAC) reviewed the data, held a long discussion (Work in Progress: Reading FDA Transcripts So You Don’t Have To!), and, in a 12-to-1 vote, decided to recommend that the FDA withdraw Avastin’s indication for metastatic breast cancer (MBC).
Here are the important data:
In the initial study, conducted by the Eastern Cooperative Oncology Group and known as ECOG 2100, MBC patients given Avastin plus paclitaxel (chemotherapy) had a progression-free survival that was about 5.5 months longer than MBC patients given chemotherapy alone. (Here is another PDF about it) In the two subsequent studies — known as AVADO and RIBBON-1 — there was no benefit in OS seen among patients given Avastin plus chemotherapy compared to those given chemotherapy alone. Sometimes OS was shorter for those given the monoclonal antibody regimen. PFS differences were observed. In the AVADO study, PFS was less than a month longer among patients given Avastin. In RIBBON-1, which tested a couple of different regimens, PFS was 1-3 months longer for the experimental arm, and was statistically significantly different to the control arm.
After many months of considering ODAC’s recommendation, and after many bitten-down nails among industry insiders, the FDA has recommended that Genentech voluntarily withdraw Avastin’s MBC indication. Genentech says it won’t withdraw the indication on its own, which means the FDA will likely take that action on its own very soon.
But that doesn’t mean that people with this type of cancer won’t have access to Avastin. The National Comprehensive Cancer Network, generally considered the authority when it comes to treating cancer, just opted to keep Avastin on its compendium for metastatic breast cancer (yet anotherPDF). That decision was made based on a vote of 15-0 in favor of keeping Avastin on the NCCN breast cancer compendium. In many states, insurers are required to the NCCN compendia when deciding what treatments to cover. Ultimately, whether a patient has insurance-covered access to Avastin may be a matter of what state they live in.
From a business standpoint, Genentech stands to lose out from Avastin’s dismissal. The drug costs about $88,000 for the treatment of breast cancer. The company earned an estimated $885 million per year after the FDA approved this indication in 2007. (Note – I wrote about this for Oncology Business Review in September 2010.)
The story continues to unfold, with all kinds of vitriol being thrown around. But several vital questions are raised by this event:
When it comes to deadly diseases, should the FDA approve a drug that could benefit some patients? It’s important to realize that at least a handful of patients on the RIBBON-1 and AVADO trials who were treated with Avastin experienced a stunning benefit. One patient, who testified to the ODAC committee, talked about living to see her grandchildren grow up as a result of this drug. That’s the few, not the many. If there’s a chance that patients could benefit, should the drug be made available? It’s also important to know that the RIBBON-1 and AVADO studies showed detrimental side effects associated with Avastin therapy. Some patients might benefit, but for others the treatment would do more harm than good.
Should the FDA consider cost? ODAC did not weigh cost into its considerations – that would be completely against the FDA’s mandate to consider drugs only in terms of safety and efficacy. Still, it’s hard to believe that cost doesn’t come into it at all. If Avastin cost $10 per treatment, would it be approved for MBC? In the UK, the FDA equivalent – NICE – does consider cost and has refused to approve Avastin for MBC (and even for other cancers for which Avastin is approved here), along with many other drugs that are approved in the US. It’s unlikely that the FDA will grow another limb to evaluate the cost-benefit ratio, and in the current political climate it also seems highly unlikely that a separate government branch will sprout anytime soon.
If most private insurers, along with Medicare and Medicaid providers, follow the NCCN compendia, then what is the significance of the FDA’s decision? As David Guy, longtime oncology business expert, explains, these state laws apply to about 75% of the US population, but legal loopholes could enable insurers to bypass that requirement. Already, an insurer in Oregon has withdrawn coverage of Avastin for MBC on the basis that it’s not medically necessary (in other words, the move is not based on the FDA’s decision).
There are many discussions about this story online (here is a particularly good one), and the debate spurred by the FDA’s decision is ongoing. The questions raised aren’t easy to answer. It’s impossible to know what it’s like for someone who needs access to life-saving treatments, and personally I can understand wanting anything and everything to be available. (Think of all the AIDS patients who did not have access to even experimental drugs in the 1980s, despite the fact that everyone knew they were going to die.) On the other hand, there needs to be some authority helping to evaluate whether a treatment is indeed useful or not, however imperfect that evaluation might be.
What do you think?