Timing of nivolumab with neoadjuvant carboplatin and paclitaxel for early triple-negative breast cancer (BCT1902/IBCSG 61-20; Neo-N): a non-comparative, open-label, randomised, phase 2 trial.

Background: The optimal scheduling of PD-1 inhibitors with neoadjuvant chemotherapy in patients with early triple-negative breast cancer is unknown. We aimed to investigate the activity of two differing schedules of neoadjuvant nivolumab initiation with 12 weeks of carboplatin and paclitaxel for this patient population.

Methods: Neo-N is an investigator-initiated, non-comparative, open-label, randomised, phase 2 trial conducted at 12 hospitals in Australia, one in New Zealand, and one in Italy. Participants had to be aged 18 years or older; have an Eastern Cooperative Oncology Group performance status of 0-1, clinical stage I (cT1cN0) or II (cT1cN1, cT2cN0-1, or cT3cN0), oestrogen receptor expression of less than 1%, and progesterone receptor expression of less than 10%; had to be HER2 negative; and have previously untreated operable breast cancer with adequate organ function. Participants were stratified according to age and randomly assigned (1:1) centrally using a computer-generated sequence with a minimisation algorithm to either nivolumab 240 mg then 2 weeks later nivolumab 360 mg and carboplatin AUC5 every 3 weeks with concurrent paclitaxel 80 mg/m once per week for 12 weeks (lead-in group) or concurrent nivolumab 360 mg and carboplatin AUC5 every 3 weeks with once per week paclitaxel 80 mg/m for 12 weeks then 240 mg nivolumab 2 weeks later (concurrent group). Data were collected from registration until the 100-day safety follow-up visit, and survival follow-up continues. The primary endpoint was pathological complete response (ypT0/Tis ypN0) at the time of surgery, analysed in each group separately and in all patients who received at least one dose of all three study treatment (modified intention-to-treat population). The trial is registered with the Australian New Zealand Clinical Trials Registry, ACTRN12619001308189, and EudraCT, 2019-003465-18, and is ongoing.

Findings: Between July 6, 2020, and April 1, 2022, 124 participants were enrolled and 14 were ineligible. 110 participants were randomly assigned and 108 were included in the modified intention-to treat analysis (53 in the lead-in group and 55 in the concurrent group). Median follow-up was 12 months (IQR 7-18). All patients were female with a median age of 49 years (IQR 43-60). 18 (17%) patients had clinically node-positive disease; 37 (34%) had clinical stage I, 70 (65%) had stage II, and one (1%) had stage III disease. The pathological complete response rate was observed in 27 (51% [39-63]) of 53 patients in the nivolumab lead-in group and in 30 (55% [43-66]) of 55 in the concurrent group. Treatment-related grade 3-4 adverse events occurred in 70 (65%; 32 [60%] of 53 in the lead-in group and 38 [69%] of 55 in the concurrent group) of 108 patients, with the most common being decreased neutrophil count (25 [47%] of 53 in the lead-in group vs 28 [53%] of 55 in the concurrent group), anaemia (six [11%] vs ten [19%]), and increased alanine aminotransferase (three [6%] vs three [6%]). Serious adverse events were reported in 16 (30%) patients in the lead-in group and 26 (47%) in the concurrent group. Treatment-related serious adverse events occurred in seven (13%) patients in the lead-in group and 20 (36%) in the concurrent group. No treatment-related deaths occurred during the study.

Interpretation: While this study did not support the hypothesis that lead-in nivolumab before chemotherapy was associated with a pathological complete response advantage, high pathological complete response rates were reached supporting shorter duration, non-anthracycline regimens in patients with newly diagnosed triple-negative breast cancer. Future trials are warranted to compare this regimen with the current standards of care.

Funding: Breast Cancer Trials and Bristol Myers Squibb.