Abstract LBO1-03: Randomized Phase II Study of Neoadjuvant Nivolumab (N) 2 week lead-in followed by 12 weeks of concurrent N+carboplatin plus paclitaxel (CbP) vs concurrent N+CbP in Triple Negative Breast Cancer (TNBC): (BCT1902/IBCSG 61-20 Neo-N)

Loi S, Niman S, Zdenkowski N, Francis P, Hay SB, Fox W, Punie K, Menzies AM, Angus R, Mavin C, Rennie L, Stoodley I, Mann B, Kuper-Hommel MJ, Regan M. Cancer Research, 84, no. 9_Supplement, (May 02 2024): LBO1-03-LBO1-03, doi:10.1158/1538-7445.Sabcs23-lbo1-03.

Background: A previous clinical study suggested that a priming dose of programmed cell death ligand 1 (PD-L1) inhibitor monotherapy 2 weeks prior to addition of chemotherapy could be more efficacious than starting all agents concurrently. Furthermore, the addition of checkpoint therapy may allow shorter chemotherapy duration in selected patients. The aim of this study was to investigate the strategies of lead-in Nivolumab (N) monotherapy and starting concurrent N with 12 weeks of carboplatin and paclitaxel in TNBC. Methods: In this multicenter Phase II study, eligible patients with stage I (cT1c)-II TNBC were randomized to receive either Arm A (lead-in): N 240 mg monotherapy, followed 2 weeks later by N 360mg + carboplatin (AUC 5) every 21 days x 4 cycles + weekly paclitaxel (80 mg/m2) for 12 weeks; or Arm B (concurrent): same agents initiated concurrently given over 12 weeks, followed 2 weeks later by N 240mg monotherapy, prior to surgery. The primary endpoint was pathological complete response (pCR ypT0/Tis ypN0). Secondary endpoints are residual cancer burden (RCB), safety, response according to baseline stromal tumor infiltrating lymphocytes (sTILs) quantity and PD-L1 expression and event-free survival (EFS). sTIL high was defined as ≥30% and PD-L1 positive as ≥1% (SP142 assay). A non-comparative Simon two-stage design was used to test the null hypothesis, in each Arm separately, of a pCR rate ≤40% vs an alternative pCR rate of ≥60% with 90% Confidence Intervals (CI). No adjuvant N was given, and adjuvant chemotherapy was at the discretion of the investigator in case of non-pCR. Results: 110 patients were randomized from July 2020 to April 2022; 108 patients who initiated treatment were analyzed. Baseline characteristics: median age was 49yrs (IQR 43-60yrs), 16.7% had node-positive disease, 34.3/64.8% had stage I/II disease, 33.3% were sTIL high and 47.2% were PD-L1 positive. For Arms A and B respectively, pCR rates were 50.9% (90% CI:39.0-63.2) and 54.5% (90% CI:42.7-66.2) with an overall pCR rate of 52.8% (90% CI:44.4-61.0). The RCB 0+1 rates were overall 68.5% (90% CI: 60.4-75.9); with 64.2% (90% CI: 52.0-75.1) and 72.7% (90% CI: 61.2-82.4) for Arms A and B respectively. Overall, in the sTIL high vs low subgroups, pCR was 66.7% (CI: 51.7-79.5) vs 45.7% (CI: 35.5-56.2%), and in the PD-L1 positive vs negative subgroups, pCR was 70.6% (CI: 58.4-80.9) vs 33.3% (CI: 21.8-46.6) respectively. pCR was 48.6% in stage I and 54.9% in stage II. Treatment-related adverse events were similar in both arms, with grade 3-4 adverse events in 64.8% patients. Early discontinuation of N occurred in 17 patients (15.7%). Immune related endocrine dysfunction (any grade) was observed in 22 patients (20.4%). Conclusion: The pCR rates exceeding 50% support that 12 weeks of a neoadjuvant non-anthracycline chemotherapy regimen with nivolumab is efficacious for Stage I/II TNBC with either concurrent or lead-in N. This study did not support the hypothesis that lead-in N was associated with a pCR advantage. The regimen was well tolerated, with no new safety signals. Patients with immune enriched tumors, identified by high sTILs or PD-L1 positivity, had very high pCR rates, identifying a subpopulation for whom a 12 week anthracycline-free chemotherapy regimen with N may be appropriate. EFS results are still maturing.