The flimsy case for a miracle ALS drug collapses under expert scrutiny

By his own account, Steve Perrin didn't spend much time considering the potential of a drug like GM604 after a low-profile biotech named Genervon--which has claimed biblical prophecy as an inspiration--announced the "data" it had collected in a study that recruited 12 patients with amyotrophic lateral sclerosis. After all, nothing's been published in any scientific journal since the top-line results were posted last fall. And the study was so tiny that Perrin, president and chief scientific officer of the ALS Therapy Development Institute, considered it nothing much more than a simple safety trial to prove the drug held no hidden dangers.

Given the varying ways that individual ALS patients progress, gradually growing paralyzed and dying in 3 to 5 years, he reasoned, the results from 8 patients after only 12 weeks of treatment, with four patients reserved for the placebo arm, couldn't provide a reliable look at efficacy.

Steve Perrin

But the biotech also touted the amazing story of one compassionate use patient who had made a remarkable personal improvement in swallowing capacity after he was given the drug and after the disease had made him a quadriplegic. And there was an equally amazing set of data provided to suggest a radical improvement in the decline of forced vital capacity, or FVC, among the patients in the drug arm.

According to Genervon, the FVC rate declined 5.6% among patients taking the drug, compared to a dramatic 22% decline for placebo.

The single case study, otherwise known as an anecdote, along with the "positive" FVC data, created the patient population's hallelujah moment.

Petitions followed, gathering more than 100,000 signatures, demanding that the FDA approve the drug now based on the results Genervon had promoted. Protests were mounted, with Genervon issuing releases in which it beat the drum for accelerated approval.

A few weeks ago, in a statement to "ALS patients and caregivers," the company ruled out a compassionate use program, adding that most of today's patients will be dead before Genervon can complete any late-stage study required by regulators.

"Even though the FDA has promised to help Genervon expedite the approval process for a Phase III trial, it would still take at least 3 years for GM604 to reach NDA," stated the company, which is run by Winston Ko. "This means the majority of this generation of ALS patients would not survive to try GM604."

The message came through loud and clear: If patients and their families want access to this drug, the FDA has to approve it immediately.

Driven by the massive popular outpouring of support for dying, desperate patients, a string of prestigious media outlets like the Washington Post and The Guardian published accounts that included both sides of the argument, without any examination of the data. The din grew louder after one ALS patient included a heartfelt plea in a column in the New York Times. Better to try something, anything, patients reasoned, than wait to die. And the dogs of war over access to experimental drugs were unleashed.

A skeptical Perrin, who has designed studies for ALS drugs and partnered with biotechs in the field, decided to take some time to scrutinize the results that are available, modeling the data to see how well it stands up against historical data for a much larger group. And there's even less than meets the eye after you consider some of the efficacy results being advertised as a basis for an accelerated approval.

In a blog post, Perrin notes that he turned to the public PRO-ACT database--which gathered 45,000 FVC measurements from 7,854 ALS subjects--to find a relevant group of patients that could be used to compare the efficacy of GM604. He ended up with data from 777 ALS patients covering a period of 70 to 120 days.

During that time there was a mean change from baseline of -4.33% in the rate of FVC--nothing even close to the figure that Genervon provided, and slightly better than the result the biotech provided for its drug.

"Based on the modeling demonstrated here from a very large patient cohort it does not appear plausible that the cohort of patients enrolled in the Genervon trial would represent a "normal" cohort of ALS patients," writes Perrin. "Therefore based on the model I continue to be very skeptical that there is any way that GM604 could have provided any therapeutic benefit in this small of a trial with this short of an exposure."

There are ways to explain the data, he adds. The results could have been collected improperly, though he doubts that. (The two investigators for this study, Hiroshi Mitsumoto and Merit Cudkowicz, work at at Columbia Medical Center and Mass General.) Fast-progressing patients could have been enrolled in the study, which would skew the results. Or maybe there was a problem with the way the patients were screened to make sure their FVC capacity was 65% or greater.

The FDA will only confirm that it provided orphan drug status for GM604, which is public record, declining to make any comment about Genervon's case, which is SOP at the agency. 

Perrin, though, concludes emphatically that there's a lot more long-term safety and efficacy data that must be gathered before a drug like this can be considered for an approval. And in the meantime, he notes, patients and families should consider that anyone taking this drug would be excluded from a study of a potentially much more promising therapy (and there are several in the pipeline).

"A potential treatment," he concludes, "that might actually slow down disease in ALS."

-- John Carroll (email | Twitter)