Conditional generative adversarial network driven radiomic prediction of mutation status based on magnetic resonance imaging of breast cancer

BACKGROUND

Breast Cancer (BC) is a highly heterogeneous and complex disease. Personalized treatment options require the integration of multi-omic data and consideration of phenotypic variability. Radiogenomics aims to merge medical images with genomic measurements but encounter challenges due to unpaired data consisting of imaging, genomic, or clinical outcome data. In this study, we propose the utilization of a well-trained conditional generative adversarial network (cGAN) to address the unpaired data issue in radiogenomic analysis of BC. The generated images will then be used to predict the mutations status of key driver genes and BC subtypes.

METHODS

We integrated the paired MRI and multi-omic (mRNA gene expression, DNA methylation, and copy number variation) profiles of 61 BC patients from The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA). To facilitate this integration, we employed a Bayesian Tensor Factorization approach to factorize the multi-omic data into 17 latent features. Subsequently, a cGAN model was trained based on the matched side-view patient MRIs and their corresponding latent features to predict MRIs for BC patients who lack MRIs. Model performance was evaluated by calculating the distance between real and generated images using the Fréchet Inception Distance (FID) metric. BC subtype and mutation status of driver genes were obtained from the cBioPortal platform, where 3 genes were selected based on the number of mutated patients. A convolutional neural network (CNN) was constructed and trained using the generated MRIs for mutation status prediction. Receiver operating characteristic area under curve (ROC-AUC) and precision-recall area under curve (PR-AUC) were used to evaluate the performance of the CNN models for mutation status prediction. Precision, recall and F1 score were used to evaluate the performance of the CNN model in subtype classification.

RESULTS

The FID of the images from the well-trained cGAN model based on the test set is 1.31. The CNN for TP53, PIK3CA, and CDH1 mutation prediction yielded ROC-AUC values 0.9508, 0.7515, and 0.8136 and PR-AUC are 0.9009, 0.7184, and 0.5007, respectively for the three genes. Multi-class subtype prediction achieved precision, recall and F1 scores of 0.8444, 0.8435 and 0.8336 respectively. The source code and related data implemented the algorithms can be found in the project GitHub at https://github.com/mattthuang/BC_RadiogenomicGAN .

CONCLUSION

Our study establishes cGAN as a viable tool for generating synthetic BC MRIs for mutation status prediction and subtype classification to better characterize the heterogeneity of BC in patients. The synthetic images also have the potential to significantly augment existing MRI data and circumvent issues surrounding data sharing and patient privacy for future BC machine learning studies.

Changes in myocardial blood flow in a canine model of left sided breast cancer radiotherapy

BACKGROUND

Left-sided breast cancer patients receiving adjuvant radiotherapy are at risk for coronary artery disease, and/or radiation mediated effects on the microvasculature. Previously our laboratory demonstrated in canines with hybrid 18FDG/PET a progressive global inflammatory response during the initial one year following treatment. In this study, the objective is to evaluate corresponding changes in perfusion, in the same cohort, where resting myocardial blood flow (MBF) was quantitatively measured.

METHOD

In five canines, Ammonia PET (13NH3) derived MBF was measured at baseline, 1-week, 1, 3, 6 and 12-months after cardiac external beam irradiation. MBF measurements were correlated with concurrent 18FDG uptake. Simultaneously MBF was measured using the dual bolus MRI method.

RESULTS

MBF was significantly increased at all time points, in comparison to baseline, except at 3-months. This was seen globally throughout the entire myocardium independent of the coronary artery territories. MBF showed a modest significant correlation with 18FDG activity for the entire myocardium (r = 0.51, p = 0.005) including the LAD (r = 0.49, p = 0.008) and LCX (r = 0.47, p = 0.013) coronary artery territories.

CONCLUSION

In this canine model of radiotherapy for left-sided breast cancer, resting MBF increases as early as 1-week and persists for up to one year except at 3-months. This pattern is similar to that of 18FDG uptake. A possible interpretation is that the increase in resting MBF is a response to myocardial inflammation.

Involvement of the AKT Pathway in Resistance to Erlotinib and Cabozantinib in Triple-Negative Breast Cancer Cell Lines

Resistance to protein tyrosine kinase inhibitors (TKIs) presents a significant challenge in therapeutic target development for cancers such as triple-negative breast cancer (TNBC), where conventional therapies are ineffective at combatting systemic disease. Due to increased expression, the receptor tyrosine kinases EGFR (epidermal growth factor receptor) and c-Met are potential targets for treatment. However, targeted anti-EGFR and anti-c-Met therapies have faced mixed results in clinical trials due to acquired resistance. We hypothesize that adaptive responses in regulatory kinase networks within the EGFR and c-Met signaling axes contribute to the development of acquired erlotinib and cabozantinib resistance. To test this, we developed two separate models for cabozantinib and erlotinib resistance using the MDA-MB-231 and MDA-MB-468 cell lines, respectively. We observed that erlotinib- or cabozantinib-resistant cell lines demonstrate enhanced cell proliferation, migration, invasion, and activation of EGFR or c-Met downstream signaling (respectively). Using a SILAC (Stable Isotope Labeling of Amino acids in Cell Culture)-labeled quantitative mass spectrometry proteomics approach, we assessed the effects of erlotinib or cabozantinib resistance on the phosphoproteome, proteome, and kinome. Using this integrated proteomics approach, we identified several potential kinase mediators of cabozantinib resistance and confirmed the contribution of AKT1 to erlotinib resistance in TNBC-resistant cell lines.