Antibiotic Exposure Ratio Predicts The Prognosis Of Patients With Advanced Non Small Cell Lung Cancer Treated With PD-1 Antibody And Chemotherapy

Background and Objectives:A growing body of data indicate that antibiotic use reduces the efficacy of immunotherapy(10)and chemotherapy(CO)in many types of cancers,but these findings applying to non-small cell lung cancer(NSCLC)remains controversial.The present study analyzed the effect of antibiotic exposure in different time windows and antibiotic duration on the outcomes of advanced NSCLC patients who received immunotherapy and chemotherapy by conducting a retrospective cohort.Materials and Methods:This study included advanced NSCLC patients who received PD-1 inhibitor or platinum-based double-agent chemotherapy between January 2015 and December 2020 at Nanfang Hospital.Using the time of the first immunotherapy/chemotherapy as a starting point,we defined the time within 3 months before and during antitumor therapy(-3m,during treatment)as the whole time period and further subdivided 5 different time windows:(-3m,0m)、(-1m,0m)、(0m,1m)、(-1m,1m)、(0m,during treatment).Patients were grouped according to antibiotic use in different time windows and antibiotic exposure ratio(AER)defined as the "days of antibiotic/days of IO or CO" in the whole period.AER was an empirical variable representing the length,frequency,and intensity of antibiotic use.The study endpoint was progression-free survival(PFS),and we performed adjusted Kaplan-Meier survival curves and Cox proportional hazards regression analyses for PFS using inverse probability of treatment weights(IPTW)to equalize between-group differences in the baseline characteristics.Results:In the IO cohort(122 cases)and CO cohort(62 cases),no statistical difference in PFS was observed in any of the subgroup analyses of antibiotic use in different time windows.Patients with AER higher than a certain cut-off value(approximately 7.0%both in the IO and CO cohort)had worse PFS than the other(IPTW-adjusted PFS of IO cohort:5.8 vs.13.7 months,P=0.03;CO cohort:4.0 vs.7.6 months,P=0.008).AER≥7.0%is an independent risk factor for disease progression in NSCLC patients treated with PD-1 antibody(HR 3.80,95%CI 1.80-7.98,P<0.001)and platinumbased double-agent chemotherapy(HR 2.41,95%CI 1.39-4.19,P=0.002).Conclusion:Prolonged and frequent antibiotic exposure,not the timing of antibiotic administration,was associated with poor outcomes in NSCLC patients treated with IO and CO.AER has the potential to be a helpful predictor in clinical practice.