Comparison of T-cell activation status in patients treated with retifanlimab in combination with anti-LAG3/Anti-TIM3 vs retifanlimab alone

PD-1/PD-L1-targeting antibodies have revolutionized cancer treatment but most pts fail to respond (primary resistance) or lose response (secondary resistance). Anti-PD-1 therapy resistance mechanisms are poorly understood; other checkpoint inhibitor pathways are important investigational targets based on animal model studies. Co-expression of PD-1, LAG3, and TIM3 is associated with functional T-cell exhaustion and may contribute to resistance. Antibodies targeting LAG3 and TIM3 are potential candidates for overcoming anti-PD-1 resistance; LAG3 is also clinically supported as a target for resistance in pts with melanoma. INCAGN2385-203 is a randomized, phase 2 study to evaluate the efficacy and safety of retifanlimab (anti-PD-1) + INCAGN02385 (anti-LAG3) and retifanlimab + INCAGN02385 + INCAGN02390 (anti-TIM3) combinations vs retifanlimab alone in immunotherapy naïve PD-L1-positive (CPS ≥1) recurrent/metastatic SCCHN. Objective response rate was numerically higher in anti-LAG3-containing arms (~30%) vs the retifanlimab monotherapy arm (20%). Median progression-free survival was similar in all 3 arms. To evaluate pharmacodynamic biomarkers associated with LAG3 and/or TIM3 blockade in the context of PD-1 inhibition, flow cytometry was used to evaluate frequency changes of CD4 and CD8 T cells, regulatory T cells, and memory T cells, as well as activation and proliferation markers in whole blood samples. T-cell proliferation and activation were observed in all 3 arms. T-cell proliferation (CD4 and CD8) measured by frequency change of Ki67-positive T cells and HLA-DR expression was more pronounced in both anti-LAG3-containing arms vs the retifanlimab alone arm. No significant difference in regulatory T cells was observed between the 3 arms. Naïve CD4 and CD8 T cells decreased at similar levels at cycle 1, day 8 across all arms. A steady decrease of central memory CD4 and CD8 T cells up to cycle 4, day 1 was observed with retifanlimab alone, whereas retifanlimab + anti-LAG3 showed elevation of central memory T-cell frequencies in both CD4 and CD8 T cells up to cycle 1, day 15, before decreasing steadily up to cycle 4, day 1; there was no obvious change from baseline to cycle 4, day 1 in the triplet arm. Opposite trends were observed in effector memory CD4 and CD8 T cells. Retifanlimab alone produced a steady increase, whereas retifanlimab + anti-LAG3 showed decreases in CD4 (up to cycle 1, day 15) and CD8 (up to cycle 2, day 1) effector memory T cells before increasing steadily. These preliminary results suggest that LAG3 and/or TIM3 blockade may have unique effects on T-cell function. LAG3 with PD-1 blockade may overstimulate T cells and lead to T-cell elimination, highlighting the need for dose optimization. Further evaluation of T-cell responses of PD-1/PD-L1 blockade with other checkpoint inhibitors, including those targeting LAG3 and TIM3, is warranted.