Sorry!! The article you are trying to read is not available now.
Thank you very much;
you're only a step away from
downloading your reports.

Run Away From Genitope


The FDA will not approve MyVax based on the immune responder analysis because it is contrary to the rules it has set out for drug approvals.


Genitope (GTOP) released long-awaited data from its MyVax trial in blood cancer after the bell last night. The trial failed, despite the fact management had the temerity to call it a success.

I actually received more than a little e-mail on Genitope over the last few months. The company was periodically releasing unblinded data and implying there was a substantial treatment effect for its drug. The reasoning was that if the unblinded median benefit was in excess of the "historical" median benefit, then the reason must be the drug is working.

I've seen this sort of thing before, most notably with Cell Therapeutics' (CTIC) cancer drug Xyotax. That trial was also a failure (unless you were being treated in Russia, but that's another story). Historical controls are notoriously unreliable, which is why FDA guidance states it won't accept historical controls as a basis for drug approval.

In a clinical trial, everyone lives longer. The standard of care cancer patients receive in a clinical trial is superior, if only because of regular checkups and increased attention to treating side effects. One also shouldn't, especially in cancer treatment, underestimate the placebo effect. Cancer is one disease where study after study proves positive outlooks and aggressive attention to health extends life. The 1-2% of cancer patients in the US who seek out and enroll in a clinical trial are more likely to be positive and aggressive compared to the average cancer patient, meaning they'll live longer than the "average" person in a historical control.

Stay informed, stay successful. Get your free two-week trial of the Buzz & Banter and get real-time information, observations and ideas from our Minyanville professors.

The other problem with historical controls is they are typically based upon old data. This is especially confounding in a fast-moving disease indication like blood cancer. Even though many blood cancers have not had new drugs approved recently, the standards of care change very rapidly as doctors mix and match different therapies in order to boost response rates.

When someone (management team, analyst, or stockbroker) pitches a biotech company to you using an argument about how the unblinded data exceeds historical controls, the general rule is to run away. The reality has rarely (if ever) been huge efficacy from the drug being studied. It's usually (always?) that the company didn't enroll the patients it intended or the historical outcomes cited are inaccurate or not applicable.

The sad thing about the MyVax data is the drug appears to work in some subset of patients. A strong, randomized Phase II program would likely have prevented the failure announced yesterday. This is a common theme in oncology, where Phase II trials are eight times less likely to be randomized than in other disease areas, leading to more than twice as many clinical failures.

Specifically with MyVax, Genitope found that patients who generated an immune response to this active immunotherapy had a statistically and clinically significant benefit versus patients who did not. To those less familiar with how the oncology division at the FDA works, this might sound a lot like a successful trial.

Unfortunately, the trial was not prospectively designed to ask that question. The conservative biostatisticians currently dominating oncology drug approvals will ultimately throw out those results on that basis. MyVax may indeed work, but the company will be forced to run another clinical trial to prove it.

Genitope bulls are probably already sharpening their pencils to write me notes telling me the trial was indeed prospectively designed to investigate immune response. After all, Genitope management went out of its way to say the tracking of immune response was a "pre-defined endpoint" collected as per the protocol.

There are endpoints and then there are endpoints. The following delves into the arcane art of biostatistics, so forgive me if it gets a little murky.

The priests of biostatistics brought their tablets down from their ivory towers and decreed a trial can only study one thing at a time. That means there must be a primary endpoint specified for all trials. Importantly, according to these ivory tower denizens, only the specified primary endpoint matters. If it fails, none of the other endpoints makes any difference. They become "hypothesis generating" and require a subsequent clinical trial to prove they are true.

Since the immune responder analysis was not pre-specified as the primary endpoint in the MyVax trial, it means the trial is a failure. That's the end of the discussion as far as the priests of biostatistics are concerned.

People who actually treat patients asked the priests of biostatistics if they could look at more than one thing in a trial – to have more than one primary endpoint. The priests of biostatistics retreated to their towers, contemplated this heresy, and returned with an edited tablet granting the ability to have multiple primary endpoints, but at a price.

