Predictive Model of Objective Response to Nivolumab Monotherapy for Advanced Renal Cell Carcinoma by Machine Learning...

Area of Science:

• Oncology
• Immunotherapy
• Genetics

Background:

• Anti-PD-1 antibodies, like nivolumab, are crucial for treating advanced renal cell carcinoma (RCC).
• Patient response to nivolumab therapy for advanced RCC is variable.
• Predicting treatment efficacy is essential for optimizing patient outcomes.

Purpose of the Study:

• To develop machine learning (ML) models for predicting tumor response to nivolumab in advanced RCC.
• To integrate genetic (single-nucleotide polymorphism - SNP) and clinical data for enhanced predictive accuracy.
• To evaluate the performance of different ML algorithms in predicting objective response and progression-free survival (PFS).

Main Methods:

• Utilized clinical and SNP data from Japanese patients with advanced clear cell RCC treated with nivolumab monotherapy.
• Employed point-wise linear (PWL), logistic regression with elastic-net, and eXtreme Gradient Boosting algorithms.
• Calculated Area Under the Curve (AUC) for objective response and C-indices for PFS to assess model utility.

Main Results:

• The PWL algorithm demonstrated the highest AUC values for predicting objective response across datasets.
• Three predictive models were developed using PWL: a clinical data-only model, a small SNP model (8 SNPs), and a large SNP model (49 SNPs).
• The C-indices for PFS were 0.522 (clinical model), 0.600 (small SNP model), and 0.635 (large SNP model), indicating improved prediction with SNP integration.

Conclusions:

• ML-based SNP models effectively predict tumor response to nivolumab monotherapy in advanced clear cell RCC.
• These predictive models hold significant potential for guiding clinical treatment decisions.
• Integrating genetic data with ML enhances the prediction of immunotherapy response in RCC.