Columbia Laboratories Inc. fought a losing battle Friday in a statistical throwdown with the FDA at an advisory committee meeting convened to consider whether data from the PREGNANT trial are enough to support a preterm birth indication for Columbia's Prochieve (progesterone gel 8 percent).

At the end of the day, the Advisory Committee for Reproductive Health Drugs voted 13-4 not to recommend approval, agreeing with the FDA that the single pivotal trial failed to prove efficacy in women with a singleton gestation and a short cervix, especially in the U.S. (See BioWorld Today, Jan. 19, 2012.)

The vote sent shares of Columbia (NASDAQ:CBRX) plummeting to 55 cents, a 65 percent drop from $1.58, in afterhours trading Friday.

Even some of those voting for the new indication weren't overwhelmed by the data. Michael Greene, professor of obstetrics, gynecology and reproductive biology at Harvard Medical School, said that while the trial demonstrated a modest treatment effect with progesterone gel that was consistent with other studies on the use of progesterone, it was not a home run.

The sponsor took a gamble with doing one pivotal trial and lost, he added.

But Sciences International Inc.'s Anthony Scialli, who also voted for the drug, said in reality, there were other studies that evaluated progesterone in preterm birth, referring to several that were discussed in the meeting. However, the FDA made it clear that those studies couldn't be considered in evaluating Prochieve because they weren't part of the new drug application.

Scialli called it a "peculiarity" of the U.S. system that the agency won't consider the other trials.

"We're very disappointed," Frank Condella, CEO of the Livingston, N.J.-based Columbia, told BioWorld Today following the meeting. "We believe this drug works," and there's nothing out there to address the unmet need.

He said he expects to talk with the FDA before the drug's Feb. 26 PDUFA date to discuss a path forward. If another trial is required, Columbia's partner Watson Pharmaceuticals Inc. will pick up the costs, Condella said, so decisions about Prochieve's future are more in Watson's hands. (See BioWorld Today, March 5, 2010.)

In presenting its case, Columbia's claims of statistically significant results from the PREGNANT study were supported by several physicians. "This study has and will continue to change the practice of obstetrics," Sonia Hassan, a professor at Wayne University and the lead investigator for the trial, told the panel.

As a result of the trial, all pregnant women treated at Wayne are screened for short cervix, and vaginal progesterone gel is recommended as a treatment. Professional societies also are in the process of recommending universal cervical screening and the use of progesterone, even though no products have been approved for the indication.

FDA reviewers countered Columbia's analysis, saying the bioadhesive gel didn't significantly reduce preterm birth at any gestational age or improve the neonatal morbidity/mortality index in U.S. subjects, although statistical significance was noted at sites outside the U.S.

Barbara Wesley, an FDA reviewer, said the data suggested that the at-risk population outside the U.S. is not the same as those in the U.S.

"Do you really believe that progesterone reduces preterm birth outside the U.S. but not in the U.S.?" Stuart Campbell, a Columbia consultant who didn't participate in the trial, said, adding that such a premise is illogical.

Noting the results of several progesterone studies in preterm birth and a meta-analysis conducted by Robert Romero, Campbell said, "I believe doing nothing is no longer an option."

The FDA questioned the validity of Romero's meta-analysis of five progesterone studies, suggesting that he selectively chose the included studies and criticizing some of his conclusions.

Although he wasn't on the agenda, Romero put in a surprise appearance to respond to those criticisms, saying the meta-analysis included all studies in which cervical length was measured, since the point of the study was the efficacy of using progesterone in pregnant women with short cervixes. He also corrected the FDA on its other criticisms.

Suggestions for Another Trial

The advisory committee urged Columbia to conduct another trial, but one with a different design. If the sponsor repeats what it's already done, it won't do any good, said Louis Weinstein, past professor and chairman of the department of obstetrics & gynecology at Thomas Jefferson University. Instead, he recommended a study done at private practices in the U.S. with middle-class women.

Other panel members pushed for a head-to-head comparative-effectiveness trial with other off-label treatments. Although no other studies have been conducted in the U.S. on progesterone for this indication, oral progesterone capsules used as suppositories and compounded progesterone suppositories have been used off-label, Condella said.

Although most of the committee saw no safety issues with Prochieve, which has been on the market as Crinone for 14 years to treat infertility and secondary amenorrhea, a few members wanted a patient registry or follow-up data on children born to women who used the drug years ago.

But Greene reminded them of the reality of drug trials and approvals. "We'd all love to have 14 years or 15 or 20 years' worth of data on 100,000 treated patients to know for sure" that there are no safety or efficacy concerns, he said, noting that such information comes after a drug is approved.

In Makena's Shadow

Even though the FDA isn't supposed to consider price when approving a drug, the price of Prochieve entered the conversation. "We were clearly in the shadow of Makena," Condella said, referring to KV Pharmaceutical Co.'s brand version of hydroxyprogesterone caproate, which had for decades been formulated by pharmacists, at about $10 to $20 a shot to prevent preterm birth for a different patient population.

But when KV got Makena approved with a single trial that raised both safety and efficacy issues, the drugmaker priced the product at $1,500 a shot. (See BioWorld Today, May 3, 2011.)

Panel members made it clear they didn't want another Makena "fiasco." A representative from Watson assured them Prochieve would be priced in keeping with Crinone.

In the run-up to Friday's meeting, Columbia's stock took quite a ride. After the FDA briefly posted the documents for the meeting a day early, the stock plunged 54 percent Tuesday in the heaviest trading the company has seen in at least five years. With a volume of more than 18 million, shares of Columbia closed at $1.10 Tuesday, down from an opening price of $2.42.

They made a partial comeback Wednesday, closing at $1.47, when the FDA reposted the materials, giving investors an opportunity to do their own analysis, Condella said.