The priests of biostatistics pray at the altar of alpha. In layperson's terms, a clinical trial is a success when there is a 5% or less chance of the result being due to chance. This is reflected in the "p-value" cited in clinical trial reports as p = 0.05. There is only so much alpha to go around in a clinical trial, according to the priests of biostatistics.

On the edited tablet, the priests of biostatistics said clinicians treating patients could indeed have more than one primary endpoint. The catch was that they could have no more of the precious alpha than p=0.05 so clinicians would have to divide it up amongst the endpoints. In layperson's terms, this means if there are to be two primary endpoints then the alpha must be divided between them. Instead of the hurdle for success being p=0.05, the hurdle for success is now significantly higher at p=0.025.

Genitope did not "spend" the alpha necessary to make the immune responder endpoint a real primary endpoint for purposes of drug approval. While success on this endpoint is probably a real indication the drug works in immune system responders, the priests of biostatistics will say only that the result is "hypothesis generating" and must be confirmed in a subsequent clinical trial.

Genitope's management suggested the benefit seen in the immune system responders was robust and highly statistically significant. So what happens if the p-value for that group is less than p=0.025 – less than the p-value implied if the company had indeed specified it as a co‑primary endpoint?

Nothing. The trial is still a failure.

The first commandment of the priests of biostatistics is that everything must be prespecified before the trial starts. If it isn't, then it doesn't count. The p-value in the MyVax immune responder group could be p=0.000001 and this would not change a thing – according to the ivory tower biostatisticians.

You can claim this defies logic and you would be right. It is only logical in the ivory tower world of conservative biostatistics. You can also claim this defies common sense. You would be certainly right about that. You can angrily claim this prevents a safe and effective treatment from reaching cancer patients. You'd be right again.

But this is the game. The conservative biostatisticians who currently dominate oncology drug approvals don't care that their rigid views harm patients and frustrate doctors. They see themselves, as bureaucrats often do, as defenders of the faith. It is sad and makes more than a few people angry enough to lobby Congress to force change.

As an investor, I have to separate those feelings from my portfolio decisions. The FDA will not approve MyVax based on the immune responder analysis because it is contrary to the rules they have set out for drug approvals. It is as simple as that.

One might hold out hope that Congress could shine a light on the sorry state of oncology drug approvals in a way that might make a difference for Genitope's MyVax. This is a low percentage bet, particularly with Genitope's cash burn rate implying a drop-dead date in the middle of 2008.

How low? Dendreon's (DNDN) data is stronger statistically than the MyVax data primarily because it does not rely on subgroup/responder analysis to drive efficacy. Even with that advantage, my firm places only a single-digit chance Congress could come through and enforce some common sense on the FDA in time to hasten Provenge's approval – and I characterize those odds as generous.

There are something around 250 development-stage biotech stocks to invest in. Ultimately, the cards are so stacked against MyVax based upon the data released yesterday that there are easier places to make money. That's the decision an investor must always keep in the front of his or her mind, even if in the back of my mind I'm angry and frustrated about how silly and unfair the priests of biostatistics have made cancer drug approvals.

Position in DNDN.

The information on this website solely reflects the analysis of or opinion about the performance of securities and financial markets by the writers whose articles appear on the site. The views expressed by the writers are not necessarily the views of Minyanville Media, Inc. or members of its management. Nothing contained on the website is intended to constitute a recommendation or advice addressed to an individual investor or category of investors to purchase, sell or hold any security, or to take any action with respect to the prospective movement of the securities markets or to solicit the purchase or sale of any security. Any investment decisions must be made by the reader either individually or in consultation with his or her investment professional. Minyanville writers and staff may trade or hold positions in securities that are discussed in articles appearing on the website. Writers of articles are required to disclose whether they have a position in any stock or fund discussed in an article, but are not permitted to disclose the size or direction of the position. Nothing on this website is intended to solicit business of any kind for a writer's business or fund. Minyanville management and staff as well as contributing writers will not respond to emails or other communications requesting investment advice.

Copyright 2011 Minyanville Media, Inc. All Rights Reserved.

Featured Videos