In a major advance against the deadliest type of skin cancer, a genetically targeted drug shrank tumors throughout the bodies of 80 percent of patients with terminal melanoma.

The study, led by the University of Pennsylvania and conducted at seven medical centers, is the first to successfully exploit a molecular abnormality found in about half of melanoma tumors, but not in healthy cells.

"This is the beginning of personalized medicine in melanoma," said senior author Paul Chapman, a melanoma researcher at Memorial Sloan-Kettering in New York City.

The drug - called PLX4032 until developers Plexxikon Inc. and Roche Pharmaceuticals give it a name - is the second recent breakthrough against a disease that had not had one in 20 years. In June, researchers reported that ipilimumab, an immune-stimulating drug being developed by Bristol-Myers Squibb and Medarex Inc., extended survival for patients whose melanoma had metastasized from the skin to internal organs.

The only two therapies currently approved by the Food and Drug Administration for metastatic melanoma do not work for most patients, and do not improve survival even in those who respond. Thus, a melanoma diagnosis is grim unless the malignancy is caught early enough to be cured by surgery.

And melanoma is becoming increasingly common. In the United States - where there were about 69,000 new diagnoses and 9,000 deaths last year - the percentage of people who develop melanoma has more than doubled in the last 30 years, according to the National Cancer Institute.

PLX4032 targets a mutation in a growth-promoting gene called BRAF. Research indicates that 40 percent to 60 percent of melanoma patients - and 8 percent of all cancers - have this BRAF gene mutation, according to the study, published Thursday in the New England Journal of Medicine.

Even so, focusing on BRAF was a gamble, as Keith Flaherty, the former University of Pennsylvania oncologist who led the study, was well aware. By the time the first PLX4032 patients were enrolled about two years ago, Flaherty had shown that Nexavar, an anti-BRAF drug approved to treat kidney cancer, did not work against melanoma, even when combined with standard chemotherapy.

Skeptics doubted that the BRAF gene was a good target for melanoma, a complex disease that can defy treatment by switching molecular signals and pathways.

Those doubts grew when the first group of patients did not respond to PLX4032, even at high doses.

"It was scary and disappointing," recalled Flaherty, who this year became director of developmental therapeutics at Massachusetts General Hospital in Boston.

In the hope that the dosage was still too low, the researchers interrupted the study and reformulated PLX4032 to make it easier for patients to absorb.

Suddenly, it worked.

"I have been treating patients with melanoma for 20 years, and I've never seen a drug response like this," said Lynn Schuchter, Penn's chief of hematology-oncology, who is enrolling patients in a larger trial of PLX4032. "Some patients responded in just 72 hours, going off pain meds."

To make sure the drug was hitting its target, the first group of 55 patients included 39 who did not have the BRAF gene mutation. In all 39, cancer continued to progress.

Among the 16 patients who did have the mutation, however, tumors got smaller in 10 patients, and disappeared in one, for a response rate of 69 percent. Tumor shrinkage was seen in all organs, including liver, bowel, and bone.

In a second group of 32 patients, all with the mutation, the response rate was even higher - 81 percent. Tumors disappeared in two subjects, and shrank in 24.

Side effects from the twice-daily capsule were moderate, mostly rash, nausea, and fatigue, although 18 patients developed squamous cell skin cancers, a nonaggressive type that is easily removed.

Exactly how long the drug can keep advanced cancer from progressing remains to be seen; 16 patients are continuing to benefit from ongoing therapy. So far, the duration of responses to PLX4032 has ranged from two to 19 months, with the median being more than 7 months. (Since safety and effectiveness studies are in the early stages, the drug cannot be provided outside the trials on a compassionate basis, a Plexxikon spokeswoman said.)

The ultimate question - whether the response translates to longer overall survival - will be answered by a bigger study now under way.

The hope, researchers agree, is that melanoma patients can be screened for key genetic mutations, then given a combination or succession of targeted therapies that cut off cancer's escape routes.

One possible candidate, Novartis' leukemia drug Gleevec (imatinib), is already on the market. A recent study showed Gleevec can shrink some melanoma tumors - although only a small proportion of melanomas are driven by mutations in the gene that Gleevec targets.

Of course, melanoma patients are not the only ones who stand to benefit from the BRAF breakthrough.

"The hope is that this is a bigger foot in the door for cancer therapy because of the prevalence of BRAF mutations," Flaherty said.

Contact staff writer Marie McCullough at 215-854-2720 or