Introduction: Nivolumab and ipilimumab (NIVO+IPI) in combination was more efficacious than nivolumab (NIVO) alone in advanced melanoma patients (pts) in the phase 3 CheckMate 067 study (Larkin et al, NEJM. 2015). To elucidate the differential mechanisms of action of NIVO+IPI vs NIVO, we assessed biomarkers in pts with advanced melanoma treated with these agents in the CheckMate 038 study.

Methods: IPI-naive pts without brain metastases from parts 1-3 of the phase 1b CA209-038 study (NCT01621490) who received NIVO+IPI Q3W (1 mg/kg+3 mg/kg, n=53) or NIVO alone Q2W (3 mg/kg, n=52) were included in the analysis. Tumor and peripheral immune markers were assessed (Table), and compared across treatment groups and by best overall response (BOR) per RECIST v1.1.

Results: Increases in tumor-infiltrating CD8 T cells were observed with both NIVO+IPI and NIVO treatment, with greatest increases in NIVO+IPI pts with CR/PR or SD vs other BOR groups (Table). Increases in tumor PD-L1 expression were observed with both treatments, with 10-fold greater increases in pts with CR/PR to NIVO+IPI vs pts with PD/NE to NIVO+IPI or CR/PR to NIVO alone. Reductions in circulating MDSCs were observed with NIVO+IPI treatment across BOR groups, but only in pts with CR/PR in the NIVO-treated group. Increases in CXCL9, CXCL10 and IFN-gamma were observed irrespective of BOR in pts treated with both NIVO+IPI and NIVO alone, with the greatest increases with NIVO+IPI treatment.

Conclusion: NIVO+IPI vs NIVO alone was associated with consistent reductions in MDSCs and greater increases in IFN-gamma signaling. Pts with CR/PR to NIVO+IPI had the greatest increases in CD8 T cells and PD-L1 expression, suggesting enhanced T-cell effector function. This may partly explain the higher ORR observed for NIVO+IPI vs NIVO alone in CheckMate 067. Further analyses will elucidate the differential immunomodulatory effects of NIVO+IPI in combination vs NIVO monotherapy in pts with advanced melanoma.

Table.

Summary of results data

Median change from baselinea(min, max)             
 NIVO+IPI, n=53 NIVO alone, n=52 
 CR/PR SD PD/NE CR/PR SD PD/NE 
Tumor immune markerb n % n % n % n % n % n % 
CD8 T cells 10 13.8 (1.3, 40.1) 12.6 (0.3, 39.6) 13 2.9 (-16.8, 33.0) 4.3 (-17.3, 26.2) -0.4 (-3.0, 7.8) 0.5 (-1.3, 24.6) 
PD-L1 10 41.0 (5.0, 93.0) 1.0 (0.0, 55.0) 4.0 (-1.0, 57.0) 4.0 (-17.0, 30.0) 4.0 (2.0, 70.0) 14 2.0 (-53.0, 70.0) 
Peripheral immune markerc n % n % n % n % n % n % 
MDSCs 13 -1.4 (-12.6, 3.7) -5.8 (-9.9, 3.3) 12 -1.5 (-7.7, 4.5) -4.0 (-5.2, 2.8) 2.8 (-0.4, 5.9) 5.6 (-7.8, 14.2) 
Immune serum cytokined n pg/mL n pg/mL n pg/mL n pg/mL n pg/mL n pg/mL 
CXCL9 16 4294.5 (840, 53653) 3670 (2153, 26270) 15 2354 (402, 23188) 14 934 (85, 8970) 748 (190, 1765) 17 762 (-1250, 4490) 
CXCL10 16 410 (2, 4775) 345 (131, 2403) 15 369 (-20, 1706) 14 187 (-27, 768) 95 (50, 932) 17 155 (-172, 1099) 
IFN-gamma 16 0.40 (-0.02, 38.93) 0.09 (-0.14, 0.48) 15 0.11 (-0.41, 42.97) 14 0.02 (-0.02, 0.64) 0.02 (-0.16, 4.33) 17 0.04 (-0.08, 0.19) 
Median change from baselinea(min, max)             
 NIVO+IPI, n=53 NIVO alone, n=52 
 CR/PR SD PD/NE CR/PR SD PD/NE 
Tumor immune markerb n % n % n % n % n % n % 
CD8 T cells 10 13.8 (1.3, 40.1) 12.6 (0.3, 39.6) 13 2.9 (-16.8, 33.0) 4.3 (-17.3, 26.2) -0.4 (-3.0, 7.8) 0.5 (-1.3, 24.6) 
PD-L1 10 41.0 (5.0, 93.0) 1.0 (0.0, 55.0) 4.0 (-1.0, 57.0) 4.0 (-17.0, 30.0) 4.0 (2.0, 70.0) 14 2.0 (-53.0, 70.0) 
Peripheral immune markerc n % n % n % n % n % n % 
MDSCs 13 -1.4 (-12.6, 3.7) -5.8 (-9.9, 3.3) 12 -1.5 (-7.7, 4.5) -4.0 (-5.2, 2.8) 2.8 (-0.4, 5.9) 5.6 (-7.8, 14.2) 
Immune serum cytokined n pg/mL n pg/mL n pg/mL n pg/mL n pg/mL n pg/mL 
CXCL9 16 4294.5 (840, 53653) 3670 (2153, 26270) 15 2354 (402, 23188) 14 934 (85, 8970) 748 (190, 1765) 17 762 (-1250, 4490) 
CXCL10 16 410 (2, 4775) 345 (131, 2403) 15 369 (-20, 1706) 14 187 (-27, 768) 95 (50, 932) 17 155 (-172, 1099) 
IFN-gamma 16 0.40 (-0.02, 38.93) 0.09 (-0.14, 0.48) 15 0.11 (-0.41, 42.97) 14 0.02 (-0.02, 0.64) 0.02 (-0.16, 4.33) 17 0.04 (-0.08, 0.19) 

CR/PR=complete response/partial response; SD=stable disease; PD/NE=progressive disease/not evaluable for BOR, but otherwise response evaluable

a Baseline is defined as the last evaluable data prior to first dose treatment within 29 days

b Tumor biopsies collected day 8-36 on-treatment, assessed by immunohistochemistry

c Whole-blood samples collected prior to third-dose NIVO+IPI or prior to fourth-dose NIVO alone, assessed by flow cytometry

d Serum samples collected day 15+/-3 on-treatment, assessed by immunoassays

Citation Format: Antoni Ribas, Salvador Martín-Algarra, Shailender Bhatia, Wen-Jen Hwu, Craig L. Slingluff, William H. Sharfman, F. Stephen Hodi, Walter J. Urba, Jason J. Luke, John B. Haanen, Margaret K. Callahan, Jedd D. Wolchok, Scott D. Chasalow, Petra Ross-Macdonald, Tina C. Young, Anila Qureshi, Christine E. Horak. Immunomodulatory effects of nivolumab and ipilimumab in combination or nivolumab monotherapy in advanced melanoma patients: CheckMate 038 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT073. doi:10.1158/1538-7445.AM2017-CT073