Author of

• 30943

  +

  P2.04 - Immuno-oncology (ID 167)

  • Event: WCLC 2019
  • Type: Poster Viewing in the Exhibit Hall
  • Track: Immuno-oncology
  • Presentations: 1
  • Moderators:
  • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall

  • 30969

    +

    P2.04-25 - Gut Mycobiome and Metabolic Interactions with Bacteria in Lung Cancer Patients Reveals Potential Therapeutic Vulnerabilities (ID 2080)

    10:15 - 18:15  |  Author(s): Nagy Dora
    • Abstract

    Background
    

    There is a lack in our understanding of pathogenesis and mechanisms accounting for large variability in response to systemic therapy. Recent data suggest that the gut-lung axis regulates systemic immune function. Moreover, in vivo and clinical studies, mainly in melanoma and mixed epithelial tumors, suggest the role of gut bacteria and response to systemic therapy. However, the presence and potential theranostic role of gastrointestinal (GI) mycobiome in lung cancer has not been explored. Here, we aimed to evaluate the associations of GI bacteria and Candida species in lung cancer patients.

    Method
    

    We included 124 stool samples from 98 lung cancer patients (adenocarcinoma (n=48), squamous cell (n=24), small cell lung cancer (n=15) and other (n=11). Patients underwent lung resection surgery (n=20) or treated with first line chemotherapy CHT (n=78). We analyzed the gut microbiome according clinicopathological variables using Internal transcribed spacer (ITS) and shotgun metagenomic sequencing technique. We performed Spearman correlation between gut microbial species and continuous variables and random forest model (RF) for feature selection. Pathway analysis was done using HUmann2 pipeline, bacterial species annotation was performed with Metaphaln2 pipeline and for fungi ITS, we used PIPITS pipeline. We compared Candida diversity with healthy controls from the Human Microbiome Project.

    Result
    

    We identified Candida species in 65% (71 of 124) of the stool samples. Of these, 48 were baseline 23 were follow-up samples (treatment). 77% of the patients included were diagnosed with advanced stage disease. There were significant differences in Candida abundance in healthy controls vs. cancer patients. In contrast, there were no significant differences in alpha and beta diversity between baseline and follow-up samples (treatment). In total 46 significant species and 140 pathways were significant different and fed into RF. Species belonging to Actinobacteria, Bacteroidetes and Firmicutes phyla are important features in RF and found to be negatively correlated (r <= -0.3, and p < 0.05) with Candida species. Certain bacteria pathways were significantly different according to the presence of Candida.

    Conclusion
    

    Candida is present in stool samples in the majority of lung cancer patients both at diagnosis and during systemic therapy. There were associations with certain gut bacteria and Candida species that may have potential future therapeutic implications. Further biomarker studies in well-defined homogenous subgroups are ongoing in order to identify the exact biomarker role of the mycobiome in lung cancer.