UPDATE 3-U.S. panel votes to reject J&J, Zeltia cancer drug

* U.S. panel says toxicity risks likely outweigh benefit

* Votes 14-1 against FDA approval of Yondelis

* Zeltia shares down 6.5 pct, J&J up 1 pct (Adds analyst comment, share prices)

By Bill Berkrot

NEW YORK, July 15 (Reuters) - A Johnson & Johnson (JNJ.N) and Zeltia SA (ZEL.MC) drug for ovarian cancer should not be approved, as risks of heart and liver toxicity outweigh its limited ability to keep the disease in check, an advisory panel to the U.S. Food and Drug Administration said on Wednesday.

By a vote of 14-1, the panel of cancer experts recommended the agency reject the drug, Yondelis. The FDA typically, but not always, abides by the recommendations of its advisory panels.

While the rejection is relatively inconsequential for giant healthcare conglomerate J&J, it represents a major setback for Spanish biotechnology company Zeltia.

Zeltia shares closed down 6.5 percent in Spain prior to the final panel vote, while J&J shares were up 60 cents, or 1 percent at $58.83 in afternoon trading on the New York Stock Exchange.

JP Morgan analyst Michael Weinstein said in a research note he has removed Yondelis revenue projections from his J&J forecasts, but said the earnings impact will be minimal.

"Expectations for Yondelis were low, and this was by no means one of the higher profile programs in J&J's pipeline, but it is a negative nonetheless," said Weinstein, who had previously expected Yondelis sales of $275 million in 2012.

Yondelis, known chemically as trabectedin, was approved for the treatment of recurrent ovarian cancer in the Philippines last week, raising hopes it would get a favorable reception from the U.S. advisory committee.

But panelists overwhelmingly felt that the modest 6-week benefit in progression-free survival (PFS) shown in a pivotal late-stage clinical trial did not justify approval. The trial tested Yondelis plus Doxil chemotherapy versus Doxil alone.

J&J's Centocor Ortho Biotech unit said it is committed to working with the FDA to address the committee's concerns.

It said in a statement that the company "continues to believe trabectedin has an important role in the treatment of recurrent ovarian cancer."

The panel was asked to vote on whether adding trabectedin to Doxil represented a favorable benefit/risk ratio in this patient population.

While the risk bar is considerably higher for drugs for extremely sick advanced cancer patients than those for chronic diseases, a notable increase in severe and life-threatening adverse events in the Yondelis treatment group swung the vote against the drug.

Those included more than twice as many incidents of pulmonary embolism -- a potentially lethal blockage of the arteries to the lungs. Cardiac adverse events were three times higher in the Yondelis treatment group, as well as many more cases of liver toxicity.

Panelists also expressed skepticism about the value of progression-free survival -- the time it takes for the cancer to worsen -- as a primary goal for a type of cancer for which other approved therapies exist.

Many seemed willing to revisit the viability of the drug once overall survival data is known. Final overall survival data will not be available for about 18 to 24 months, the companies said. Interim analysis, which may not be indicative of final data, has so far found no significant difference in overall survival.

Company representatives argued that PFS was a clinically meaningful goal and said the drug could be an important treatment option for relapsed ovarian cancer patients who cannot tolerate platinum-based chemotherapies.

The hearing included emotional testimony from a cancer survivor, who credited Yondelis for her recovery. She had suffered from a different type of advanced cancer that failed to respond to treatment with Eli Lilly and Co's (LLY.N) Gemzar or Pfizer Inc's (PFE.N) Sutent, but said her tumors disappeared after being treated with Yondelis in a clinical trial.

"Trabectedin has given me back my life," she said in urging the panel to vote in favor of the drug.

The one panelist who heeded that call and who had had ovarian cancer, said: "It does look like there is a relatively small but significant group who could benefit from this." (Reporting by Bill Berkrot; additional reporting by Lisa Richwine in Washington; editing by Andre Grenon)