No abstract

e21729 Background: A neutrophil/lymphocyte ratio (NLR)> 5 or its derivative (dNLR)> 3 have been proposed as a prognostic factor for non-effectiveness of nivolumab in non-small-cell lung cancer (NSCLC). Filgrastim (GSGF) increases the absolute counts of neutrophils in patients (pts) with neutropenia. The aim of this study is to determine whether the use of GSGF, during the first line therapy of nSCLC, influences the effectiveness of the anti-PD1/PDL1 inhibitors (aPD1-i) used in subsequent line. Methods: We designed a single center, observational and retrospective study. We included NSCLC pts treated with Nivolumab (NIV), Pembrolizumab (PEM) or Atezolizumab (ATE) between January-16 and November-19 after primary platinum-based chemotherapy at one hospital from south Spain. The primary endpoints were overall survival (OS) and progression-free survival (PFS). Results were assessed using Kaplan-Meier method for GSGF-use stratified by NLR, dNLR, or aPD1-i. Results: 79 pts were enrolled (91% men), mean age was 66.7 years, 78.8% of pts had ECOG = 0-1. NIV was used at 55.7%, ATE 25.3% and PEM 19.0% of pts. 50.6 % of tumors were squamous histology. GSGF was used at 19 pts (24.1%). The mean WBC at the start of therapy with aPD1-I was 9300 cell/µL (sd=3900), range 3600-25800 cell/µL. NLR was less than 5 in 46 pts (58.2%) and dNLR was less than 3 in 51 pts (64.6%). The mean time from the start of line previously to start the treatment with aPD1-i was 9.7 months (sd=8.6). The median for treatment duration was 4.1 months [IC 95%; 2.3-5.8].OS was 7.3 months [IC 95%:3.4-11.3] and PFS was 4.4 months [IC95%:2.3-6.5]. We found statistically significant difference in OS between patients with dNLR <3, median 13.9 months [IC95%; 5.4-22.4] vs patients with dNLR> 3, median 3.6 months [IC95%; 2.4-4.7], p=0.037. The median OS was higher in pts GSGF-treated (6.1 months) vs pts no GSGF-treated (3.3 months), p=0.347, no reaching statistical significance. No difference was found in OS by NLR. The analysis for OS by GSGF use stratified by dNLR is shown in the table. Conclusions: The effectiveness of aPD1-i in nSCLC patients previously treated and with dNLR>3, a priori poor prognosis, is greater in the group of patients treated with GSGF, reaching OS like pts with dNLR <3. [Table: see text]

Purpose: To determine whether the use of G-CSF during the first-line therapy of nSCLC influences the effectiveness of anti-PD1/PD-L1 drugs (aPD1-i) used in subsequent therapies.Methods: We designed a single-center, observational, retrospective study with nSCLC patients treated with aPD1-i (nivolumab, pembrolizumab, or atezolizumab) after primary platinum-based chemotherapy in southern Spain. The primary endpoints were overall survival (OS) and progression free survival (PFS). The results were assessed using the Kaplan-Meier method for G-CSF-use.Results: We enrolled 79 patients (91% male) with a mean age of 66.7 years. Nivolumab, atezolizumab, and pembrolizumab were used in 55.7%, 25.3%, and 19.0% of the patients, respectively. G-CSF was administered to 19 patients. The median treatment duration was 4.1 months (95%CI: 2.3-5.8). Overall survival (OS) and progression free survival (PFS) were 7.3 (95%CI: 3.4-11.3) and 4.4 months (95%CI: 2.3-6.5), respectively. The median OS was 15.8 months (95%CI: 1.2-30.3) in G-CSF-treated patients and 5.3 months (95%CI: 2.3-8.3) in the non-treated group (p=0.186). Higher OS was observed in all subgroups.Conclusions: The effectiveness of aPD1-i in previously treated nSCLC patients was higher in the G-CSF-treated group than in the non-treated group, although the difference was statistically insignificant. These results may contribute to improving the effectiveness of anti-PD1/PD-L1 drugs, although confirmatory prospective studies are required.