Validation and Derivation of miRNA-Based Germline Signatures Predicting Radiation Toxicity in Prostate Cancer

Area of Science:

• Oncology
• Genetics
• Radiotherapy

Background:

• Radiotherapy (RT) is a primary cancer treatment, but RT-induced genitourinary (GU) toxicity is a significant challenge for prostate cancer survivors.
• Identifying patients at risk for GU toxicity remains difficult, hindering personalized treatment strategies.

Purpose of the Study:

• To validate the PROSTOX biomarker (miRNA-based germline biomarkers or mirSNPs) for predicting late RT-induced GU toxicity.
• To investigate the potential of mirSNPs in defining other forms of RT-associated GU toxicity, including acute and chronic toxicity.

Main Methods:

• Utilized data from 148 patients in the MIRAGE trial comparing MRI-guided versus CT-guided prostate stereotactic body RT.
• Employed linear regression to assess the association between the PROSTOX score and late GU toxicity.
• Applied machine learning models to develop predictive signatures for acute and chronic GU toxicity, evaluating accuracy with AUC metrics.
• Conducted comparative Gene Ontology analysis to identify unique pathways for different toxicity types.

Main Results:

• The PROSTOX biomarker accurately predicted late GU toxicity with an AUC of 0.76 (p < 1.2E-9).
• mirSNP-based signatures effectively distinguished acute and chronic RT-associated GU toxicity, with AUCs of 0.770 and 0.763, respectively.
• Gene Ontology analysis revealed distinct molecular pathways associated with acute, chronic, and late GU toxicity.

Conclusions:

• These findings strongly support the use of mirSNPs for predicting RT-induced GU toxicity.
• This research paves the way for personalized radiotherapy, potentially improving patient outcomes by mitigating toxicity.