After two days of hearings between the FDA and Roche/Genentech, an advisory committee voted unanimously to revoke the previously granted accelerated approval of Avastin for treatment of advanced breast cancer. The result has been confusion over the basis for this decision, fear that the drug will no longer be available to patients currently receiving it, and anger that this represents another manifestation of the dreaded “ObamaCare” menace which denies healthcare to save money. Immediately after the decision was announced, CMS (Medicare and Medicaid) stated they would continue to reimburse for Avastin for the above indication. To further keep us in limbo, the committee’s decision has been referred to the final arbiter for his ruling which is not expected for 6-8 weeks.
In an effort to provide some clarity for patients and their families, I contacted representatives of Genentech who kindly provided me the clinical trials discussed at these hearings. First some general comments:
- The gold standard of clinical research is prospective, randomized, double-blind clinical trials, abbreviated RCT. Negative features of RCT’s are they are very expensive to conduct (up to $1 billion) and take up to 10 years to complete. This leads to frustration that cancer research is not progressing rapidly enough.
- Less understood is that Inclusion/Exclusion Criteria select younger, healthier patients for entrance into these studies, and tend to exclude older, sicker patients who comprise a sizeable number of people seen in community based oncology. Thus, RCT’s measure the drug’s efficacy, not its effectiveness, putting the test drug in the best possible light. All 3 of the studies discussed were RCT’s
In February 2008, the FDA granted accelerated approval of Avastin for treatment of metastatic/advanced breast cancer after an FDA advisory committee voted 5-4 to deny such an approval based on a company-supported study showing a delay of tumor growth of 1-2 months with Avastin. Publication of the E2100 trial resulted in this change to a positive outcome for Genentech and Avastin. The study involved 722 patients treated with weekly IV Taxol (standard treatment) vs. Taxol + Avastin (experimental treatment) 10 mg/kg IV every 2 weeks. Addition of Avastin resulted in a doubling (5 months) in time to tumor regrowth and doubling in response rate (tumor shrinkage) from almost 25% to just under 50%. There were no complete remissions where all evidence of cancer went away, and overall survival was not improved. Therefore no one was cured. There was risk in taking Avastin (6 patients died as a result of the drug). But, the study demonstrated that Avastin had an effect in breast cancer. So, benefit (efficacy) was judged to outweigh risk, and the FDA granted conditional approval pending completion of 2 other studies that were in progress.
The second trial called the AVADO Trial included 736 patients treated with weekly IV Taxotere (almost identical to Taxol) vs. Taxotere + Avastin 7.5 mg/kg or 15 mg/kg IV every 3 weeks. Note the different doses and schedules of Avastin. Whether or not this explains the less impressive results is unclear. But tumor regrowth was delayed 0.7 months with the lower dose of Avastin, and 0.8 months with the higher dose. There was no improvement in overall survival.
The final study was the RIBBON I Trial involving 1,237 patients. Study design was similar (standard arm vs. standard + Avastin) . But here the similarity ends. Standard chemotherapy was left up to the oncologist. Avastin was administered at 15 mg/kg IV every 3 weeks. Looking at just the group treated with a taxane drug (Taxol, Taxotere or Abraxane) the delay in tumor regrowth was 1.2 months. And I should point out that the taxane was given every 3 weeks as in the AVADO Study, not weekly as in the E2100 Trial. Again no prolongation of overall survival was noted, i.e., patients didn’t live any longer whether or not they received Avastin. In this study Avastin was used at the 15 mg/kg every 3 weeks dosing schedule. I found it interesting that best results were noted when Avastin was combined with Xeloda (about 3 months) since this combination is used commonly in colon cancer where it has been shown to be beneficial.
As a result, an FDA advisory committee voted 12-1 against Avastin’s use in breast cancer in June, 2010. They accurately pointed out that the above 2 trials didn’t show more than a couple of months delay of tumor spread, and that benefit (efficacy) did not outweigh risk. By December 2010, the FDA announced plans to revoke approval for breast cancer, but adding that the drug will stay on the market for other cancers as well as “off label” use for breast cancer. This led to Genentech’s appeal and the current hearings.
It should be noted that RCT’s are designed so that patients receive at least what is considered standard of care. Thus, all patients received chemotherapy. They were not denied treatment in order to cut costs. I find this politically motivated objection absurd. The purpose of these trials is to identify what works and pay for it. If all 3 study designs were as close to identical as possible, including drugs used, doses and scheduling, perhaps the results of the second two studies would have confirmed the first. We are also entering the age of “personalized oncology” where we can identify subsets of patients with certain biomarkers on their cancer cells that predict for response, or lack of it, to a given agent. In fact, a promising marker for response to Avastin is currently being studied. Perhaps in the near future we will be able to individualize all cancer treatment. Wouldn’t it be nice not to need large, expensive and time consuming RCT’s to what will and won’t work in a given patient?
One final note: All physicians are now expected to practice “Evidence-Based Medicine”. In order to determine whether or not we are following guidelines appropriately, our government and health plans have mandated we purchase Electronic Medical Records systems. I estimate the total cost of purchase, update and maintenance of our system to date is over $1 million. And if we don’t follow agreed upon clinical pathways, reimbursement for providing these expensive chemotherapeutic agents is lowered. I agree that EMR’s are an improvement over paper charts. I agree we should practice Evidence-Based Medicine. But when faced with evidence that Avastin adds no benefit to patients with advanced breast cancer, Medicare states it will still reimburse for it, “who’s zoomin’ who”?