When is a trial "failure" really a failure?

"When the FDA says so," is the obvious and most important answer, but data last week from a Phase II study by Corgentech Inc. with its previously successful pain drug, ALGRX 4795, brought up questions.

In patients undergoing groin-hernia repair, the ALGRX 4795 trial showed no significant differences in pain scores between patients undergoing groin-hernia repair who were given the drug and those given placebo. The 41-patient trial measured efficacy of the compound - dripped directly into the wound for one minute just before the wound was closed - at one week and four weeks after surgery.

But all patients in both the treatment and placebo arms got paracetamol (acetaminophen) and non-steroidal anti-inflammatory drugs (ibuprofen) for a week after surgery, which is a mandated standard of care in Denmark, where the trial was conducted. That, along with lower thresholds of pain to begin with, could have skewed the results, said George Fulop, analyst with Needham & Co., in a research report.

Corgentech gained the compound, a vanilloid-1 receptor (VR-1) agonist based on capsaicin, in the merger with AlgoRx Pharmaceuticals Inc. Capsaicin is the substance that confers on chili peppers their distinctive heat, and plant extracts have been used as topical pain treatments for centuries. In the same way that people who often eat Mexican food can find themselves less likely to sweat, gulp water and gobble soft tortillas to undo the effects of chili peppers, pain patients whose VR-1 receptors are overstimulated can end up desensitized.

Acting as a C-neuron anesthetic, ALGRX 4795 has worked in Phase II studies against pain from elbow tendonitis, as well as pain in patients undergoing bunionectomy and those with the neuropathic foot condition known as Morton's neuroma. Phase II studies in total knee replacement and cholecystectomy (gall bladder removal) have completed enrollment, with data due in the second quarter of the year.

"A lot of people are asking why they even did the [hernia-surgery] trial, because it was such low pain," Fulop told BioWorld Financial Watch. "I can't answer that question. It must have looked good when they did it."

The company couldn't be reached, but another firm working in the same area but with a topical compound is NeurogesX Inc., which calls their compound the transient receptor potential vanilloid 1 receptor, or TRPV1.

Keith Bley, senior vice president of nonclinical research and development at the firm, declined to call the Corgentech hernia-surgery study a failure.

"Patient numbers were small, and if the surgeons are doing their job right, they don't induce much pain" for the drug to have an affect on, he noted.

NeurogesX's product is called NGX-4010 (brand name Transacin), a high-concentration form of synthetic capsaicin, known as trans-capsaicin, applied directly to the site of pain via a rapid-delivery dermal application system. Near the end of last year, NeurogesX raised $20 million through a private equity financing to finish pivotal trials with Transacin in HIV-associated neuropathy and post-herpetic neuralgia.

"There are quite a few relatively low concentration creams" based on capsaicin for pain, Bley said, but patient compliance is low because of "pungency or stinging," and the creams must be administered several times per day, whereas Transacin is a single procedure, done in a doctor's office.

Antagonists of the receptor also are being explored by pharma firms, he told BioWorld Financial Watch, and those "have gotten a lot of play" with such conditions as inflammatory and irritable bowel disease.

Allan Basbaum, professor and chairman of the anatomy department at the University of California at San Francisco, has served as an adviser to NeurogesX and is editor-in-chief of the journal Pain. He said whether the other pain drugs given with ALGRX 4795 muddied the trial results "depends on how you think the [Corgentech] drug is working. It wouldn't interfere at all with acetaminophen, and ibuprofen is working in the periphery to prevent sensitivization of the pain fibers. If you destroy the pain fibers, it's moot."

Also working to develop pain drugs that target the receptor is Neurogen Corp., which has a major deal with Merck & Co. Inc., formed in late 2003. The partners in February started Phase I trials with NGD-8243. As part of that collaboration, Merck made a $15 million license fee payment to Neurogen and bought $15 million of Neurogen stock, agreeing to pay $12 million in research funding and license maintenance payments, with another $118 million possible in milestone payments. Renovis Inc. has a pharma-backed VR1 pain program, as well. In May, the firm entered a potential $187 million preclinical research deal with Pfizer Inc.

Corgentech, though, occupied last week's headlines. After the AlgoRx merger, the company's technology value is about $65 million, Fulop wrote in his report, "despite a Phase III and a Phase II compound in development in important areas of unmet medical need."

The Phase III product is ALGRX 3268, expected to be the subject of a new drug application this year. A topical local anesthetic, the compound is designed to reduce pain from needle-sticks and intravenous line placements through its use of a needle-free injection system to accelerate lidocaine particles into the epidermis in order to anesthetize nerves.

"We believe the company is undervalued at its current enterprise value, and we are maintaining our 'buy' rating," with the target price under review, Fulop wrote. As for ALGRX 4795, he expects better news ahead.

"The mechanism of action is sound," he said. "That's what I fall back on." n