Abstract
Background: NIVO inhibits the interaction of the PD-1 receptor with its ligands, PD-L1 and PD-L2. While high tumor PD-L1 expression results in greater clinical benefit with NIVO, MEL patients (pts) with low-to-no PD-L1 expression also benefit from NIVO. This may be partially attributed to PD-L2. Here, we analyzed the association between PD-L1/PD-L2 expression and efficacy from a phase II study (CheckMate 064) evaluating NIVO followed by ipilimumab (IPI) vs IPI followed by NIVO in MEL.
Methods: Pts were randomized 1:1 to receive sequential treatment with NIVO (3 mg/kg IV Q2W x 6) then IPI (3 mg/kg IV Q3W x 4) (cohort A) or IPI then NIVO (cohort B). Pts were biopsied prior to treatment and at Wk 13; pts had only received one agent at Wk 13. PD-L1 and PD-L2 expression were assessed by immunohistochemistry on formalin-fixed, paraffin-embedded tumor samples with the Dako PD-L1 28-8 pharmDx assay and a novel PD-L2 assay (antibody clone 9E5). Dichotomization of the percent expression of PD-L1 and PD-L2 was optimized by ROC analyses.
Results: In pre-treatment samples, 32/97 (33%) pts had PD-L1 expression ?5% and 38/83 (46%) pts had PD-L2 expression ?70%. No consistent change from baseline in PD-L1 or PD-L2 was observed in either cohort with treatment. The frequency of pts with PD-L2 expression ?70% was higher among pts with PD-L1 ?5% vs PD-L1 <5%: 63% (19/30) vs 36% (17/47). In cohort A, there was a greater mean tumor burden reduction and higher response rates at Wk 13 among pts with high vs low PD-L2 expression, even in pts with low-to-no PD-L1 (Table). No association between PD-L2 and efficacy was observed in cohort B.
Conclusions: Similar to PD-L1, PD-L2 expression may enrich response to NIVO in MEL. PD-L2 expression may partially explain the efficacy of NIVO in pts with low-to-no PD-L1. However, NIVO is still active in pts with low-to-no expression of both ligands, suggesting other factors may explain the activity of NIVO. Further analysis is required to delineate these factors.
ExpressionStatus, % . | N . | CheckMate . | CheckMate . | |||
---|---|---|---|---|---|---|
. | . | . | 064 . | 064 . | ||
. | . | . | Cohort A . | Cohort B . | ||
PD-L1 . | PD-L2 . | . | Tumor response, n/N (%) . | Reduction in tumor burden, % . | Tumor response, n/N (%) . | Reduction in tumor burden, % . |
≥5 | ≥70 | 19 | 7/13 (54) | -26.6 | 2/6 (33) | 4.9 |
≥5 | <70 | 11 | 2/7 (29) | -13.3 | 0/4 (0) | 49.8 |
<5 | ≥70 | 17 | 4/10 (40) | -28.4 | 0/7 (0) | 58.0 |
<5 | <70 | 30 | 4/18 (22) | -2.8 | 0/12 (0) | 31.3 |
ExpressionStatus, % . | N . | CheckMate . | CheckMate . | |||
---|---|---|---|---|---|---|
. | . | . | 064 . | 064 . | ||
. | . | . | Cohort A . | Cohort B . | ||
PD-L1 . | PD-L2 . | . | Tumor response, n/N (%) . | Reduction in tumor burden, % . | Tumor response, n/N (%) . | Reduction in tumor burden, % . |
≥5 | ≥70 | 19 | 7/13 (54) | -26.6 | 2/6 (33) | 4.9 |
≥5 | <70 | 11 | 2/7 (29) | -13.3 | 0/4 (0) | 49.8 |
<5 | ≥70 | 17 | 4/10 (40) | -28.4 | 0/7 (0) | 58.0 |
<5 | <70 | 30 | 4/18 (22) | -2.8 | 0/12 (0) | 31.3 |
Citation Format: Scott J. Rodig, Evisa Gjini, Kaushal Desai, Chelsea Jin, Christine Horak, Mary Ruisi, Jeffrey Weber, Gordon J. Freeman, F. Stephen Hodi. Association of programmed death-ligand 1 (PD-L1) and 2 (PD-L2) expression with nivolumab (NIVO) efficacy in advanced melanoma (MEL). [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr CT133.