Background: INDUCE-1 is a first-in-human trial investigating FE, an IgG4 ICOS agonist non-T-cell depleting mAb, in monotherapy (mono) and combo with PD-1 inhibitor PE. ICOS is highly expressed in melanoma, an immunotherapy responsive tumor, and is a biomarker of response to anti-CTLA-4 treatment; in nonclinical models, ICOS agonism has enhanced activity in combo with immune checkpoint blockade (ICB). Preliminary data from INDUCE-1 R/R melanoma ECs are presented to support FE expansion in ICB experienced (exp) R/R melanoma patients (pts).

Methods: Eligible pts had non-uveal R/R melanoma, ≤5 prior lines of systemic therapy, achieved response or stable disease (SD) on prior anti-PD-1/L1 treatment, and no prior immunotherapy-related Grade ≥3 toxicities leading to treatment discontinuation. In FE mono and FE + PE 200 mg combo ECs, pts were randomized to FE 0.3 or 1 mg/kg; treatment was given every 3 wks up to 35 cycles until disease progression or unacceptable toxicity. Safety and efficacy were assessed. Biomarkers in tumor biopsies were analyzed.

Results: By 30 Mar 2020, 39 pts in the mono EC were evaluable (21 pts at 0.3; 18 pts at 1 mg/kg FE); all pts were anti-PD-1/L1 exp; 21/39 (54%) were anti-CTLA-4 exp; 7/39 (18%) had ≥4 prior lines of therapy. Of the 39 pts, 29 (74%) discontinued treatment due to progression. Response rate was 10% (4/39; 1 complete response, 3 partial responses); 8/39 (21%) pts had SD and 2/8 (25%) pts with SD had tumor size reductions (RECIST v1.1) at ≥19%; 5 pts (13%) had SD at ≥18 wks. Duration of response (DoR) was 2.0+ to 6.3 mo. BRAF and NRAS status did not impact clinical activity. Anti-CTLA-4 exp pts had a numerically higher response rate than anti-CTLA-4 naïve pts.

In the combo EC, 17 anti-PD-1/L1 exp pts were evaluable (6 pts at 0.3; 11 pts at 1 mg/kg FE); 12/17 (71%) were anti-CTLA-4 exp; 15/17 (88%) discontinued treatment, all due to progression. Response rate was 18% (3/17), DoR was 7.9+ to 8.2 mo, and disease control rate (response or SD) was 53% (9/17); 1 pt with SD had tumor size reduction of 10%.

Ongoing analyses with mono EC showed a 28.6% (2/7) response rate, DoR of 2.0+ to 6.3 mo, and mOS of ~26 mo in pts with ICOS high tumor-infiltrating T cells.

Sample size limits comparison; however, clinical activity was similar between FE doses and between BRAF+ and BRAF− pts. Adverse events (AEs) were consistent with AEs reported for PE.

Conclusions: FE is the first ICOS agonist with reported single-agent activity in ICB exp R/R melanoma, supporting ICOS as a target. FE + PE in combo shows promising clinical activity and manageable safety in R/R melanoma. Continued survival follow-up of ECs is warranted. Updated efficacy and PK/PD data to be presented.

Funding: Study 204691 (NCT02723955) funded by GSK in collaboration with Merck Sharp & Dohme Corp, a subsidiary of Merck & Co, Inc, Kenilworth, NJ, USA.

Citation Format: Michele Maio, Jeffrey S. Weber, Maria Vieito Villar, Frans L. Opdam, Victor Moreno, Omid Hamid, José Trigo, Michael Chisamore, Marco Balas, Sapna Yadavilli, David C. Turner, Courtney Henry, Xiao Ji, Catherine Ellis, Marc Ballas, Axel Hoos, Antoine Italiano. Inducible T cell costimulatory (ICOS) receptor agonist, feladilimab (FE), alone and in combination (combo) with pembrolizumab (PE): Results from INDUCE-1 relapsed/refractory (R/R) melanoma expansion cohorts (EC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr CT033.