Pathological regression patterns following neoadjuvant chemo-immunotherapy in head and neck squamous cell carcinoma: a pilot study.

Introduction: Neoadjuvant chemoimmunotherapy (NACI) has drawn considerable attention in Head and neck squamous cell carcinoma (HNSCC) owing to its potential in functional preservation and treatment-failure reduction. Yet whether the surgical extent can be narrowed following NACI is largely debatable due to a potential non-centripetal tumor regression may result in scattered microfoci residing beyond the narrowed margin.

Methods: In this pilot study, we characterized the tumor regression pattern in a post-NACI HNSCC cohort using a whole-mount histopathological approach. The MRI examinations before and after NACI were used to evaluate the objective response rate (ORR).

Results: Of the 52 patients enrolled, the ORR was 75%. Pathological complete response (pCR) rate was 15.4%, and the major pathological response (MPR) rate was 40.4%. Two major regression patterns were identified in whole-mount tumor sections, centripetal regression and non-centripetal regression. Centripetal regression was observed in 37 patients (71.2%) and was subcategorized into complete regression (Ia, 15.4%), unifocal centripetal regression (Ib, 36.5%), and multifocal centripetal regression (Ic, 19.2%). Non-centripetal regression was seen in 15 patients (28.8%) and was subcategorized into scattered regression (IIa, 25.0%) and non-regression (IIb, 3.8%). Moreover, we found a pre-NACI CPS higher than 20 or post-NACI (18)F-FDG SUVmax reduction exceeding 50% were potential predictive factors for the centripetal regression pattern.

Discussion: We revealed for that centripetal regression was the predominant pattern of regression after NACI in HNSCC. Hence, our data presumably supports a reduced surgical extent in post-NACI HNSCC patients. Future studies should focus on identifying accurate predictive factors for the regression pattern, which may eventually assist in risk stratification and surgical decision making.

Conclusions: The pattern of tumor pathological regression after NACI for HNSCC is mainly divided into centripetal and non-centripetal regression, with the former accounting for the major portion of the regression.