Editor's note: In an earlier version of this column, the item on MELA Sciences contained erroneous information as provided by the U.S. Food and Drug Administration. The agency does in fact enter into formal or binding agreements for experimental medical device clinical trials. MELA's public statements to that effect are correct. The story has been corrected.
) -- I will not complain, but keeping up with the volume of your email and tweets is quickly becoming a Sisyphean task. To help catch up, here's a special Thursday edition of the Biotech Stock Mailbag.
Tiffany M. emails,
Ziopharm is launching a pivotal phase III study of its experimental drug palifosfamide in sarcoma at a time of great uncertainty and debate over how to define and measure clinical benefit in cancer drug studies.
I spoke with Ziopharm CEO Jon Lewis Wednesday morning about the study and my sense is the company (with the help of outside experts) has done an admirable job navigating some tricky waters with the palifosfamide study to make it as "approvable" as possible. Does this eliminate all the regulatory risk and worry? No way.
Food and Drug Administration statute has long granted accelerated approval to cancer drugs based on clinical trials that measure a drug's ability to delay tumor growth or worsening. This "progression-free survival," or PFS endpoint, was long considered to be a surrogate measure for survival. In other words, if a drug lengthens the time before a patient's tumor starts growing again, that patient would also likely live longer.
Drug companies like running cancer drug studies with PFS endpoints because data can be collected and analyzed faster than from studies using overall survival as a primary endpoint.
Yet this correlation between PFS and survival benefits in cancer studies is now being challenged. On Tuesday, an advisory panel recommended that the FDA revoke the accelerated approval of
Avastin as a breast cancer treatment because follow-up studies failed to demonstrate a benefit for patients.
In 2008, the FDA granted Avastin accelerated approval in breast cancer based on a single study showing the drug delayed tumor growth. It was hoped that follow-on studies of Avastin would both confirm this PFS benefit as well as document that breast cancer patients were living longer.
Those subsequent breast cancer studies of Avastin failed to accomplish either objective. In fact, the new studies demonstrated a smaller benefit in delaying tumor growth and no survival benefit for patients. Moreover, patients treated with Avastin were at greater risk for side effects.
These findings compelled the FDA advisory panel Tuesday to recommend that the agency take away Avastin's approval in breast cancer. It also raised tons of questions and concerns for other companies with plans to seek FDA accelerated approval based on PFS studies.
The FDA has not said that PFS is no longer an approvable endpoint for cancer drugs, and there are examples of drug studies where a PFS benefit does also confer a survival benefit for patients. Still, the "PFS or survival endpoint" controversy is ratcheting up, and companies using PFS endpoints to get cancer drugs approved today are facing a barrage of investor questions and concerns.
Which brings us back to Ziopharm because the just-started palifosfamide phase III study in first-line metastatic sarcoma uses PFS as its primary endpoint for accelerated approval. Moreover, the FDA and Ziopharm could not reach agreement on a special protocol assessment for this study.
Yeah, it's a worry but not necessarily a reason to panic. The palifosfamide study is designed with enough patients and statistical power so that a three-month improvement in tumor worsening (PFS) should also be enough to demonstrate a survival benefit for patients, according to Ziopharm CEO Lewis.
"If PFS is positive, it is very likely, knowing what we know about the nature of this disease and the design of our study, that the other
overall survival will be also be positive," Lewis told me in a Wednesday morning telephone call.
But what happens if palifosfamide delays tumor growth but doesn't prolong survival? Would this prevent Ziopharm from seeking the drug's approval? Or, if the company pushed ahead with a filing, would the FDA be more likely to deny approval?
"We believe that palifosfamide is approvable based on a PFS benefit alone," said Lewis. That opinion, he adds, is based on feedback from sarcoma experts, as well as regulatory consultants who point out that PFS is still very much an in-vogue and adequate endpoint for drug approvals in Europe.
Understood, but Jon, if you don't hit overall survival in the palifosfamide study, you know that investors are still going to freak out, right?
Lewis laughed, uncomfortably. I took that as a sign of agreement.
The palifosfamide data from a phase II study in sarcoma, presented at the ASCO meeting in June, looked really good. That should give investors confidence that Ziopharm can replicate the strong results in the phase III study. I do worry about the PFS-to-survival issue -- and a lot of investors will, too -- which does create an overhang of added risk to the Ziopharm stock story. Ziopharm also needs 18 months to enroll the palifosfamide phase III study, so don't expect data until 2012.
I'm dipping deep into the Mailbag for an email from Rebecca S. sent in late March.
"Hi Adam. I love your column, and no matter how tough it is to hear sometimes. (As an investor and caregiver to husband with lymphoma, the truth is the truth. What is going on with BioVest International (BVTI.PK) ? After the failure of Genitope and Favrille, we thought the vaccine route was over, but BioVest seems to be hanging in there. The stock has risen from less than 50 cents to $1.65 in recent weeks. Is there any promise to this company or is this just dumb speculation."
Any investor interest in BioVest and its lymphoma vaccine BiovaxID is simply dumb speculation. Sorry. I hope your husband is doing well, but I cannot be optimistic at all about BiovaxID's chance for approval.
Common sense should tell you I'm right. The initial "positive" results from the phase III study of BiovaxID in follicular non-Hodgkin's lymphoma patients were released in 2008. In the spring of the following year, the data from this phase III study were presented during the plenary session of the American Society of Clinical Oncology annual meeting. How and why ASCO allowed the BiovaxID data to be highlighted at the most important and closely watched session of its annual meeting still escapes me, but there it was.
