Supervised Learning and Multi-Omics Integration Reveals Clinical Significance of Inner Membrane Mitochondrial Protein (IMMT) in Prognostic Prediction, Tumor Immune Microenvironment and Precision Medicine for Kidney Renal Clear Cell Carcinoma

Int J Mol Sci 2023 AI 6 Explanations View Original
Original Paper (PDF)

Unable to display PDF. Download it here or view on PMC.

Plain-English Explanations
Pages 1-2
IMMT and Its Relevance to Clear Cell Renal Cell Carcinoma

The Inner Membrane Mitochondrial Protein (IMMT), also known as Mitofilin, is a structural component of the inner mitochondrial membrane that maintains cristae morphology. Mitochondrial dysfunction is a hallmark of many cancers, and IMMT alterations have been linked to metabolic reprogramming in several tumor types.

Clear cell renal cell carcinoma (ccRCC) is the most common subtype of kidney cancer, characterized by dysregulated metabolic pathways including the Warburg effect and altered lipid metabolism. The mitochondrial inner membrane is a central hub of oxidative phosphorylation, making IMMT a plausible regulator of the metabolic vulnerabilities in ccRCC.

Despite its structural importance, the clinical significance of IMMT expression in ccRCC, its relationship to the tumor immune microenvironment, and its potential as a therapeutic target or prognostic biomarker have not been systematically explored prior to this study.

TL;DR: IMMT is a mitochondrial structural protein that may regulate the metabolic vulnerabilities of clear cell RCC, but its prognostic and immunological significance in kidney cancer had not previously been characterized.
Pages 2-4
Multi-Omics Data Sources and Analytical Framework

The study integrated data from multiple large-scale resources. Gene expression, mutation, copy number variation, and clinical outcome data for 545 ccRCC patients were obtained from The Cancer Genome Atlas (TCGA-KIRC), providing a comprehensive molecular and clinical portrait of each case.

Additional validation and functional data were drawn from GEO transcriptomic datasets, the TIMER immune cell deconvolution platform, and the Cancer Cell Line Encyclopedia (CCLE), enabling cross-resource validation of key findings and investigation of IMMT function across cell line models.

Supervised machine learning methods including LASSO regression, random forest, and support vector machine were applied to identify IMMT-related gene signatures predictive of overall survival, constructing a multi-gene prognostic model that outperformed IMMT expression alone.

Immune cell infiltration analysis using CIBERSORT, ESTIMATE, and ssGSEA algorithms quantified the composition of the tumor immune microenvironment in relation to IMMT expression levels, linking molecular biomarker status to immunological context.

TL;DR: Multi-omics data from 545 TCGA-KIRC patients were analyzed using supervised machine learning and immune deconvolution algorithms to characterize IMMT's prognostic value and immunological associations in ccRCC.
Pages 4-6
IMMT Expression and Prognostic Significance

IMMT was found to be significantly downregulated in ccRCC tumor tissue compared to adjacent normal kidney, and lower IMMT expression correlated strongly with worse overall survival, higher tumor grade, and more advanced pathological stage in the TCGA-KIRC cohort.

Kaplan-Meier survival analysis and multivariate Cox proportional hazard regression confirmed IMMT expression as an independent prognostic factor after adjusting for clinical covariates including age, gender, stage, and grade, establishing its potential as a standalone biomarker.

The machine learning-derived prognostic gene signature incorporating IMMT and its co-expressed network genes achieved a time-dependent AUC of approximately 0.75 at 5-year survival prediction, outperforming clinical staging alone and several previously published ccRCC gene signatures.

Functional enrichment analysis of IMMT-correlated genes revealed significant enrichment in oxidative phosphorylation, mitochondrial metabolism, and fatty acid oxidation pathways, consistent with IMMT's known role in maintaining the inner mitochondrial membrane structure required for efficient ATP production.

