Customizing Vaccines for Melanoma

Even as checkpoint inhibitors gain ground in cancer immunotherapy, scientists are investigating additional ways to alert the immune system to the presence of tumors. Preliminary results from a small phase I study at Washington University in St. Louis, MO, suggest that personalized vaccines directed at specific tumor mutations may have potential, at least in metastatic melanoma.

While cancer vaccines are not a new concept, their success to date has been limited because “they target shared antigens expressed in both normal and cancer cells, so the immune system doesn't see these antigens as foreign and is difficult to stimulate,” says trial leader and medical oncologist Gerald Linette, MD, PhD. The key difference with this study, he adds, is that its three participants, all of whom have advanced melanoma, received vaccines designed to harness immune activity against distinct neoantigens, peptides encoded by mutant genes in their respective tumors.

“Neoantigens are like flags on the tumor surface that can attract the immune system's attention,” explains immunologist Beatriz Carreno, PhD, the study's first author. Exome sequencing revealed that each patient had “hundreds of these flags,” she says, and “the challenge was identifying the ones that would elicit the strongest immune response.” Aided by a computer algorithm and validation assays, the researchers selected seven unique neoantigens for each patient, which were then incubated with the patient's own dendritic cells—responsible for presenting foreign material to cytotoxic T cells—and incorporated into a personalized vaccine.

Prior to this trial, the three patients had undergone surgery and received the checkpoint inhibitor ipilimumab (Yervoy; Bristol-Myers Squibb) upon relapse. Two had disease progression despite the drug, whereas one had a complete remission. Over approximately 18 weeks, each patient received three doses of their customized vaccine, which was well tolerated and did not cause any immune-related adverse events.

Analyzing blood samples, the researchers detected robust immune activity as early as 14 days post-vaccination. The patient who began in complete remission remains so, while the other two whose disease had been progressing have stable disease. Overall, each patient mounted responses to three of their seven neoantigens. The researchers also observed that vaccination expanded both the number and diversity of cytotoxic T cells recruited by the immune system, and confirmed that these T cells were capable of killing melanoma cells bearing the same neoantigens in vitro.

Given that the study was meant to assess safety and proof-of-principle, “it would be premature to conclude that this treatment had any therapeutic benefit,” Linette says. However, because checkpoint inhibitors have proven effective in only a fraction of melanoma patients, he hopes personalized vaccines, if determined to be therapeutically feasible, “will convert non-responders to responders” and broaden the population of patients who benefit from cancer immunotherapy.

“Right now there's no good way to identify these patients,” he adds, “but we have some clues to build on, and that's where this research is headed.”