NeoIRX trial: Immunologic induction with peri-lymphatic cytokines to enhance pembrolizumab (pembro) response in stage II/III triple-negative breast cancer (TNBC).

604 Background: The addition of pembro to neoadjuvant chemotherapy (NACT) improves pathological complete response (pCR) rate and recurrence-free survival in stage II/III TNBC, albeit with substantial chemotherapy-attributed toxicity. We hypothesize that locoregional cytokine therapy could be combined with pembro to optimize immune response and clinical outcome. In a phase Ib early-stage breast cancer trial, we previously demonstrated that peri-lymphatic cytokine injection (IRX-2 regimen, comprised of physiologic doses of IL-2, IFNg, and other cytokines derived from activated donor lymphocytes, given with low-dose cyclophosphamide) is well tolerated and associated with increased intratumoral lymphocytes, T-cell activation, and PD-L1 expression. Here, we report preliminary outcomes of neoIRX, a phase II trial evaluating induction IRX-2 + pembro preceding NACT + pembro. Methods: Subjects with stage II/III TNBC were randomized to receive induction pembrolizumab (all subjects: 200mg IV) +/- peri-areolar IRX-2 (IRX-2 arm: 1 ml SQ x2 daily for 10 days + cyclophosphamide 300 mg/m2 IV x 1) preceding initiation of NACT + pembro. The primary endpoint was pCR rate following NACT + pembro (n = 15 subjects/arm planned); secondary endpoints were safety and tolerability. We explored post-induction/pre-NACT radiographic (ultrasound) and histologic outcomes (TILs, tumor regression) as a biomarker strategy to predict pCR. Results: The trial terminated after n = 12 subjects due to withdrawal of drug support for IRX-2. The IRX-2 arm achieved 83% pCR (n = 5/6, CI 36-100%) compared to 33% pCR with pembro alone (n = 2/6, CI 4-78%). The regimen was well-tolerated with minimal IRX-2-attributed toxicities (67% grade I skin bruise). Toxicities during NACT + pembro were similar to Keynote-522. 67% (n = 4/6) of IRX-2 subjects experienced week 3 radiographic regression, with evidence of brisk lymphocyte infiltration on week 3 biopsy, and with 100% pCR rate (n = 4/4) following NACT + pembro. n = 2/6 subjects receiving pembro+IRX-2 experienced pCR on week 3 biopsy, versus n = 0/3 evaluable in control arm. Conclusions: Induction IRX-2 + pembro is well tolerated and is associated with encouraging outcomes, supporting further study of peri-lymphatic induction cytokine therapy in stage II/III TBNC. Post-induction radiographic and histologic outcomes may identify patients with immune-responsive tumors, for whom NACT de-escalation may be a promising therapeutic approach to mitigate toxicity. Clinical trial information: NCT04373031 .