ESTRO 2022 - Abstract Book

S433

Abstract book

ESTRO 2022

S. Ronchi 1 , B. Vischioni 1 , A. Cicchetti 2 , M. Bonora 1 , R. Ingargiola 1 , E. Rossi 1 , S. Russo 1 , M. Ciocca 1 , E. Orlandi 1

1 National Center for Oncological Hadrontherapy (CNAO), Radiation Oncology Clinical Department, Pavia, Italy; 2 Fondazione IRCCS Istituto Nazionale dei Tumori, Prostate Cancer Program, Milan, Italy Purpose or Objective To retrospectively evaluate clinical outcome and toxicity in patients (pts) with head and neck mucosal melanoma (HNMM) treated with carbon ion radiotherapy (CIRT) at our institution. Materials and Methods HNMM pts enrolled for CIRT after surgery or diagnostic biopsy from June 2013 to June 2020 were included in this analysis. Local relapse-free survival (LRFS), progression-free survival (PFS), distant metastasis-free survival (DMFS) and overall survival (OS), were calculated according to the Kaplan-Meier Method for different patient characteristics. Prognostic factors were also collected and analyzed using KMM and Cox Regression Analysis. Early and Late side effects were assessed using CTCAE v.5.0 criteria. Results Overall, 40 pts were included for analysis. Table 1 shows pts’ characteristics. Twenty-eight (70%) pts were treated after surgery, 10 (25%) with exclusive CIRT, 2 (5%) pts received systemic therapy before and after CIRT (chemotherapy in one case, immunotherapy in the other). CIRT treatment schedules were given in 16 fractions (4 fr/week) for a total dose of 65.6 RBE Gy and 68.8 RBE Gy in 22(55%) and 18(45%) pts, respectively. Moreover, 18 pts (44%) received immunotherapy after CIRT as part of the treatment for the primary tumor or at the relapse. Median follow-up (FU) time was 18 mo (range 5-81 mo). Two-years- LRFS, PFS, DMFS and and OS were 84.5%, 33.2%, 37.3% and 58.6% respectively (see first row in Fig1). No difference in clinical outcome was found for resected compared to unresected patients. Naïve tumor status was associated with better LRFS (HR=8.4 and p-value<0.01, Fig1 E) and PFS (HR=2.4 and p=0.03, Fig1 F). Better OS was reported for T3 stage (HR=15.6, p= 0.001, Fig 1G) and age years using a univariate cox regression model. These variables were considered in a bivariate model. HR was 4.3 (p<0.01) for T4 and 3.9 (p=0.03) for age years, with an AUC=0.81. To analyse the relationship between immunotherapy and CIRT we found that combined treatment after CIRT in T4 stages was associated with better OS outcome at 30 months (40%) than patients not receiving immunotherapy (20%). In patients not receiving immunotherapy after CIRT, T3 and T4 stages maintained significant differences in 3y-OS (100% vs 19.2%), whereas T3 and T4 patients receiving immunotherapy reached similar OS rates (71.4% vs 72.7%). As far as concerned side effects, acute tox at the end of CIRT was G1-G2 in 95% and G3 in 5% of pts. Late tox was G0 in 10% and G1-G2 in 81% of pts; late toxicity ≥ G3 consisted in one G3 unilateral hear impairment and 2 unilateral visual loss of grade G3 and G4, respectively (expected tox).