|
Senior Member
Join Date: Aug 2006
Posts: 3,380
|
ODAC Says Toxicity Trumps Meager PFS and TTP Benefit for Yondelis, Doxil
The Pink Sheet Daily. 2009 Jul 20, J Hammon
Changing endpoints midway through the clinical trials for the two Centocor/Ortho Biotech oncology drugs reviewed by FDA's Oncologic Drugs Advisory Committee July 15 did not help their regulatory prospects - but in the end it was the poor efficacy and unacceptable toxicity that doomed both Yondelis and Doxil. The sponsors, both divisions of Johnson & Johnson, were tasked with defending two applications: Yondelis (trabectedin) in combination with Doxil (doxorubicin HCl liposomal injection) for relapsed ovarian cancer, and Doxil in combination with docetaxel for locally advanced or metastatic breast cancer.
Committee members voiced concerns about the change in endpoints, but regardless of the efficacy measure being considered, the poor findings led them to vote against both applications. The vote against Yondelis was 14-1 and the vote against Doxil was unanimous.
Oncology drug sponsors often look to surrogate endpoints as a faster route to approval. The trials for both Yondelis and Doxil originally were designed with overall survival as the primary endpoint. But the sponsor subsequently changed both endpoints, to progression-free survival (PFS) for Yondelis and time-to-progression (TTP) for Doxil.
Though the Office of Oncology Drug Products has made many public efforts - through a series of workshops and guidances - to modernize the range of acceptable endpoints for cancer drug trials, office Director Richard Pazdur started out the meeting by clarifying misconceptions that surrogate endpoints offer an easier standard for approval.
"We would expect not only a statistically persuasive finding, but one that would be clinically relevant in the proposed indication and have a positive risk-benefit analysis," Pazdur stated.
Pazdur's remarks set the tone for both sessions of the meeting, making it clear that the agency was not impressed with either application - that for both, efficacy was questionable, toxicity was substantial, and that overall survival findings were either yet to be seen (interim results were presented for the Yondelis trial, and were not favorable) or clearly failed (the Doxil data were final).
Though the FDA official stated that the purpose in holding the meeting was to seek ODAC's advice on the balance of benefit and risk, Pazdur used the meeting to drive home the high standards that FDA holds for both efficacy and toxicity, even in life-threatening conditions.
Yondelis PFS Results Fail To Impress
The combination of Yondelis + Doxil showed a median PFS of 7.3 months compared to 5.8 months for Doxil monotherapy. In addition, the interim analysis of overall survival found no benefit.
That six-week PFS benefit came at the cost of greater toxicity: In the Yondelis combination therapy arm, grade 3-4 neutropenia was three times more frequent (63% vs. 22%), febrile neuropenia was four times more frequent (8% vs. 2%), and grade 3-4 thrombocytopenia was six times more frequent (18% vs. 3%) than with Doxil alone. Additionally, grade 3-4 transaminase elevations were 25 times more frequent with the Yondelis combination arm (50% v. 2%), which also contained six cases that met Hy's law criteria for hepatotoxicity and a higher number of cardiac events.
FDA questioned the clinical relevance of the six-week improvement, and Pazdur pointed out that there are other therapies on the market for the indication: "In a situation where there are no other therapeutic options and all other therapeutic options have been exhausted, there may be a greater acceptance of risk. With the current application, other treatments are available, including re-treatment with platinum."
He also said that for a novel agent, more evidence should be presented. "This is a new molecular entity - a new NDA - not a supplement, and it is supported by only one randomized trial," he told the panel. Pazdur noted that Lilly's Gemzar (gemcitabine) was approved for ovarian cancer based on a PFS endpoint, but that was after the drug had been marketed for over a decade in the U.S. for indications in non-small cell lung cancer, pancreatic cancer and metastatic breast cancer.
Panel member Wyndham Wilson, National Cancer Institute, seconded this point: "We are not looking at a drug that has been on the market where there is extensive experience with it. We are looking at a drug for approval for the first time ... the bar needs to be much higher. We are also looking at a setting where there is a fairly significant risk profile associated with it - liver function, cardiac, hematological - and we only have a glimpse of how that might play out in large numbers of folks over long period of time. I don't think it's fair to put the standards of other drugs along-side this one for those reasons."
A Path Forward For Yondelis?
Despite its current problems, there could still be hope for a Yondelis approval. The data presented at the meeting was from an interim analysis, and a subsequent submission could be made when the data are mature. The sponsor said final survival data will be available 18 to 24 months from now, and FDA statisticians predict a 45 percent chance of seeing statistical significance in this data.
Additionally, there was a subset of patients in the trial that the committee noted as potentially good candidates for the Yondelis combination: those sensitive to platinum therapy, but intolerant of it.
There was disagreement, however, over how large an opportunity that would be. The company suggested that could mean as much as 20 percent of the population, but panelists suggested these patients represent a very small part of the ovarian cancer population - perhaps around 5 percent. The committee also noted that the subset would need to be prospectively studied for the benefit to be confirmed.
Beyond the fact that the Yondelis combination showed a relatively small effect on a surrogate endpoint with a high toxicity profile, committee member David Harrington, Dana-Farber Cancer Institute, said, "more important to me is that we are all speculating on a subset of the patients in this trial, who may be anywhere from five to 20 percent, who may or may not benefit from the therapy, and that has not been tested in any sort of prospective way. It's an important difference in that subset if we can rely on it and if it translates into a survival difference," but "this leaves as much on the table as it solves."
Doxil Offers Modest Gain, Significant Cost
Though there is still hope that mature data will show a survival advantage for Yondelis, the OS data in the Doxil breast cancer trial are mature and showed no statistically significant advantage for the combination of Doxil and docetaxel compared to docetaxel alone. In fact, a slight disadvantage was shown: the median survival was 20.5 months for the combination compared to 20.6 months for single agent docetaxel.
An improvement in TTP was reported - the median TTP was 9.8 months for the combination versus seven months for the docetaxel monotherapy arm - but the reliability of those results was questionable because of different reports: the independent review committee reported a difference in TTP of 2.8 months, while the trial investigator reported 1.5 months.
FDA and ODAC were also concerned about the applicability of the results to U.S. populations. After accruing poorly in the U.S., the trial was conducted mostly in Russia and Eastern Europe, where prognostic tests for HER2 and ER/PR status - which determines first line MBC treatment in the U.S. - were not conducted for many subjects.
Also calling into question the relevance of the data to a U.S. population was the choice of control arm. FDA advised the sponsor to consider a docetaxel + capecitabine combination for the control arm since it is associated with a three-month survival advantage, but the sponsor "disregarded the advice," Pazdur said, and went with docetaxel monotherapy as the control.
On top of the lack of survival benefit and issues regarding the reliability and applicability of the data, the combination was so poorly tolerated that 34 percent of subjects discontinued Doxil due to adverse events, mostly for hand-foot syndrome and stomatitis. There was also an increase in pneumonia in the Doxil arm - 75 percent of which were serious adverse events and one of which was fatal during cycle 1 of the treatment.
ODAC voted unanimously against Doxil for breast cancer, citing the increased toxicity, the modest TTP effect, and questionable applicability to U.S. populations - all of which could possibly have been overlooked if the combination had demonstrated a survival advantage.
"Survival trumps everything, and this has failed to show any utility in that regard," said committee member Ronald Richardson, Mayo Clinic.
Panel member Carmen Allegra, University of Florida, added that "Certainly there are other available doublets that may be as toxic as this one, but at least they have a survival advantage."
Hopeful
|
Linear Mode
