Skip to main content
Premium Trial:

Request an Annual Quote

Early Results Show Personalized Vaccines for Melanoma Patients May Boost Immune Response


NEW YORK (GenomeWeb) – Vaccines that are tailor made to target melanoma patients' specific mutations in order to jumpstart an immune response may be a promising avenue of future therapy, according to new research published today.

In a proof-of-principal study published today in Science Express, researchers from Washington University School of Medicine showed that they could use genomics to design a personalized vaccine for three patients with advanced melanoma and that the vaccine boosted those patients' immune responses. One patient, who had achieved remission prior to receiving the vaccine, is still in remission, while the other two are currently in stable condition. In addition, the vaccine appeared to be well tolerated and the patients did not have any immune-related adverse events.

The group now plans to test the strategy in a phase I clinical trial in combination with checkpoint inhibition therapy that will likely begin in the next nine months to  one year.

"This proof-of-principal study shows that these custom-designed vaccines can elicit a very strong immune response," senior author Gerald Linette, a medical oncologist at Washington University School of Medicine and leader of the clinical trial at Siteman Cancer Center and Barnes-Jewish Hospital, said in a statement.

To design the personalized vaccines, the researchers first performed both exome sequencing and RNA sequencing on patients' tumor samples. They looked only for missense mutations causing an amino acid substitution in peptides that bind to immune cells, with the hypothesis that designing a vaccine based on those neoantigens would help elicit a tumor-specific immune response.

"It's a new approach that merges cancer genomics with immune cell therapy to develop a personalized dendritic cell vaccine that activates killer T cells against neoantigens," Beatriz Carreno, lead author of the study and an associate professor of medicine at Wash U, said during a conference call with reporters.

The three patients all had advanced melanoma that had been removed surgically. In addition, all three were at high risk for recurrence and did recur and were all treated with the checkpoint inhibitor ipilimumab prior to receiving the vaccine. One patient achieved complete remission in response to ipilimumab while the other two continued to progress.

Following exome and RNA sequencing by researchers at Wash U's Genome Institute, the team chose seven mutations from each patient, all of which were unique to the patient's specific tumor, on which to base the vaccine.

Elaine Mardis, co-director of the Genome Institute, said that the RNA sequencing component was critical for making sure that the mutations found from the exome sequencing component were actually being expressed in the tumor. Sequencing RNA helped to "eliminate 40 to 50 percent of predicted amino acid changes" because the gene or variant is not expressed, she said during the conference call. "That winnows the list down quickly," she added.

Next, a computer algorithm scoured the remaining candidate mutations to determine which were likely to bind tightly to immune molecules. The final step was to do a series of lab assays using the patients' tumor cells to determine that the predicted neoantigens were indeed being expressed on the surface of immune cells.

Carreno said that from the results of these three patients as well as an additional two patients that have since entered into the study, it seems unlikely that neoantigens, or so-called "flags," will overlap between patients.

The team administered three doses of the vaccine to the patients over 18 weeks and none had any immune-related adverse events. In addition, the vaccine appeared to boost patients' immune response as well as increase the diversity of neoantigen-specific T cells, a somewhat surprising finding that suggests that "cancer patients have a potentially rich pool of naïve tumor-specific T cells that remains dormant unless activated by vaccination," Carreno said.

The group evaluated vaccination response by analyzing blood before vaccination and at weekly timepoints after. All of the patients were found to have pre-existing immunity to at least one neoantigen — a finding that was not surprising, as naturally occurring immunity has been known to occur. Each patient displayed enhanced T cell immunity to three of the seven peptides in their respective vaccine. The three peptides with the strongest reaction had the highest binding affinity to immune cells. In addition, in all three patients the vaccine revealed new clonotypes that responded to the neoantigen.

"We're very encouraged by the quality of the T cell immune response directed against the melanoma neoantigens in all three patients," Linette said. Of the three patients, the one that previously went into remission is still in remission, while another patient has stable disease, and the third patient experienced a slight two-month regression of lung metastasis. That patient has not received additional therapy and the cancer has since grown back to where it was, but has not progressed further, Linette said.

Although the "natural history" of the patients' melanoma is that it would progress, Linette said, it is nonetheless too early to say whether the vaccine itself provided any therapeutic benefit to the patients. "The primary goal was to see if this was safe and if we could elicit an immune response and what that response would look like," he added.

The turnaround time from receipt of the patient's sample for genomic analysis to administration of the vaccine was approximately three to four months. The sequencing and analysis portion took approximately two weeks. Carreno said that she anticipates that the entire process could be shortened to around six weeks, eventually, which Linette said would be a clinically relevant time frame.

"The ideal patient would have disease that could be surgically resected," he said, enabling two to three months to analyze the tumor and develop a vaccine. However, even for patients with metastatic disease, the approach could still be feasible, he said. Often those patients still have a portion of their disease that is resectable, and while the vaccine is being developed they may be administered other targeted therapies like BRAF or MEK inhibitors, or even a checkpoint inhibitor prior to receiving the vaccine, Linette said.

The next step, said Linette, is to conduct a small phase I trial where patients would receive the vaccine in conjunction with a checkpoint inhibitor. He said that a number of questions regarding the clinical administration of the vaccine still need to be addressed, including whether it makes a difference whether the vaccine is given prior to the checkpoint inhibitor, concurrently with the drug, or after the therapy.

Mardis added that the approach could potentially be amenable to other types of cancer. Previously, she said that the Wash U team planned to apply the approach to triple-negative breast cancer and lung adenocarcinoma. Additionally, under a five-year grant from the Susan G. Komen for the Cure Foundation, Wash U researchers began pre-clinical work in animal models in 2011 to design a personalized vaccine for breast cancer patients.