Updated with new survival data. CHICAGO ( TheStreet) -- A herpes simplex virus re-engineered to kill cancer cells shrank tumors and produced an early trend towards prolonging the lives of patients with skin cancer, according to updated results from a phase III study conducted by Amgen ( AMGN) that will be presented Saturday afternoon at the American Society of Clinical Oncology (ASCO) annual meeting. The treatment of melanoma, or skin cancer, is a hot topic at this year's ASCO annual meeting, with a lot of attention focused on experimental but very promising drugs from Bristol-Myers Squibb ( BMY) , Merck ( MRK) and Roche ( RHHBY) that prime a patient's immune system to identify and kill cancer cells. Amgen is the first company to demonstrate that a so-called "oncolytic virus" -- known as talimogene laherparepvec (T-Vec) -- can eliminate skin cancer in a late-stage clinical trial. T-Vec also works, in part, by harnessing the immune system of patients, which suggests to Amgen that the two different immune therapy approaches to killing cancer cells could be combined. "Scientifically, it's a compelling idea to combine T-Vec with a PD-1," said Amgen's David Chang, who oversees the T-Vec program. PD-1 refers to the drugs being developed by Bristol-Myers, Merck and Roche. "We're doing early safety studies, it's too early to comment but the interest in pursuing this approach is much higher today." In the Amgen phase III trial, more than 400 patients with advanced skin cancer were randomized to treatment with the T-Vec or the immune booster GM-CSF. Sixteen percent of the T-Vec patients reported meaningful tumor shrinkage for at least six months, compared to 2% for patients treated with GM-CSF. The difference in durable tumor response was statistically significant in favor T-Vec, achieving the primary endpoint of the study. The overall response rate in the study was 26 percent for T-Vec compared to 6 percent for GM-CSF. T-Vec was even more effective in patients with earlier-stage melanoma which had spread to the skin and lymph nodes but had not yet invaded organs like the liver or lungs. T-Vec induced durable tumor shrinkage in 33 percent of these earlier-stage patients. T-Vec's tumor shrinkage rates in the 20-30 percent range for melanoma are slightly lower than responses seen by the so-called PD-1 inhibitors being developed by Bristol-Myers and Merck. "We view the overall treatment effect to be similar to what the other agents are showing," said Amgen's Chang. "The only notable side effect with T-Vec is flu-like symptoms that is transient. Put everything together and melanoma patients with extensive skin and lymph node involvement but without visceral metastases may be well served with T-Vec." Unlike the Bristol-Myers and Merck trials to date, Amgen's is capable of measuring a survival benefit against a comparator treatment. In the phase III study, treatment with T-Vec has reduced the risk of death by 21 percent compared to GM-CSF alone. The survival data is not yet statistically significant although it came close with a p value of 0.07. (A p value of 0.05 would have been statistically significant.) At one year, the difference in the percentage of patients still alive in the study was 4.3 percent, but widened to 8.3 percent at two years and 12.8 percent at three years. This is consistent with the idea that cancer immunotherapy takes longer to be effective, Change says. Amgen expects to conduct another survival analysis of the T-Vec study later this year. Amgen's development of T-Vec began in January 2011 when it acquired the therapy's original developer, BioVex, for $425 million plus another $575 million in potential future payments. In an interview at the time of the deal, Roger Perlmutter, Amgen's chief scientist, said, "What impresses me about about BioVex is the firm scientific approach the company has taken in engineering the virus but also the responses seen to date in clinical trials ."