Testimony of Dr. Jennifer Yttri at FDA Advisory Committee on Tivozanib

May 2013

I am Dr. Jennifer Yttri and I am speaking on behalf of the Cancer Prevention and Treatment Fund of the National Research Center for Women & Families. Our non-profit research center scrutinizes medical data, evaluating scientific evidence of benefits and risks for patients.  We analyze and review research and provide objective and understandable health information to patients, health care providers, and policy makers.  Our organization does not accept funding from pharmaceutical companies and therefore I have no conflict of interest.

Today’s fundamental question is whether the one completed Phase 3 clinical trial is enough to approve the new drug tivozanib as a treatment for patients with renal cell carcinoma.

FDA guidelines recommend two trials that support efficacy of a drug. In some cases, FDA will approve a new drug based on only one trial if that trial shows a significant improvement over existing therapy. In the FDA’s own words: “A conclusion based on two persuasive studies will always be more secure than a conclusion based on a single, comparably persuasive study. Reliance on only a single study will generally be limited to situations in which a trial has demonstrated a clinically meaningful effect on mortality, irreversible morbidity… and [when] confirmation of the result in a second trial would be practically or ethically impossible.”

The NDA submitted for tivozanib relies on one phase 3 study comparing tivozanib to sorafenib. While tivozanib was shown to increase progression free survival by 3 months in patients with renal cell carcinoma, it did not improve overall survival. In fact, there was a non-significant lower overall survival for tivozanib, suggesting the drug may actually harm patients more than it helps them. This one study alone is not enough to meet the FDA’s guidelines for drug approval.

In addition, the study results have inconsistencies that raise red flags about the research and about the drug itself.

1.       Patients receiving tivozanib had increased progression free survival, so why were they more likely to die in the first 30 days due to disease progression, compared to sorafenib. This does not make sense. The deaths in the first 30 days could be due to chance, but the disease free survival could also be due to chance.  More research is needed to clarify the risks as well as the long-term benefits.

2.       FDA noted that 70% of sorafenib patients stopped taking the drug at least temporarily and 44% ended up with a reduced dose. This is much higher than other studies – for example one study highlighted by the FDA had only 14% interruption and 10% reduction.  There is no logical explanation for this, but these unusual problems could make sorafenib seem inferior to tivozanib, when in fact it might be superior.

These inconsistencies may be due to chance or to irregularities in how the studies were conducted in other countries.  Standard of care and assessment of disease varies in the US and other countries, and fewer than 10% of patients enrolled in the trial were in the US.  We agree with the FDA reviewer that it is preferable to enroll US patients, so that the study reflects the disease burden and treatment in the US and can provide better insight into treatment outcomes for US patients.

Regardless of the reason for the inconsistencies in the study, a second, independent, phase 3 study would help determine the safety and efficacy of this drug for treatment of renal cancer.

We urge you to recommend that the FDA require the sponsor to complete a second trial to confirm the positive effect of tivozanib on PFS; address the concern over lower OS; and provide better information on how generalizable the results are in the US population. Based on the data provided, there are no ethical concerns with requiring another trial. With the additional information, the FDA can make an informed decision as to whether tivozanib meets their standards of safety and efficacy.  That is not possible based on this one study with such inconsistent results.