Universal Cancer Vaccine -- Still a Pipe Dream?

Kristin Jenkins

March 01, 2017

Advances in targeted immunotherapy strategies, including clinical success with immune checkpoint modulators, have revived interest in the development of a cancer vaccine, researchers say.

However, results from a "best-case scenario" analysis of genetic alterations in more than 60,000 unique tumors from many different types of cancer indicate that "the holy grail of cancer therapy" — a universal cancer vaccine — may still be a bit of a pipe dream, thanks to the "seemingly limitless genetic diversity in cancer."

Even if a semiuniversal HLA-specific cancer vaccine strategy could be developed, it would only be relevant to a subset of people representing about 0.3% of the general population, say Ryan J. Hartmaier, PhD, a senior scientist at Foundation Medicine, Inc, in Cambridge, Massachusetts, and colleagues.

"It is possible to identify a set of alterations shared across patient tumors for the production of a non-individualized, poly-neoantigen cancer vaccine in an HLA subtype-specific manner," the researchers explain in an report published online February 24 in Genome Medicine.

"However, with current neoantigen prediction methodologies, this approach will be applicable to only a small proportion of the population. Similar analysis of whole exome/genome sequencing, although not currently feasible at scale in a clinical setting, will likely uncover further diversity," they write.

"Early on, the research team nicknamed the analysis Project Snowflake when it revealed the vast genetic diversity of cancer tumors," Dr Hartmaier said in an interview. "Each tumor is unique. Even tumors that rely on the oncogenic effects from the same specific driver mutation share few other alterations."

It would be good to keep this in mind, he added, "when we are perplexed by the heterogeneous response in clinical trials."

Targeted cancer vaccines created for individual patients not only are impractical from both a time and cost perspective, but the technologies needed to develop them cannot be scaled to large populations. In light of this, the researchers decided to investigate whether "nonindividualized" immunotherapies could be developed to target many different neoantigens.

A broad vaccine targeting many different tumors "could be seen by some as the 'holy grail' of cancer therapy," Dr Hartmaier said in a statement.

Using next-generation sequencing-based genomic profiling from 63,220 unique tumors, the researchers found that approximately 45% of mutanomes — the mutations contained in each tumor — had at least one mutation from the set of 10 mutations they had chosen to maximize the number of unique tumors.

Computer-aided analysis, used to predict which of the alterations would be targetable, showed that between 2% and 12% would produce a neoantigen.

"This held true for tumors driven by KRAS G12C, PIK3CA E545K, or EGFR L858R alterations, which define distinct sample subsets," Dr Hartmaier and colleagues report. "We therefore hypothesized that sets of carefully selected mutations/neoantigens may allow the development of broadly applicable semiuniversal cancer vaccines."

To test this approach, they selected a panel of 10 neoantigens that could be applied to the maximum number of unique tumors in the data set. The researchers found that between 0.7% and 2.5% of the two most common HLA-A/B subtypes in North American/European populations contained at least one alteration that would produce one of the predicted neoantigens. They estimated that this would be relevant in fewer than 0.3% of the population.

In addition, KRAS G12C–driven tumors produced similar results, with 0.8% and 2.6% for each HLA subtype, respectively, indicating that "neoantigen targets still remain highly diverse even within the context of major driver mutations," the researchers say.

Importantly, the analysis led to a refined definition of "shared neoantigens" based on unique peptides rather than mutations. This may have important implications for the development of a cancer vaccine, the researchers say, "since a given mutation can produce many distinct peptides, each with their own MHC-I affinities." They point out that for this reason, the study's main conclusions "are unlikely to be significantly altered by the reliance on targeted sequencing data."

Although it is too early to predict how successful targeted immunotherapy strategies will be, early experimental and clinical data are "very encouraging," Dr Hartmaier told Medscape Medical News. "The successes of checkpoint inhibitor immunotherapies have taught us that leveraging the immune system as a therapeutic tool can be a game changer. Many believe that cancer immunotherapy and genomic profiling are scientific disciplines mature enough to make targeted cancer immunotherapy strategies feasible. As such, there is a lot of excitement about the potential they can offer."

However, there are many major technical and logistical barriers to making these strategies scaleable, widely available, and affordable, Dr Hartmaier said. The study provided "some data to help guide these efforts and [determine] whether some of these barriers, namely, the requirement for individualized therapies, could be bypassed," he noted, adding that individualized cancer vaccine approaches could be feasible with significant advances to scale the technology.

He also pointed to the study's limitations, "especially around MHC-II-presented neoantigens, which are notoriously hard to predict but offer hope for nonindividualized cancer vaccine strategies. More work needs to be done."

Targeted immunotherapies are unlikely to be "the whole story" and may be used with other immunotherapies, such as checkpoint inhibitors, he predicted.

"Should the clinical efficacy be strong enough, I have little doubt that the field will find ways to overcome the technical and logistical barriers in order to make it [targeted cancer immunotherapy strategies] widely available and more cost-effective," he said.

This study was funded through Foundation Medicine. Dr Hartmaier and coauthors David Fabrizio, Michael E. Goldberg, Lee A. Albacker, and Juliann Chmielecki have financial relationships with Foundation Medicine, including patent ownership. Coauthors Jehad Charo and William Pao are employees of Hoffman La Roche Ltd, which owns an equity stake in Foundation Medicine.

Genome Med. Published online February 24, 2017. Full text

Follow Medscape Oncology on Twitter for more cancer news: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.