Neoadjuvant nivolumab (N) plus weekly carboplatin (C) and paclitaxel (P) in resectable locally advanced head and neck cancer.

6583 Background: Despite multimodality standard therapy, patients (pts) with resectable locally advanced squamous cell carcinoma of the head and neck (LA SCCHN) are at high risk for recurrence. Pts with pathologic complete response (pCR) or major pathologic response (MPR) to neoadjuvant chemotherapy have improved overall survival. PD-1 checkpoint inhibitors are approved in combination with platinum-based chemotherapy in the 1st-line treatment of recurrent/metastatic SCCHN. We hypothesize the addition of N to wkly carboplatin C and P will increase the pCR rate at the primary site compared to historical controls. Methods: This is an investigator-initiated trial for pts with newly diagnosed (AJCC 8 th ) stage III-IV HPV- (oral cavity (OC), oropharynx (OP), hypopharynx (HP), and larynx (L) or stage II-III HPV+ OP SCCHN without distant metastasis who are surgical candidates. Neoadjuvant chemo starting d1 is C AUC 2 IV wkly x 6 plus P 100 mg/m2 IV wkly x 6 plus N 240 mg IV q 2 wks x 3 with surgery on wk 8. The primary endpoint is pCR at the primary site. To estimate pathologic response, the resected pathology specimens are cut >1 section/cm. Using the Aperio Digital scanning system, slides are imaged, and then annotated by at least 2 pathologists for viable tumor vs. treatment effect with areas automatically calculated to yield the percentage of viable tumor. Our primary endpoint will be reached if 11/37 planned pts have a pCR at the primary site. Results: From 11/17-12/19, 27 pts received the study regimen and had surgery (1/27 had an unknown primary; thus, inevaluable for the primary endpoint). Of 27 pts, median age was 59 (46-83), women 31%, HPV+ 15%, OC 73%, OP 19%, HP 7%, L 4%; stage III 33%, stage IVA 67%. Gd 3 toxicities were in 37% pts; 1 pt febrile neutropenia, 3pts anemia, 1pt diarrhea, 1pt cellulitis and 1pt rash. Four pts had gd 3-4 neutropenia. Dose reductions were in 2 pts, and 4 pts had 1 wkly dose dropped. All 27 pts went to surgery, none with PD by CT; all with negative margins. One pt died with rapid recurrence; no other recurrences (median f/u 13 mos). Our primary endpoint was met; 11/26 (42%) pts (excluding pt with unknown primary) had a pCR at the primary site. 9/23 (39%) HPV- pts, had a pCR. MPR or pCR was 18/26 (69%) and in HPV- pts, 15/23 (65%). 2/11 pts had microscopic residual disease in 1 LN each. Conclusions: The combination of N and wkly PC was well tolerated. The primary endpoint of pCR at the primary site in > 11/37 pts was met with the 27 th pt. Accrual continues. Exploratory outcomes assessing markers of immune bias in tumor tissue and plasma are in process. Clinical trial information: NCT03342911 .