Evidence Suggests Likely Approval of Cabozantinib for Metastatic Medullary Thyroid Cancer

Exelixis (NASDAQ:EXEL) has a Prescription Drug User Fee Act (PDUFA) action date on November 29th, 2012 for cabozantinib as a treatment for patients with progressive, unresectable, locally advanced, or metastatic medullary thyroid cancer (MTC).  The FDA’s Oncologic Drugs Advisory Committee (ODAC) had a discussion of cabozantinib on its schedule for the November 8th-9th meeting but the company recently announced that cabozantinib was removed from the schedule.  The question is whether this cancellation is a positive or negative indicator.  To get at this question, it is useful to work through two of the most likely reasons that the cancellation is negative: (1) the drug is so marginally effective the FDA sees no reason to discuss it or (2) the drug is effective but so toxic that the side effects clearly outweigh the benefits.  If these are not plausible (or not supported by what we know of the efficacy and safety of cabozantinib) then the risk/reward profile of the bullish case looks attractive, especially after a recent pullback.

Is Cabozantinib an Effective Treatment for MTC?

The MTC phase III trial data were recently presented at ESMO 2012 with additional PK data.  The presentation had three critical pieces of information on the effectiveness of cabozantinib.  First, the progression free survival (PFS) was longer (and statistically significant) for patients receiving cabozantinib as compared to placebo (11.2 versus 4.0 months leading to a hazard ratio [HR] of 0.28 and P <.0001).  The PFS benefit was seen in all RET mutation subgroups with PFS hazard ratios of 0.24, 0.47, and 0.30 (RET mutation positive, negative, and unknown).  Second, the progression free survival rate at 12 months was 47.3% for cabozantinib treated patients compared to 7.2% for placebo.  Finally, the objective response rate (ORR) for the cabozantinib treated patients was 28% compared to 0% for those receiving placebo.

While the data look impressive, it is important to place them into context of other treatments for MTC.  The most recently approved product for the treatment of MTC is vandetanib (trade name of CAPRELSA and marketed by AstraZeneca (NYSE:AZN) and approved in 2011 for late-stage, metastatic MTC in adult patients who are ineligible for surgery).  Vandetanib did have an ODAC panel and the FDA presentation is insightful.   While vandetanib and cabozantinib data should not be considered apple to apple comparisons, we can use the vandetanib efficacy and safety profile as rough guides to the standards for approval in MTC.  First, both the vandetanib and cabozantinib studies used PFS as the primary endpoint and in its pivotal study vandetanib had a PFS HR of 0.35.  This is slightly worse than cabozantinib but it is always difficult to compare across trials (perhaps especially in this case as the median PFS of this trial for both treatment and placebo are much higher than the cabozantinib trial, implying differing baseline characteristics for the patients in these studies).  At the very least, one could say that cabozantinib does not appear materially worse and might be slightly more effective in terms of the PFS HR.  Second, when looking at ORR cabozantinib is once again on par if not slightly better (47.3% for cabozantinib compared to 45% for vandetanib).  The same caveat applies to comparing across trials but this is another piece of evidence that the effectiveness of cabozantinib is not materially worse than recently approved MTC treatments.

In general, it seems a difficult case to argue that the ODAC panel was cancelled because of a perceived lack of effectiveness.  Its effectiveness as measured by PFS HR and ORR appear similar, if not slightly better, than the recently approved vandetanib.  This is not to say that cabozantinib is the clearly better drug.  The question is whether the efficacy of cabozantinib is weak enough to deny it outright without an ODAC panel.  The only time a significant difference appears between cabozantinib and vandetanib is comparing the median PFS of the two compounds with 11.2 months (cabozantinib) versus 30.5 months (vandetanib).  This stark difference, however, is likely driven by different patient populations as the control group median PFS show a similar difference with 4 months for the cabozantinib control group versus 19.3 months for the vandetanib control group.  While it is odd to see such a major difference, the fact that the control group is similarly different implies that these are two different populations and the proper comparison is the PFS HR.  In addition, if this was a major concern, then this is likely something that the FDA would want to discuss at an ODAC panel as opposed to rejecting the compound outright.

