Skip to main content

Biotech Medarex Misses the Mark

The case looks weak for a positive phase III cancer drug trial.

BOSTON -- I'm not one to shy away from controversy, so I'll get right to the point: I've spent a few days digging into



skin cancer drug MDX-010. What I found wasn't very impressive.

I don't believe the current phase III studies are going to yield positive data strong enough to warrant the drug's approval.

The drug, which also goes by the name ipilimumab, just isn't very potent when used by itself (as monotherapy) in patients with advanced melanoma, or skin cancer. That's the conclusion I draw from a look at previously conducted clinical trials.

If a drug fails to impress in phase II trials, it's hard to have much confidence in a successful outcome from phase III trials. That's conventional wisdom in the biotech sector -- and it usually serves investors well.

Medarex shares had been moving strongly in the past three months as investors await the results from a trio of phase III monotherapy studies of MDX-010 in patients with advanced melanoma. The stock closed Friday at $15.55, off more than 6%.

The data are expected before the end of the year, hopefully in September or October.

These are registration studies, which means that if they're successful, Medarex and partner

Bristol-Myers Squibb

(BMY) - Get Free Report

say they will file for the drug's approval with the Food and Drug Administration.

Obviously, the results are very important for Medarex in the short term. The company does have a deep pipeline of drugs in development, but has yet to advance any to approval and the marketplace.

Because Medarex has so many drugs in development, the company is in a much stronger position than



, for example, which is totally reliant on the prostate cancer drug Provenge for its fortunes.

If MDX-010 fails in melanoma, Medarex has a lot to fall back on -- that's important because while the stock could suffer in the short term, there are ample opportunities later for Medarex to succeed.

Study the Study

The focus for Medarex investors right now, however, is a 150-patient study of MDX-010 as monotherapy in patients with advanced metastatic melanoma. (There are two companion studies that will report data at the same time, but for simplicity sake, it's OK to focus on just this one.)

These are very sick patients who have already failed a first line of treatment. Life expectancy is short and there are no approved drugs to treat them.

The high unmet medical need and dearth of good drugs in the second-line setting for metastatic melanoma works in MDX-010's favor. The regulatory bar set by the FDA should be somewhat lower, which means the drug probably doesn't have to produce stellar results in order to get approved.

What sort of efficacy is good enough? In the past, an objective response rate of 10% (translation: 10% of patients report tumor shrinkage of at least 50%) has been deemed a minimal efficacy standard for cancer drugs in similar situations.

More recently, FDA officials have been growing skeptical of such low response rates, especially in the absence of other data showing how patients might be benefiting. This year, especially, regulators have taken tough stances on cancer drug approvals (Dendreon and

GPC Biotech


are just two examples.)

The agency seems to want very strong evidence of efficacy, including concrete data pointing toward a survival benefit.

Medarex hasn't given clear guidance on a response rate hurdle for the MDX-010 melanoma trial. But I'm assuming that an MDX-010 objective response rate in the mid-teens (let's say 14%-16%) puts the drug in a good position for approval.

A response rate in the low-teens (11%-13%) puts the drug's chances for approval on the fence. A response in the single-digits, or even 10%, won't be good enough.

You'll notice that my efficacy thresholds seem high. That's because Medarex is using a less stringent measure of tumor shrinkage for its MDX-010 study. The so-called WHO classification for measuring tumor response (the one being used by Medarex) generally results in higher response rate numbers than the so-called RECIST criteria, which is more commonly used.

Simply put: The response rates to be reported by Medarex from the MDX-010 study will be inflated a bit by using the WHO measurement system, probably by two or three percentage points.

The problem is that when you dig into previous melanoma studies for the drug, the objective response rates reported (all using RECIST criteria, by the way) are in the 5%-10% range. That's not a very robust response, and it doesn't bode well for the upcoming phase III study.

A Closer Look

Some details:

A phase I/II MDX-010 study presented at ASCO in June reported a 5% objective response rate in patients who had failed a previous treatment for melanoma. These patients were given various dose levels of MDX-010. A subset of 23 patients in this study dosed with 10 milligrams of MDX-010 (the same dose being used in the phase III study) performed marginally better -- a 9% response rate.

In a similar study presented at the 2005 ASCO meeting, 37 melanoma patients given MDX-010 as monotherapy managed just a 5% objective response rate. This study used a lower dose (3 mg) of MDX-010.

Another study of 56 melanoma patients published in 2005 also used a 3-mg dose of MDX-010, yielding an objective response rate of 13%. But some of the patients in this study hadn't failed previous treatment, so when they're excluded from the analysis, the response rate drops to 8%.

This isn't to say that MDX-010 doesn't have previous data that works in its favor. For one thing, the drug does seem to produce a high level of stable disease, which translated, means tumors that shrink minimally or just stay the same size. Moreover, the duration of the response seen with MDX-010 can be quite long.

When Medarex releases data from the phase III trials, the company will likely include stable disease patients together with those reporting an objective response. This will produce a "disease control rate," or DCR, that will probably look quite robust.

If significant tumor shrinkage (i.e., an objective response) is a best-case scenario, stable disease is a second-best outcome for cancer patients, because a tumor that isn't growing can also help patients live longer.

A key aspect of the Medarex bull case for MDX-010 is that while the objective response rate might be low, the DCR rate and the duration of response will be high. These latter data, the bull argument goes, will be enough to convince the FDA to approve MDX-010, especially since this is a desperate patient population lacking good treatment options.

Tracking Survival Rates

But I don't buy it for a couple of reasons. For starters, stable disease might be suggestive of a survival benefit, but the FDA will want to see actual survival data.

The phase III studies for MDX-010 aren't set up to collect survival data, at least not in a robust and objective format, because they're designed with a single arm only (i.e., no patients are being given a placebo or best supportive care, by which it could be determined if MDX-010 is really helping melanoma patients live longer.)

Medarex bulls point to the fact that the current MDX-010 studies are being conducted under a Special Protocol Assessment -- essentially an agreement between Medarex and the FDA that the trial design is robust enough to warrant a drug approval (contingent on posititve data, of course.)

That's true, but there's a lot of wiggle room in so-called SPAs, by which the FDA can change its mind or interpret things differently. The blowup of GPC Biotech and its prostate cancer drug satraplatin in July is a prime example.

Medarex is a good company with a deep and diversified pipeline. It has solid partnerships, and MDX-010 is an interesting drug that may end up succeeding against melanoma when combined with other drugs (as some other data suggest.) There are other, ongoing phase III studies of MDX-010 in melanoma that are tracking survival, but they won't be completed for some time. The drug is also being tested against prostate cancer.

I say that because I don't want to come across as a total curmudgeon. But when it comes to this most immediate and important study, which uses MDX-010 as a monotherapy in melanoma, the evidence on hand simply doesn't provide enough confidence to predict a successful outcome. That's especially true given the current mood at the FDA, which is taking a hard line on all drugs, even those for advanced cancer.

There's a 23% short interest in Medarex right now, so I know I'm not the only one skeptical about MDX-010's near-term prospects. We'll know soon enough whether the skeptics like myself are right.

Adam Feuerstein writes regularly for In keeping with TSC's editorial policy, he doesn't own or short individual stocks, although he owns stock in He also doesn't invest in hedge funds or other private investment partnerships. Feuerstein appreciates your feedback;

click here

to send him an email.