Outcomes Following Immunotherapy in Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma: A Single-Institution Experience

Abstract

<h3>Purpose/Objective(s)</h3> Pembrolizumab and nivolumab are approved in the treatment of recurrent/metastatic head and neck squamous cell carcinoma (HNSCC). However, real-world effectiveness may vary from clinical trial efficacy, partly due to strict exclusion criteria. We reviewed the effectiveness of immunotherapy (IO) in recurrent/metastatic HNSCC at a single tertiary referral center. <h3>Materials/Methods</h3> All HNSCC patients who received single-agent palliative IO for recurrent/metastatic disease were identified from the electronic medical record via ICD codes under an IRB approved protocol. Wilcoxon rank sums test and chi squared test of independence were used to compare characteristics between IO responders (any response) and non-responders (stable or progressive disease) respectively. Kaplan-Meier method was used to estimate overall (OS) and progression free survival (PFS). Multivariable cox proportional hazards method was used to identify predictors of OS. <h3>Results</h3> A total of 138 patients were identified with the following primary tumor sites: p16+ oropharynx (n=45), p16- oropharynx (21), oral cavity (29), larynx (21), hypopharynx (7), nasal/sinus (7), or unknown primary (8). Concurrent chemoradiation was given to 68.8% (95) of patients and palliative chemotherapy was given prior to IO in 26.1% (36). IO was given for unresectable recurrent or metastatic disease in 23.2% (32) and 76.8% (106) of patients respectively, with 95.7% (132) receiving pembrolizumab and 4.3% (6) receiving nivolumab. Median age was 64 years (IQR 58-71), with ECOG status of 0, 1, and 2 in 23.9% (33), 60.9% (84), and 15.2% (21) of patients respectively. Combined positive scores (CPS) were available in 45.7% (63) of patients, with 33.3% (46) having CPS≥1 and 23.9% (33) having CPS≥20. The median number of IO cycles received was 4 (IQR 2-7). Of the 96 patients who survived to undergo imaging after ≥3 cycles, 43.8% (42) of patients had progression, 15.6% (15) had stable disease, 36.5% (35) had partial responses, and 4.2% (4) had complete responses. There were no significant differences in age, time from definitive radiation to IO, or absolute lymphocyte count (ALC) at IO initiation between responders and non-responders. Median PFS from IO initiation was 3.2 months (95%CI 2.8-4.4) months while median OS from IO initiation was 6.6 months (4.7-10.3). In a multivariable model, ECOG status of 1 (HR=2.00, 1.16-3.45) or 2 (HR=3.03, 1.51-6.09) vs ECOG 0 and decreased ALC at IO initiation (100 cells/µL, HR=1.06, 1.01-1.12) were significant predictors of worse OS. <h3>Conclusion</h3> In this real-world cohort, survival with palliative IO in HNSCC was similar to that in randomized control trials. Though ECOG status and ALC were significant predictors of OS, further work into predictors of OS mediated through evidence of tumor response is needed.