If the BiovaxID data were credible and supportive of approval, someone, some company, would have swooped in and either licensed the vaccine or acquired BioVest outright. During this entire two-year period, BioVest was in deep financial trouble. In fact, I believe, the company was in bankruptcy. That means BiovaxID would have been available cheap!
Yet, here we are, three years since the first BiovaxID data were announced and no one is interested in the vaccine. No partnerships. No acquisition offers. Nothing. BioVest is now emerging from bankruptcy and plans to seek approval for the vaccine on its own. Common sense should tell you that the reason Biovest is operating solo is because BiovaxID doesn't really work. Or, at least, data from the pivotal, phase III study doesn't provide the evidence to prove the vaccine works. That is the most likely verdict from the FDA if and when Biovest files for approval.
If you want a detailed explanation for why the phase III study data lacks credibility, write me back and I'll take it up in a future Mailbag. For now, I'll just say that the disease-free survival benefit that favored BiovaxID in the study was skewed heavily by the way patients were randomized and included/excluded from analysis, as well as the fact that the study never fully enrolled.
, its obesity drug Qnexa, and last week's FDA advisory panel (which voted against the drug's approval) are still hot-button issues for readers. Scott B. writes:
"Adam, I'm a Vivus shareholder and thank you for your fair approach to everything going on with Qnexa. My question is do you think the FDA will disregard the special protocol assessment (SPA) that Vivus worked out for the Phase III trials? If the FDA wanted two or three years of data then it could have set that up in the SPA, right? Maybe the FDA should state they will make a decision with the available phase III data in addition to the upcoming two-year data in September. I see a six-month delay, maybe."
Vivus and the FDA did negotiate a special protocol assessment (SPA) covering the design of the phase III Qnexa studies, including the need for one year of safety data. What investors may not understand, however, is that an SPA is a fungible document. If circumstances change, the SPA changes or becomes invalid.
Vivus and the FDA aren't willing to talk specifically about the Qnexa SPA, but my guess is that the outcome of the panel vote forces the FDA to rethink its safety data requirements for Qnexa and will delay the weight-loss drug's approval beyond its Oct. 28 approval decision date.
It seems pretty clear going in that FDA believed one year of safety data on Qnexa was going to be enough, probably because the two drugs making up Qnexa -- phentermine and topiramate -- are already approved.
What the FDA may not have anticipated when it agreed to the Qnexa SPA is that the experts on its advisory panel would be more conservative and risk-averse than its own drug reviewers. Usually, it's the other way around, but these outside experts had just been through two days of intense testimony and controversy over the heart risks of
If these panel members gave a collective thumb's up to Qnexa's safety on just one year of data, was there a risk that some unforeseen side effect would come back to haunt their decision in two or three years? I bet this question weighed on their deliberations.
The FDA is not obligated by statute to abide by the recommendations of its advisory panels. I have witnessed cases in which the FDA has ignored the advice and either approved or rejected drugs against the wishes of its outside advisors. I don't believe that will happen with Vivus because obesity, while a serious disease, is not immediately life threatening or without alternative treatment.
FDA and Vivus thought that one year of safety data for Qnexa was enough. Both were wrong. As a result, the SPA must change. Vivus says it will have two year's worth of Qnexa safety data later this quarter. Hopefully, that data support the drug's safety and approval. If so, FDA will review the new data, but a final approval decision on Qnexa isn't likely to come until next year at the earliest.
For more on the
FDA and SPAs, read a primer on the topic
I wrote in the Nov. 20, 2009, Mailbag.
Patrick C. asks,
"In relation to your most recent article about MELA Sciences (MELA) , don't your concerns about how the company's trial design fly in the face of the fact that the FDA entered into a special protocol assessment (SPA) for its trial? All indications point to the fact that MELA did everything by the FDA's own guidance. Your comments would be appreciated."
Another SPA question.
The FDA does not have an SPA process for medical devices, FDA spokesman Dick Thompson confirmed to me via email Wednesday. The FDA does have a mechanism, known as an "agreement meeting" in which a medical device maker can get FDA's agreement on an investigational plan or clinical trial design, he said, adding that, "It is a little used and little known process."
According to FDA law, this agreement between the FDA and the medical device maker is "binding" but can be changed by FDA when there is a "substantial scientific issue essential to determining the safety or effectiveness of the device."
In a recent press release, MELA said, "the company and the FDA entered into a
binding protocol agreement
to stipulate the study design, including the sensitivity and specificity endpoints that should be used to determine the safety and effectiveness of MelaFind."
In the most recent 10-Q, MELA said the MELAFind study was conducted "under the auspices of a
Emphasis mine but capitalization MELA's.
And in last year's 10-K, the FDA agreement was described by MELA as a "binding Protocol Agreement."
MELA has not disclosed the specifics of its binding agreement with FDA. In March, MELA announced that FDA sent a letter to the company stating the application for MELAFind was "not approvable" at that time. MELA later submitted responses to FDA questions and the agency scheduled an advisory panel meeting in November to review the MELAFind application.
Tomorrow, another Mailbag.
-- Reported by Adam Feuerstein in Boston.
Follow Adam Feuerstein on
Adam Feuerstein writes regularly for TheStreet.com. In keeping with TSC's editorial policy, he doesn't own or short individual stocks, although he owns stock in TheStreet.com. He also doesn't invest in hedge funds or other private investment partnerships. Feuerstein appreciates your feedback;
to send him an email.