ESTRO 2023 - Abstract Book

S1087

Digital Posters

ESTRO 2023

was delivered by Tomotherapy in 70% of pts and by VMAT in 30% of pts. One pt interrupted RT (COVID+). Median number of ChT cycles was 5 (1-6), 70% pts received full ChT dose. The main acute toxicity was an unexpected lymphopenia (93/98 pts: 95%). G3-G4 lymphopenia was observed in 68% pts and 27% of pts had G2 lymphopenia. Three consecutive pts (3%) had bacterial pneumonia, 1 pts had aorto-esophageal fistula and underwent early surgery. No difference in term of mean PTV in pts with G0-G2 (843.1 cc) and pts with G3-G4 lymphopenia (852.3 cc) and in term of median tumor length (G0-G2: 5 cm (1-12 cm); G3-G4: 5 cm (1-15)) was seen. Median time to restaging was 42 days (14-87), 78/99 pts underwent surgery (21 pts excluded from surgery: PD 8 pts; worsening clinical condition 6 pts; died 1pt; lost 2 pt; cCR 2 pts; not yet operated 2 pts). Post-surgical stage was T0: 18 pts, T1: 12 pts, T2: 16 pts, T3: 31 pts, T4: 1 pt; N0: 50 pts, N+: 28 pts. Mandard TRG was TRG1: 18 pts (23%), TRG2: 16 pts (21%), TRG3: 32pts (41%), TRG4: 11 pts (15%), TRG5: 1 pt (1%). In pts with TRG1 (pCR) and TRG2, G3-G4 lymphopenia was observed in 56% (19/34 pts) vs 73% (32/44) of pts who were TRG3 and TRG4. One-year mortality was 24% (16/68) in pts with G3-G4 lymphopenia vs 19% (6/31) for G0-G2. Two-year mortality was 38% (26/68) in pts with G3-G4 vs 32% (10/31) for G0-G2 lymphopenia. Conclusion Lymphopenia seems to impact on responses to the neoadjuvant treatment and to be predictor of the prognosis of pts with EC or EGJC. Other pretreatment predicting factors should be studied to limit the incidence of lymphopenia, in order to assess benefit from checkpoint inhibitors or other lymphocyte-mediated immunotherapies. 1 Fondazione Policlinico Universitario Campus Bio-Medico, Operative Research Unit of Radiation Oncology, Rome, Italy; 2 Università Campus Bio-Medico di Roma, Fondazione Policlinico Universitario Campus Bio-Medico, Research Unit of Radiation Oncology, Rome, Italy; 3 Università Campus Bio-Medico di Roma, Research Unit of Radiation Oncology, Rome, Italy; 4 Università Campus Bio-Medico di Roma, Fondazione Policlinico Universitario Campus Bio-Medico, Research Unit of General Surgery , Rome, Italy; 5 Università Campus Bio-Medico di Roma, Fondazione Policlinico Universitario Campus Bio Medico, Research Unit of General Surgery, Rome, Italy Purpose or Objective Despite advances in the multidisciplinary management of pancreatic cancer, overall prognosis remains poor, due to early progression of the disease. We have proposed a model to select better patients for multimodal treatments in borderline resectable and locally advanced unresectable pancreatic ductal adenocarcinoma (PDAC). Materials and Methods From 2008 through 2022, patients with borderline resectable or unresectable PDAC were enrolled in three consecutive prospective studies of chemoradiotherapy (CRT) with or without induction chemotherapy (IC). In all cases an accurate pre treatment staging including CT scan, FDG-PET/CT and laparoscopy with peritoneal washing was performed. After IC and 4 weeks after the completion of CRT, a re-evaluation was performed regarding tumour response and resectability with CT scan, FDG-PET/CT scan. The primary objective of this secondary analysis was the resection rate. Secondary objectives included progression-free survival (PFS), overall survival (OS), local progression-free survival (LPFS), metastases free survival (MFS). Results Of the 129 eligible patients (69 men, 60 women), median age was 64 years (range, 36–75 years). According to the results of the pre-treatment workup, thirty-nine patients (30.2%) had metastatic disease and were therefore excluded from the protocols. Sixty-eight patients who underwent IC (Gem-Ox or FOLFIRINOX schemes) were evaluated for clinical response after 2 months by using CT scan and PET-CT scan. Eleven patients (16%) experienced disease progression; two patients refused to continue the protocols. Overall, seventy-seven (59.7%) patients received concomitant CRT and were evaluated for clinical response (Figure 1). The median follow-up for all patients was 21 months (range, 5 to 132 months). Thirty-eight patients (60%) underwent surgical radical resection. R0 resection was achieved in 37 of the 38 resected patients (97.4%). Two patients died due to perioperative complications. For the 77 patients undergoing CRT, the median PFS was 13 months ((95% CI, 8 to 20). One-year, 2-yr and 3-yr PFS were 56%, 32%, 22%, respectively. One-year, 2-yr and 3-yr LPFS were 83%, 68% and 60%, respectively. One-year, 2-yr and 3-yr MFS were 65%, 40%, 30%, respectively. Median OS was 17.5 months (95% CI, 12.1 to 22.8). One-year, 2-yr and 3-yr OS were 80%, 37% and 29%, respectively. Patients who underwent resection had a significantly longer median OS compared with non resected patients (37.6 months vs 13 months, p < 0.001). The median PFS for resected patients was 22.5 months compared with 9.5 months for non resected patients (p< 0.001). The median OS and PFS for patients treated by upfront CRT were 14.6 and 9.9 months compared with 19.2 and 17.8 months for patients treated by IC followed by CRT (p<0.05). Figure 1: Study flowchart PO-1356 A new algorithm in pancreatic cancer: secondary analysis of three prospective studies G.M. Petrianni 1 , M. Fiore 2 , P. Trecca 1 , G. D'Ercole 3 , M. Benincasa 3 , E. Ippolito 2 , D. Caputo 4 , R. Coppola 5 , S. Ramella 2