Yesterday was a dark day for those holding Aveo Oncology (AVEO) or Delcath Systems (DCTH) through the FDA’s Oncologic Drugs Advisory Committee meeting. I was personally very bullish on the prospects for both of these agents to receive a positive vote, and wanted to share my perspectives on the proceedings.
AVEO’s Tivozinib for the treatment of advanced renal cell carcinoma (RCC)
In general, I see 4 main areas that contributed to a negative panel vote for this drug:
1. Tone set by Richard Pazdur, MD, Director of Division of Oncology Drug Products, and by Mikkael Sekeres, MD, Chair of ODAC Voting Committee
FDA framed the tivozinib review in the context of their concerns about the increased potential detrimental survival signal; the availability of other approved therapies in this setting; and their concerns with flaws in the study design and the unexplained 30 day safety/mortality discrepancies. Pazdur was very vocal throughout the session in reiterating these points, and the committee’s questions certainly focused extensively in these areas. Sekeres opened the questions to the sponsor with a scathing criticism directed to Robert Motzer, MD (principal investigator for the AVEO pivotal study) on the poor study design. In my opinion, the position of the FDA and the ODAC Chair definitely set a negative and somewhat hostile tone for the proceedings.
2.The pivotal study design was flawed
FDA themselves acknowledged that they did not intend to change the regulatory standard for the approval of new therapies for RCC, in other words, progression free survival (PFS) was an acceptable end point, but the issues with the study design were too great in the eyes of the panel. Unfortunately, this is a dilemma for regulators because it would be quite hard for a new therapy to demonstrate highly statistically significant incremental benefit above and beyond a drug like sunitinib in PFS and OS, so how should future studies be designed given the landscape of current standard of care? One of the panelists did mention this as a question; if two drugs are tested for non-inferiority in an advanced RCC setting, is that good enough for approval? The major issues with the study protocol cited were:
- Patients in the active comparator arm were not treated with current standard of care. (In the US, sunitinib is most frequently used in the front line setting, not sorafenib)
- Per protocol, why was the active comparator arm allowed to cross-over to experimental therapy. (Crossover studies are common but it was suggested as almost unethical that patients on the active treatment arm were allowed to crossover to experimental therapy. The protocol should have allowed for the experimental arm to crossover to active therapy upon progression, not the other way round)
- Per protocol, both arms of the study were allowed to receive standard of care/subsequent lines of targeted therapy upon documented progression. However, given that the majority of patients were enrolled in eastern European countries, access to 2nd line TKI’s was very restricted, and these patients were managed differently to current practices in the US (where multiple lines of therapy are available). Hence the natural history of these patients was not indicative of care in a US practice setting
3. There was a potential detrimental overall survival disadvantage with tivozanib
Although the absolute difference in overall survival between the two arms was only 2 weeks, the sponsor’s presentation of the cross over theory (i.e. patients on sorafenib who progressed, crossed over and received treatment with a second active drug and therefore lived longer) was not enough to convince the panel. One panelist fixated extensively on the imbalance observed in the initial 30 day mortality rate. Despite the fact that these patients were documented to have progressed on active treatment, the panel seemed concerned that latent issues with safety could be at play, both as early as 30 days, and in terms of overall survival. This was compounded by the FDA presentation outlining reviews of other targeted therapy trials where patients have been treated sequentially with two targeted agents, and this has never lead to a worsening hazard ratio. Sekeres brought up a valid point with the committee in terms of his struggle as to how he would envisage counseling his patients about the risk/benefit of this drug in light of improved PFS but a potential detrimental OS. As pointed out quite belligerently by Pazdur, the sponsor didn’t exactly help themselves by omitting any reference to OS in their proposed labeling and instead chose to omit this completely.
4. The improved toxicity profile did not translate to a definitive improvement in patient quality of life
Unfortunately, despite a far superior adverse event profile, once daily dosing regimen, and lower level of dose modifications or discontinuations, the sponsor was unable to present any compelling evidence from their studies that patients experienced an improved quality of life. While there was a number of compelling patient and caregiver testimonies during the open public hearing session, the panel were not swayed, and voted against the overall risk benefit for this product.
So where does this leave AVEO?
A complete response letter is highly inevitable by their PDUFA goal date of 7/28/2013, following which the sponsor can request a type A meeting to discuss a potential regulatory path forward. I think FDA will need to work closely with the sponsor to provide input on any future studies, given that benchmarks for approval probably need to be shifted. Placebo based studies are no longer ethical, and it is unrealistic to expect drugs to significantly outperform current standards of care. I still believe tivozinib is an active agent in RCC, we shall have to see whether the sponsor has the time and resources available to complete a second pivotal study that is likely to succeed.
In connection with AVEO, PropThink has taken a long position.