Microbiota proteomics profiles in muscle-invasive bladder carcinoma related to response to neoadjuvant chemotherapy.

Abstract

4589 Background: Muscle-invasive bladder carcinoma (MIBC) poses significant challenges due to high recurrence and mortality rates, coupled with the toxicity of neoadjuvant chemotherapy (NACT). This has driven the search for biomarkers to improve treatment management and patient quality of life. Methods: Fifty-eight FFPE samples from MIBC patients obtained from transurethral resection (TURBT) were studied. Microbiota analysis was performed by amplification and sequencing of the V4 variable region of the 16S rRNA gene and using Qiime2 software for taxonomic identification. Proteins were extracted and digested from TURBT samples and analyzed by mass spectrometry with data-independent acquisition. For protein identification, a reference database was built, including both the human proteome and bacteria genera proteomes identified by 16S experiments. Proteomics data were processed with Perseus and analyzed using probabilistic graphical models (PGMs) and hierarchical clustering. Results: We have information about treatment response for 56 patients. Twenty-four patients achieved a pathological complete response (43%), with a median disease-free survival of 22 months, and a median overall survival of 29.23 months . 151 bacteria genera identified by 16S experiments, were included in the metaproteomics database. In proteomics experiments, 42 bacteria and 5,111 human proteins were identified. After applying quality criteria, 13 bacteria proteins were used for the subsequent analyses. Hierarchical clustering analysis identified three groups with different microbiota protein profiles: Microbiota1, Microbiota2 and Microbiota3. These groups showed significant differences in response to NACT. A higher proportion of non-responders (73%) vs. responders (27%) was observed in Microbiota2 compared to the other groups (54% in Microbiota1 and 36% in Microbiota2), whereas a predominance of responders (64%) was observed in Microbiota3 (46% in Microbiota1 and 27% in Microbiota2) (p= 0.0481). In a previous work, our group defined three proteomics-based groups related to response to NACT (Layer1 (1.1, 1.2 and 1.3)) (Pinto et al., SEOM 2024). Significant differences were also observed in the distribution of Layer1 between the microbiota clusters. Microbiota2 had a higher representation of patients belonging to Layer1.3, characterized by a majority of non-responders and Microbiota3 had a higher proportion of patients from Layer1.1. Conclusions: To our knowledge, this is the first metaproteomics study in FFPE samples from bladder carcinoma patients for biomarker discovery. Three distinct microbiota protein profiles were identified, one with higher proportion of non-responders. The potential role of these bacteria in NACT response needs further study, highlighting metaproteomics as a promising avenue for biomarker development.