BioWorld International Correspondent

PARIS - An independent panel set up by NicOx SA to review the results of the Phase II endoscopic trial of AZD3582, its COX-inhibiting nitric oxide-donating drug carried out by AstraZeneca plc, concluded that "a number of failures and deficiencies affected the outcome of the study."

In particular, it found "major protocol violations" in about 30 percent of the patients (310 out of 970) and concluded that AstraZeneca was not justified in arriving at a negative evaluation of the product.

London-based AstraZeneca is co-developing AZD3582 under agreements signed with NicOx, of Sophia-Antipolis, in 1998 and 2002. The Phase II trial was designed to test the compound for the treatment of acute and chronic nociceptive pain. After communicating the preliminary results to NicOx, the two companies issued a joint statement in mid-February announcing that the trial "did not reach its primary endpoint with respect to gastrointestinal ulcers" but did meet the majority of secondary endpoints.

Now, NicOx is questioning the validity of the trial. After requesting the full data file from AstraZeneca, NicOx asked an independent panel of experts specialized in gastroenterology, biostatistics and clinical methodology to review the trial's clinical protocol, the available data and AstraZeneca's analyses and conclusions.

The most serious protocol violation it found was an excessive delay between last treatment and endoscopy, which averaged 5.5 days (and ranged from three to 29 days), whereas it should have been no more than two days, NicOx corporate relations director Sylvain Goyon said. Due to the small size of ulcers stipulated in the protocol (3 mm), this delay "is likely to have affected the healing process and consequently the final ulcer incidence response," NicOx said.

In effect, the rate of response was only 13.7 percent for the reference compound, whereas a rate of 20 percent to 25 percent was expected.

Another major violation mentioned by the panel was the exposure of patients to nonpermitted drugs, including proton pump inhibitors, H2 antagonists and other non-steroidal anti-inflammatory drugs. Furthermore, it suggested that patient compliance was "inadequate for drug treatment" and pointed out that there was "a large variability in clinical response" between countries. The panel also questioned the quality of the study due to the "lack of uniform evaluation and of recorded endoscopic examination findings," and noted the "lack of a bioequivalence study comparing the in-house formulation of the reference compound and the commercial formulation."

And the panel arrived at this conclusion: "In aggregate, these deficiencies, and the fact that all relevant secondary endpoints were statistically significant in favor of AZD3582, do not allow a conclusion of negative trial outcome."

AstraZeneca is sticking to its guns, however. A press spokesman told BioWorld International that the Phase II trial was "well powered and well conducted and fulfilled the standards of good clinical practice." He reiterated the initial conclusion that it had "failed to meet the primary endpoint with respect to gastrointestinal ulcers." Nevertheless, AstraZeneca remains committed to completing the Phase II development program for the drug.

Goyon said, however, the "trial results do not disqualify either the product or the technology" of NicOx, since other factors could explain its inefficacy with regard to ulcers. He suggested the trial ought to be carried out again.

The NicOx panel consisted of eight experts from five countries (the U.S., Italy, France, Spain and Austria). It included Edmund Gehan, director of biostatistics at Georgetown University, and Gabriele Bianchi Porro, director of the gastrointestinal unit at the University of Milan, both of whom criticized the trial for the low incidence of ulcers in the reference drug group. Bianchi Porro said, "In view of the questionable quality of the study, it is almost impossible to draw any conclusions."

Meanwhile, NicOx has started clinical development of another compound, NCX1020, a nitric oxide-donating derivative of budesonide, which it describes as the first in a new class of inhaled nitrosteroids designed to treat severe respiratory diseases such as asthma and chronic obstructive pulmonary disease. The Phase I trial is being conducted in the UK and will assess not only the drug's general safety and tolerability but also its effects on pulmonary function at three different dose levels. It is the ninth compound NicOx has taken into clinical development.