Study explores biological implications of MHC-II expression in tumor cells

FCRL6

“[FCRL6] actually has an [immunoreceptor tyrosine-based inhibitory motif] domain in the intracellular portion of the human ortholog, which suggests that it could have some inhibitory function,” Dr. Balko said, adding that it has been shown to be expressed in a substantial proportion of natural killer cells and CD8 cells, and in a minor fraction of CD4+ T cells, which have been described as “cytotoxic CD4 cells.”

An immortalized FCRL6-negative natural killer cell line know as NK-92 was used to test for inhibitory function.

“We co-cultured it with K562 cells, which are a leukemia cell line that is both class I and class II negative; because they have a missing-self signal, the natural killer cells will naturally lyse the K562 cells, which can be measured by chromium release,” he explained.

When MHC-II was reconstituted on K562 cells, the natural killer cells still had effective lysis of the K562 cells, but when FCRL6 was also transduced on the natural killer cells, this interaction was stopped, and there was suppression of cytotoxic activity, or chromium release, in the co-cultures, suggesting that FCRL6 may have a checkpoint-like functionality, he said.

In the melanoma dataset, a look at FLCR6 mRNA in the tumor microenvironment showed that it was also much more highly expressed in HLA-DR–positive tumors and in the relapsed specimens.

In the tumors with paired specimens (three of which were MHC-II positive and three of which were MHC-II negative), IHC for FCRL6 identified greater enrichment of lymphocytes in the MHC-II–positive tumors, but the difference was not statistically significant.

In the breast cancer samples, where more LAG-3 and FCRL6 was seen in the triple-negative breast tumors, quantitative immunofluorescence showed that FCRL6-postive lymphocytes and LAG-3-positive lymphocytes had a substantial suppression of CD8-sel-positive granzyme B-positive cells within the microenvironment that was more substantial than that observed with PD-L1 expression, he noted.

“So our conclusions are that MHC-II tumors demonstrate enhanced T cell-mediated inflammation and immunity and anti-tumor immunity is circumvented through adaptive resistance by PD-1 and potentially LAG-3/MHC-II engagement in some tumors, and that ... FCRL6 may be a novel MHC-II receptor with inhibitory functionality, and could be a new immunotherapy target,” he said.

MHC-II expression could be useful for stratifying patients to combined anti-PD-1/anti-LAG-3 therapy, and eventually to combined anti-PD-1/anti-FCRL6 therapy, he added.
 

Combined anti-PD-1 and anti-LAG-3 therapy

The findings are of particular interest given recent findings regarding LAG-3 antibodies in development, said invited discussant Antoni Ribas, MD.

In a study reported by Ascierto et al. at ASCO 2017, for example, combined anti-PD-1 and anti-LAG-3 therapy had a 13% overall response rate in metastatic melanoma patients who progressed on anti-PD-1 therapy alone (20% and 7.1% in those with and without LAG-3 expression, respectively), said Dr. Ribas of the University of California, Los Angeles.

The 20% response rate seen in those with LAG-3 expression suggests “there could be a biomarker for this combined therapy,” he said, noting that while the overall response rate of 13% is low, “it is relevant because it is rescuing some patients who progressed on therapy, and it follows Dr. Balko’s science of why that would be the case.”

Dr. Balko has received research funding from Incyte, and holds a patent on use of HLA-DR/MHC expression to predict response to immunotherapies. Dr. Ribas owns stock in Advaxis, Arcus Ventures, Compugen, CytomX Therapeutics, Five Prime Therapeutics, FLX Bio, and Kite Pharma, and has served as a consultant or adviser for Amgen, Genentech/Roche, Merck, Novartis, and Pierre Fabre.

sworcester@frontlinemedcom.com

SOURCE: Balko J et al., ASCO-SITC abstract 180