Exploring oral microbiome in oral squamous cell carcinoma across environment-associated sample types.

The relationship between the oral microbiome and oral squamous cell carcinoma (OSCC) has been extensively investigated. Nonetheless, most previous studies were single-center, resulting in the absence of systematic evaluations. To address this gap, we performed a comprehensive meta-analysis on 1,255 samples from OSCC-related 16S rRNA gene data sets, representing a diverse range of OSCC phenotypes. It is recognized that the progression of cancer is related to the alterations in the microbiome among different phenotypes. Our findings revealed distinct microbiome characteristics among different sample types, with Biopsy (Bios) and Swab samples exhibiting significant differences between phenotypes. In Bios samples, the microbiomes of the Cancer group and the normal tissue adjacent to the tumor (NAT) group display a higher similarity, while both differ from the microbiome of the Fibroepithelial polyp (FEP) group. Moreover, the identified differential genera and pathways corresponded with these observations. We developed a diagnostic model using the random forest algorithm on Swab samples, achieving an area under the receiver operating characteristic curve (AUC) of 0.918. Importantly, this model exhibited considerable effectiveness (AUC = 0.849) when applied to another sequencing platform. Taken together, our study provides a comprehensive overview of the oral microbiome during various OSCC progression stages, potentially enhancing early detection and treatment.IMPORTANCEThis study answers key questions regarding the universal microbial characteristics and comprehensive oral microbiome dynamics during oral squamous cell carcinoma (OSCC) progression. By integrating multiple data sets, we examine the following critical aspects: (1) Do different sample types harbor distinct microbial communities within the oral cavity? (2) Which sample types offer greater potential for investigating OSCC progression? (3) How are the oral microbiomes of the Cancer group, normal tissue adjacent to the tumor group, and Fibroepithelial polyp group related, and what is their potential association with OSCC development? (4) Can a diagnostic model based on microbial signatures effectively distinguish between Cancer and Health groups using Swab samples?