There’s a yin and yang to neoantigens, Alberto Bardelli told the audience at the 2019 annual conference of the European Society of Medical Oncology (ESMO) in Barcelona, Spain, last month.
They contribute to tumorigenesis, resistance and tumor heterogeneity.
But they are also often specific to tumor cells but not normal cells and “some,” he said, “are actionable targets.”
How to identify and capitalize on those actionable targets, and how to predict who will benefit from immunotherapy, are currently among the pressing questions in immuno-oncology.
On the one hand, tumors with a high tumor mutational burden (TMB) are more likely to respond to immunotherapy, because they are the most likely to have, among their many neoantigens, one that the immune system can really sink its teeth into.
But mutation load alone, Bardelli said, is “informative but certainly not predictive.”
In an ESMO special symposium on tumor neoantigens, Yardena Samuels, principal investigator of the melanoma research group at the Weizmann Institute of Science, gave an overview of the contributors to mutational load, and how to separate it from the related concept of tumor heterogeneity.
Tumors are made up of different subclones, and the composition of those subclones changes over the course of the illness, both as a natural evolutionary process, and as a response to pressures brought on by treatment.
Heterogeneity, Samuels explained, consists of both the number of overall mutations burden a tumor has, and how that burden is distributed within the different subclones that make up the tumor. Overall, those aspects can be jointly expressed by the “Shannon Diversity index.” A low diversity index means that most mutations are in one clone, and predicts a greater likelihood of successful immunotherapy.
Bardelli, who is director of the laboratory of molecular oncology at Candiolo Cancer Institute and associate professor at the University of Turin, argued that at least some DNA repair-deficient tumors may be a special case.
DNA repair defects lead to different types of cancer mutations
In the microsatellite instability- high (MSI-hi) space, he told the audience, “I’m not convinced we need clonal neoantigens.”
MSI-hi tumors have an overall high mutations burden, and Keytruda (pembrolizumab, Merck & Co. Inc.) is approved for the treatment of MSI-hi tumors regardless of their anatomical site of origin.
Bardelli pointed out that MSI-hi tumors make up around 17% of colorectal cancers, making them a relatively frequent subtype.
But MSI-hi tumors make up only 3% to 5% of stage IV cancers, suggesting that something about having mismatch repair deficiencies protects patients from progression.
Such protection might stem from the tumors being highly visible to the immune system, and Bardelli is co-founder of Neophore Ltd., a British startup that is developing mismatch repair inhibitors in the hopes of turning tumors MSI-hi-like.
In 2017, Bardelli and scientific co-founder Giovanni Germano published a paper in Nature reporting that inactivating mismatch repair generated neoantigens and impaired tumor growth in preclinical models. Bardelli was frank about the seeming oddity of inactivating tumor suppressors to treat cancers, telling the audience that “I’m completely aware of the risk of exposing a patient to a mutagen. But I’m also interested in the ability of these molecules to cause indels and frameshifts. . . . Perhaps there is space to develop a small molecule that can inhibit mismatch repair.”
Another company focused on neoantigens is Pact Pharma Inc. Scientific co-founder (and principal investigator of the Yervoy trials) Antoni Ribas, also a professor of medicine and of surgery at the University of California, Los Angeles, described the approach, which focuses on finding and empowering the T cells that will respond to neoantigens, in what he called “personalized adoptive cell transfer.”
Pact Pharma’s approach – currently in phase I trials – is to capture patient-specific T cells, and predict which ones will lead to a strong antitumor immune response, and engineer those cells to create a personalized cell therapy for each patient.
Endogenous T cells, Ribas said, are “by and large dysfunctional,” owing to the immunosuppressive wiles of the tumors they are fighting. But “patients who do not respond will still have T cells that could respond.” The company’s goal is to identify and engineer those cells.