br Objectives Immunotherapy IO is e ective in metastatic
Objectives: Immunotherapy (IO) is eﬀective in metastatic Non Small Cell Lung Cancer (NSCLC). Gut microbiota has an impact on immunity and its imbalance due to 2642-37-7 may impair the eﬃcacy of IO. We investigated this topic in a case series of NSCLC patients treated with IO.
Materials and Methods: Data about all metastatic NSCLC patients treated with IO between 04/2013 and 01/2018 were collected. Patients were stratified according to antibiotic use during the Early IO Period (EIOP), and ac-cording to the Antibiotic-Immunotherapy Exposure Ratio (AIER) defined as “days of antibiotic/days of IO” during the Whole IO Period (WIOP). Survival was estimated using the Kaplan-Meier method. Log-rank test was used to compare the curves. Multivariate analyses were performed with the Cox model.
Conclusion: Although no diﬀerence in outcome were observed with antibiotic use in the EIOP, a detrimental eﬀect became evident for patients with a higher AIER in the WIOP. If its relevance is confirmed, AIER may become an innovative variable for estimating the impact of antibiotics on IO eﬃcacy.
Metastatic Non Small Cell Lung Cancer (NSCLC) is the leading cause of cancer-related deaths worldwide .
Only a limited subgroup of patients harbors genetic alterations amenable of targeted treatments (i.e. EGFR, ALK, ROS-1, MET). For the remaining ones, platinum-based chemotherapy was the only available option until a few years ago .
In the last years, diﬀerent clinical trials showed superiority of im-munotherapy (IO) over standard chemotherapy in terms both of eﬃ-cacy and toxicity, in metastatic NSCLC [3–8]. This has led first to the approval of anti-PD1 and anti-PD-L1 agents in second and more ad-vanced lines of therapy [3–7], then in first line setting . Many other
E-mail address: [email protected] (G. Lo Russo).
compounds with similar or slightly diﬀerent mechanism of action are currently under investigation, with promising results [9–15]. These studies will probably lead to an expansion of the indications for IO, likely also extending to non-metastatic stages of disease and to com-binations with other immunotherapeutic or cytotoxic agents [9–15]. Despite such promising premises, only 25–30% of patients appear to derive a durable benefit from IO and the event of primary resistance, up to its worst expression arising as hyper-progression, is well known . This variability strengthens the importance to identify predictive fac-tors of response to IO [17–22].
Emerging evidence suggests that not directly tumor-associated fac-tors may contribute to the response to cancer therapy. Bacteria in-habiting the gut, collectively named as gut microbiota, maintain host
physiology and health by exerting fundamental functions, spanning from metabolic to immunomodulatory properties .
It was recently demonstrated that gut microbiota constitutes one of the environmental factors aﬀecting response to chemotherapeutic and immunotherapeutic drugs through its ability to regulate the immune response. Indeed, a clear cause-eﬀect relationship between the com-position of gut microbiota and the eﬃcacy of both chemotherapy and IO has been showed in mouse models [24–27]. Accordingly, antibiotic-induced dysbiosis in tumor-bearing mice has been associated with the failure of IO containing anti-CTLA4 and anti-PD1 antibodies [26–28]. Moreover, fecal microbiota transplantation of mice with stool of re-sponders and non-responders to IO transfers the capability to respond or not to IO [26,28,29]. In addition to these experimental findings, analyses in human cohorts of metastatic patients aﬀected by diﬀerent malignancies showed that the use of broad-spectrum antibiotics which are known to severely reduce the bacterial diversity and the function of intestinal flora, around the beginning of IO, has a detrimental eﬀect on patient response and Progression-Free Survival (PFS) [28,29]. In par-ticular, Derosa et al. in a retrospective cohort of patients with meta-static renal cell carcinoma and NSCLC treated with IO, showed that the administration of antibiotics within one month before the beginning of treatment has a detrimental eﬀect on Response Rate (RR) and PFS . Routy et al. confirmed this evidence in a large cohort of patients with diﬀerent malignancies (NSCLC, renal and urothelial carcinomas), showing that cases receiving antibiotics between 2 months before and 1 month after the first IO administration had worse PFS and OS than their non-treated counterparts. The same researchers performed a molecular characterization of microbiota through shotgun sequencing of stool DNA, finding that the clinical response to Immune Checkpoint In-hibitors (ICIs) is correlated to the abundance of Akkermansia mucini-phila. These data were subsequently confirmed in mouse models, in which the transplant of a fecal microbiota rich in Akkermansia and Alistipes induced enhanced response to ICIs potentiating T-mediated response . Another work by Gopalakrishnan et al. prospectively studied patients with metastatic melanoma treated with IO. The pa-tients were classified as responders if they achieved at least disease stability for 6 months, or as non-responders, provided that they showed progressive disease within the first 6 months. The researchers identified significant diﬀerences in the composition of bacterial flora between the two subgroups, with a predominance of Clostridiales, Fecalibacterium and Ruminococcaceae in the stool specimens obtained from responders, and a prevalence of Bacteroidales, Escherichia and Anaerotruncus in the stool specimens obtained from non-responders. Again, these results were confirmed and replicated in germ-free mice, which achieved a better response to IO after fecal transplants with stool specimens of responding patients . A fourth work by Matson et al. analyzed a similar population of melanoma patients treated with IO reporting that cases showing an objective tumor response had basal stool samples enriched in Bifidobacterium longum, Collinsella aerofaciens and En-terococcus faecium. Germ-free mice transplanted with these specimens showed improved response to anti-PD-L1, with an increase in T-cell infiltrates in tumor masses .