Radiomics Method Predicts FOXA1 Mutations in Prostate Cancer

Strategic Insights -

Researchers at Fudan University Shanghai Cancer Center have developed a radiomics method using magnetic resonance imaging (MRI) to predict FOXA1 gene mutations in prostate cancer patients. The study, published in Scientific Reports, focused on analyzing clinicoradiological and MRI radiomics data from 367 prostate cancer patients to identify features associated with FOXA1 mutations.

Key Takeaways:

  • The study developed a model to predict FOXA1 mutations in prostate cancer patients using radiomics features extracted from MRI images.
  • The model was trained and validated on data from three centers, with a total of 109 patients with FOXA1 mutations and 258 without.
  • The radiomics method using a random forest classifier was found to have higher accuracy in predicting FOXA1 mutations compared to logistic regression and support vector machine models.
  • The study selected three radiomics features to build models, with area under the receiver operating characteristic curve (AUC) values of 0.82 and 0.81 in internal and external validation sets, respectively.
  • The research concluded that radiomics method with random forest classifier could effectively predict FOXA1 mutations in prostate cancer.
  • Ruchuan Chen, Dept. of Radiology, Fudan University Shanghai Cancer Center, led the research team.
  • Additional authors include Lin Deng, Bingni Zhou, Guoqing Hu, Hualei Gan, Ling Zhang, Liangping Zhou, Kefu Liu, and Xiaohang Liu.

Statistics:

  • 367 prostate cancer patients were included in the study.
  • 109 patients had FOXA1 mutations, while 258 did not.
  • The radiomics method using random forest classifier achieved AUC values of 0.82 and 0.81 in internal and external validation sets, respectively.
  • The AUC values of logistic regression and support vector machine models were 0.74 and 0.71, and 0.60 and 0.65, respectively.

Sources:

  • The value of MRI-based radiomics and clinicoradiological data for the detection of forkhead box protein A1 gene mutated prostate cancer. Scientific Reports, 2025;15(1):22929.
  • Nature Portfolio, Heidelberger Platz 3, Berlin, 14197, Germany.
  • Fudan University Shanghai Cancer Center, No. 270, Dong-An Road, Shanghai, 200032, People's Republic of China.