The Composition Alteration of Respiratory Microbiota in Lung Cancer

Background: Previous studies have reported mixed results regarding the composition of respiratory microbiota in lung cancer patients. Therefore, relying on previously published studies, we sought to estimate the relative proportion of respiratory microbiota between lung cancer cases and controls.Methods: MEDLINE, Embase, The Cochrane Library, and Web of Science online databases were systematically searched from inception up to October 14, 2019, to retrieve relevant studies. The relative abundance of each predominant taxon of respiratory microbiota in lung cancer patients and controls was pooled using the reported outcome data.Results: A total of 8 studies comprising 530 participants were included in the final analysis. The pooled phylum level analysis revealed that Bacteroidetes and Proteobacteria were the most abundant bacterial phyla among all participants, recording 17.5%, 47.5% in lung cancer patients, 28.2%, 39.27% in patients with benign pulmonary diseases and 40.62%, 32.09% in healthy controls, respectively. In addition, Actinobacteria and Firmicutes phyla were abundant in lung cancer cases compared to other groups (14.8%, 17.62% for lung cancer versus 13.04%, 13.16% for benign pulmonary nodules and 12.43%, 12.45% for healthy controls). At genus level, Prevotella was predominant in all the participants, and its proportion was relatively lower in cancer patients (25.74% for lung cancer versus 35.59% and 36.75% for benign pulmonary nodules and healthy controls, respectively). Comparatively, Streptococcus was more abundant in lung cancer cases (9.65% in lung cancer versus 7.98%, 7.26% in benign pulmonary nodules and healthy controls).Conclusions: The respiratory microbiota composition of respiratory microbiota significantly differs between lung cancer patients and healthy individuals, and may be used as potential biomarker of lung cancer. In addition, larger sample size, standardized procedures, dynamic monitoring, metabolomics, and culturomics are needed to confirm these results.