An analysis of immunologic biomarkers in a phase II trial of patients with advanced colorectal cancer treated with chemotherapy and PD-1 checkpoint inhibition identified several factors associated with a patient’s beneficial response to the chemo-immunotherapy regimen, according to findings presented during the 33rd Annual Meeting of the Society for Immunotherapy of Cancer.
Matthew R. Farren, PhD
An analysis of immunologic biomarkers in a phase II trial of patients with advanced colorectal cancer (CRC) treated with chemotherapy and PD-1 checkpoint inhibition identified several factors associated with a patient’s beneficial response to the chemo-immunotherapy regimen, according to findings presented during the 33rd Annual Meeting of the Society for Immunotherapy of Cancer (SITC 2018).1
In the ongoing, single-arm, phase II trial (NCT02375672), a modified FOLFOX regimen (oxaliplatin, folinic acid [leucovorin], and 5-fluorouracil; mFOLFOX6) was given in combination with pembrolizumab (Keytruda) to enhance response to antiPD-1 therapy for patients with advanced or recurrent CRC.
Thirty patients were treated with 200 mg of intravenous pembrolizumab every three weeks plus mFOLFOX6 on days one and 15 of every 28-day cycle.
Results from the trial demonstrated a response rate of 26.7 percent with one complete response and seven partial responses by RECIST criteria. By immune-related response criteria (irRC), the response rate was 53.3 percent (16/30 PR).
At a median follow-up of 17.7 months, the median progression-free survival (PFS) was 11.3 months and the median overall survival (OS) was not yet reached.
Matthew R. Farren, PhD, of the Winship Cancer Institute of Emory University, discussed the biomarker analysis portion of the trial during a rapid-fire oral abstract presentation.
To complete the analysis, blood samples were taken from the patients at baseline, cycle 1 day 15, and cycle 3 day 1. The investigators assessed levels of 40 cytokines and chemokines and approximately 200 peripheral blood mononuclear cell immunophenotypes for their relation to PFS, OS, and clinical response.
Circulating CD4+ and CD8+ T-cell checkpoint molecules was assessed at baseline and following treatment in responders and non-responders. Among CD4+ T cells, reduced LAG- 3 and VISTA were both negatively associated with response. “The patients who had lower levels of LAG-3 and VISTA molecules with CD4+ cells were less likely to have a partial or complete response,” Dr. Farren explained. After treatment, patients with an increased frequency of CTLA-4 CD4+ T cells were more likely to have responded to treatment.
Farren noted that an analysis of Ki-67 represented T cells that were more highly proliferative. “As you might expect...we saw that for both CD4 and CD8 cells, in patients that had an increased frequency of Ki-67 following treatment, they were likely to [have] a response.”
By both RECIST and irRC, patients who had several circulating cytokines at baseline, such as G-CSF, were independently associated with beneficial response to the combination. Changes in such levels were independently associated with clinical responses to pembrolizumab in combination with mFOLFOX6.
“We found that several cytokines interchange at even two cycles of therapy that was predictive of what the patient’s response would be down the road,” Dr. Farren said.
Investigators have uncovered potential methods to overcome genetically-based resistance mechanisms to PD-1 checkpoint blockade in melanoma, according to the results of an analysis presented in a rapid oral abstract session yesterday afternoon.
Tumors with JAK1/2 loss-of-function (LoF) resistance could achieve benefit from antiPD-1 therapy with the addition of a TLR9 agonist, and those with β2-microglobulin (B2M) LoF resistance could be overcome by a new-generation IL-2 drug, explained Davis Torrejon, MD.
The abstract, presented by Dr. Torrejon, of the University of California, Los Angeles, received an Abstract Travel Award from SITC’s Forward Fund to recognize young investigators’ excellent novel research.
In the study, the investigators used CRISPR/Cas9 genome editing to create JAK1, JAK2, and B2M knockout sublines of a murine MC38 carcinoma model. The model represented a cancer with a high mutational load that would respond well to antiPD-1 therapy.
In JAK1/2 knockout sublines, the cells showed a loss of sensitivity to interferon (IFN), but that did not impair T-cell recognition and cytotoxicity. In B2M knockout sublines, the cells showed a lack of antigen presentation to the T cells.
RNA sequencing analysis demonstrated that in the JAK1/2 knockout cell lines, IFN-γinduced increased expression of an- tigen-presenting machinery, IFN-γ signaling, and the CXCL9 and CXCL10 chemokines were lost.
“JAK1, JAK2, and B2M [knockout] results in resistance from the PD-1 [therapy] in this model,” Dr. Torrejon said, as both JAK1/2 and B2M LoF tumors induced in vivo resistance to PD-1 checkpoint inhibition. Additionally, antiPD-1 therapy was unable to increase CD8 T cells in the JAK1/2 and B2M LoF–mutant tumors and were predominantly CD8 exhausted, he explained.
Dr. Torrejon and his colleagues attempted to overcome the mechanisms of these resistances with combination therapy. With intratumoral administration of SD-101, a TLR-9 agonist, the JAK1/2 knockout cell lines overcame local resistance to checkpoint inhibition,even in an abscopal manner. And with NKTR-214, a kinetically engineered IL-2 receptor agonist, the B2M knockout cell lines also overcame resistance to PD-1 inhibition and it increased survival significantly in these cell lines.
Torrejon D, Abril-Rodriguez G, Tsoi J, et al. Overcoming genetically-based resistance mechanisms to PD-1 blockade. Presented at: Presented at: 33rd SITC Annual Meeting; November 7-11, 2018; Washington, DC. Abstract P557.