CHICAGO – Two immune checkpoint inhibitors were better than one as neoadjuvant therapy for patients with resectable early-stage non–small cell lung cancer (NSCLC) in the phase 2 NEOSTAR trial.
Among 44 patients with stage I-IIIA NSCLC who were randomized to either a combination of nivolumab (Opdivo) and ipilimumab (Yervoy) or to nivolumab alone, the combination was associated with higher rates of the primary endpoint of major pathological response (MPR), defined as a reduction in viable tumors cells to 10% or less, reported, from the University of Texas MD Anderson Cancer Center in Houston.
“Nivolumab/ipilimumab induced a 44% MPR rate in resected patients, met the trial prespecified boundary with seven MPRs in the intention-to-treat population, and induced pathologic complete responses in 38% of resected patients,” she said at the annual meeting of the American Society of Clinical Oncology.
To test whether neoadjuvant monotherapy or combination therapy could improve outcomes of standard induction chemotherapy, NEOSTAR investigators enrolled patients with NSCLC stage I-IIIA, including patients with a single involved mediastinal node (N2 single station) who were eligible for surgical resection.
The patients were randomized on a 1:1 basis to receive nivolumab 3 mg/kg on days 1, 15 and 29 alone or in combination with ipilimumab delivered 1 mg/kg on day 1, followed by surgery 3-6 weeks after the last study dose and then postoperative standard-of-care chemotherapy.
Of 53 patients screened, 44 were eligible, with 23 randomized to nivolumab monotherapy and 21 randomized to nivolumab/ipilimumab. Of this group, five did not proceed to surgery (one in the monotherapy arm, four in the combination arm) because of either high surgical risk, lack of respectability, or refusal of surgery. The mean age at randomization was 65.6 years. In all, 18% were never smokers, and the remaining 82% were former or current smokers.
The MPR rate in the intention-to-treat population – the primary endpoint – was reached in four patients (17%) in the monotherapy arm and in seven patients (33%) in the combination arm. As noted, the combination arm reached the prespecified boundary of six or more patients with an MPR. All patients in each arm who had an MPR also had a pathologic complete responses.
Of the 39 patients who went on to resection, 37 were evaluable, and in these patients the respective MPR rates were 19% and 44%. Two patients on nivolumab alone and six on nivolumab plus ipilimumab had 0% viable tumor detectable at the time of surgery. Radiographic responses included one complete response in the combination arm and eight total partial responses, in five and three patients, respectively. The objective response rated was 20%. The responses, assessed by Response Evaluation Criteria in Solid Tumors (RECIST) were positively associated with MPR, Dr. Cascone said.
In 11% of patients, the investigators observed apparent radiographic progression after neoadjuvant immune checkpoint inhibitors in mediastinal and or in nonregional nodes. However, pathological assessment and evaluation of the flaring nodes did not reveal evidence of disease, but instead showed noncaseating granulomas that were not present at baseline.
“Awareness of this phenomenon, which we named the ‘nodal immune flare,’ is of critical importance, as if the clinician fails to distinguish the nodal immune flare from disease progression, potential curative surgery for these patients could be avoided,” she said.
Grade 1 or 2 treatment-related adverse events included rash, itching, fatigue, anemia, cough, and diarrhea. Grade 3 or greater treatment-related adverse events included hypoxia, pneumonia, and pneumonitis in the nivolumab monotherapy arm and diarrhea and hyponatremia in the combination group. One patient treated with nivolumab monotherapy, who had achieved 0% viable tumor, had grade 3 pneumonia and pneumonitis, which was treated with steroids that impeded the healing of a bronchopleural fistula and subsequent empyema. Other surgical complications included air leaks, which occurred in five patients in the nivolumab group and three in the nivolumab plus ipilimumab arm.
Two patients died, one in the monotherapy arm from steroid-treated pneumonitis 4.1 months after randomization and one in the combination arm who had progressive disease 2 months after randomization, and died from the disease 15 months later.
Invited discussant, from Stanford (California) University School of Medicine, commented that the choice of neoadjuvant immunotherapy was not based on molecular markers, “and I think we have a major unmet need for developing biomarkers for personalized treatment in this area.
“Ideally, the biomarkers that we would have in this setting would, A, allow us to identify which patients have micrometastatic disease and therefore are likely to benefit from the upfront systemic therapy and, secondly, also could tell us which neoadjuvant therapy they would respond to, be it immunotherapy, chemotherapy, or the combination,” he added.
The study was supported by Bristol-Myers Squibb. Dr. Cascone disclosed honoraria from the company. Dr. Diehn reported stock ownership, consulting, research funding and travel expenses from various companies.
SOURCE: Cascone T. et al. ASCO 2019, .