Is Cabozantinib Not Safe Enough?

If we can rule out efficacy as a potential reason for the FDA to cancel the ODAC, then the second most likely explanation is safety.  In general, the side effect profile of cabozantinib has been characterized as manageable and not problematic.  Patrick Schöffski, MD, professor at the Department of General Medical Oncology at the University Hospitals of Leuven, Catholic University Leuven, Belgium noted that, “adverse events were generally manageable, allowing treatment with the oral drug for extended periods of time” (see link).  In terms of grade 3 or greater reactions, the most common (cabozantinib versus placebo) were diarrhea (15.9% versus 1.8%), palmar-plantar erythrodysesthesia (12.6% versus 0%), fatigue (9.3% versus 2.8%), hypocalcemia (9.3% versus 0%), and hypertension (7.9% versus 0%).  This side effect profile looks even better when compared to vandetanib.  The FDA found (slide 30) that over 35% of patients treated with vandetanib experienced QTcF prolongation of more than 60ms, which is considered a grade 4 adverse event.  This led to a safety conclusion that vandetanib “has considerable toxicity, which in some instances mirrors or is worse than the symptoms of untreated medullary thyroid carcinoma” (slide 41).  Ultimately, this was not sufficient to reject vandetanib but it does have a Risk Evaluation and Management Strategy (REMS).

Of course, the point is not to question the effectiveness or side effect profile of vandetanib but to compare it to cabozantinib.  Even with those side effect concerns, vandetanib was approved.  So if the ODAC panel was cancelled due to concerns over the safety of cabozantinib, they would have to have been very severe side effects.  In the data presented on the side effect profile of cabozantinib, there does not seem anything on that scale.  Of course, it is always possible that a new signal emerged that has not yet been presented.  Three points discount that possibility.  First, it would likely be a material event that would have been reported to investors, although what an investor thinks needs to be disclosed and what a company thinks needs to be disclosed are not always the same.  Second, if a new safety signal emerged, would this not be the type of event that the FDA would want to discuss with the company and experts in an ODAC panel?  The FDA clearly had questions over the safety profile of vandetanib and used the ODAC panel to delve into this issue.  One would expect if a similar type of issue arose that the FDA would want more information.  Third, and perhaps most compelling, if there was a new and severe safety signal then it seems unlikely that the FDA would allow cabozantinib to have an expanded access program.  Exelixis has had an expanded access program ongoing and it has been confirmed as recently as October 2012.  The continuing expanded access program is completely inconsistent with a drug that has a new and severe safety issue (one that is significant enough to reject without an ODAC).  Overall, it seems likely that cabozantinib has a safety profile that is likely superior to the approved vandetanib and no new safety signals have emerged.

Why Cancel the ODAC?

In general, the two most plausible bearish reasons to cancel the ODAC do not seem supported by available evidence.  While it is always possible that some other factor is at work, the risk/reward favors the bullish case going into the PDUFA.  The efficacy appears on par if not slightly better than the recently approved vandetanib and the safety profile has not seen any of the major concerns that were associated with vandetanib.  In addition, it is unlikely that a new safety problem arose given the continued expanded access program.  Exelixis shares have been under pressure since their most recent secondary and the price has not been able to gain any real traction to the upside.  As such, those wanting to play the MTC (likely) approval should buy at the low end of its trading range and not chase it on the runs.

It should also be noted that over the long-term, the MTC approval is going to be a rounding error compared to the potential in prostate cancer.  As such, this note should only be read in the context of the FDA action as related to potential approval in MTC and not a longer term call on the potential of cabozantinib in prostate cancer.  In addition, the company has recently raised a significant amount of cash, generating a large number of warrants that can pressure the price over a longer time horizon.  Regardless of the potential pitfalls of holding Exelixis for the longer term, an approval of cabozantinib as a treatment for patients with progressive, unresectable, locally advanced, or metastatic medullary thyroid cancer (MTC) would certainly be a bullish catalyst over the short term, and an analysis of recent events and known data point towards approval.