TL;DR: IMMT downregulation in ccRCC independently predicts worse survival, and a machine learning-derived IMMT-related gene signature achieves an AUC of 0.75 at 5-year survival prediction, outperforming clinical staging.
Pages 6-8
IMMT and the Tumor Immune Microenvironment

Immune deconvolution analysis revealed that low IMMT expression was associated with increased infiltration of immunosuppressive cells including M2 macrophages and regulatory T cells, alongside decreased infiltration of cytotoxic CD8+ T cells. This pattern is characteristic of an immunosuppressed tumor microenvironment that supports tumor immune evasion.

Tumor purity, estimated by the ESTIMATE algorithm, was lower in IMMT-low tumors, further indicating greater immune infiltration that paradoxically correlates with worse outcomes, consistent with the concept that immunosuppressive infiltrates rather than cytotoxic infiltrates dominate the tumor microenvironment in poor-prognosis ccRCC.

IMMT expression also correlated significantly with immune checkpoint gene expression, including PD-L1 and CTLA-4, suggesting that IMMT status may serve as a predictor of response to immune checkpoint inhibitor therapy in ccRCC patients.

These findings position IMMT as a link between mitochondrial function and immune landscape, potentially explaining why metabolically dysregulated tumors with low IMMT expression also tend to have immunologically cold or immunosuppressed microenvironments.

TL;DR: Low IMMT expression is associated with immunosuppressive infiltrates, higher immune checkpoint gene expression, and lower cytotoxic T cell abundance, suggesting IMMT status may predict immune checkpoint inhibitor response in ccRCC.
Pages 8-10
Drug Sensitivity Analysis and Precision Medicine

Using the GDSC drug sensitivity database linked to CCLE cell line genomic profiles, the study identified compounds whose efficacy correlates with IMMT expression levels across cancer cell lines. Lestaurtinib, a multi-kinase inhibitor targeting FLT3, JAK2, and TrkA, emerged as the top candidate with the strongest predicted sensitivity in IMMT-low ccRCC models.

Several other targeted agents including sunitinib-related kinase inhibitors showed differential sensitivity based on IMMT expression, consistent with the known role of VEGF pathway inhibitors in ccRCC and suggesting that IMMT status might stratify patients likely to benefit from specific targeted therapies.

The precision medicine analysis demonstrates how integrating biomarker expression data with drug sensitivity databases can generate testable hypotheses about which patients are most likely to respond to specific agents, moving toward molecularly guided treatment selection beyond standard histological subtyping.

TL;DR: Drug sensitivity analysis identified lestaurtinib as a top candidate for IMMT-low ccRCC, demonstrating how multi-omics biomarker profiling can generate specific drug-response hypotheses for precision medicine.
Pages 10-13
Conclusions and Translational Significance

This study establishes IMMT as a clinically significant biomarker in ccRCC, integrating evidence from prognostic modeling, immune landscape analysis, and drug sensitivity prediction into a comprehensive multi-omics portrait of IMMT's role in this disease.

The multi-disciplinary approach illustrates the power of combining supervised machine learning with large-scale genomic databases to extract biological insight that would be difficult to obtain from any single data source or analytical technique alone.

Translational next steps include experimental validation of lestaurtinib efficacy in IMMT-low ccRCC cell lines and patient-derived organoids, prospective biomarker validation studies using clinical cohorts with matched treatment and outcome data, and investigation of the mechanistic link between IMMT-mediated mitochondrial integrity and immunosuppressive microenvironment formation.

If validated, IMMT expression profiling could be incorporated into routine molecular workup of ccRCC to stratify patients for checkpoint inhibitor versus targeted therapy, representing a step toward truly personalized treatment for a cancer that remains difficult to manage in its advanced stages.

TL;DR: IMMT is a prognostic and immunological biomarker in ccRCC with drug sensitivity implications, and its validation as a precision medicine tool could help stratify patients between checkpoint inhibitor and targeted therapy approaches.
Citation: Open Access, 2023. Available at: PMC10218256.