DrABC: deep learning accurately predicts germline pathogenic mutation status in breast cancer patients based on phenotype data

Development and validation of ultrasound-based radiomics model to predict germline BRCA mutations in patients with breast cancer

BACKGROUND

Identifying breast cancer (BC) patients with germline breast cancer susceptibility gene (gBRCA) mutation is important. The current criteria for germline testing for BC remain controversial. This study aimed to develop a nomogram incorporating ultrasound radiomic features and clinicopathological factors to predict gBRCA mutations in patients with BC.

MATERIALS AND METHODS

In this retrospective study, 497 women with BC who underwent gBRCA genetic testing from March 2013 to May 2022 were included, including 348 for training (84 with and 264 without a gBRCA mutation) and 149 for validation(36 patients with and 113 without a gBRCA mutation). Factors associated with gBRCA mutations were identified to establish a clinicopathological model. Radiomics features were extracted from the intratumoral and peritumoral regions (3 mm and 5 mm) of each image. The least absolute shrinkage and selection operator regression algorithm was used to select the features and logistic regression analysis was used to construct three imaging models. Finally, a nomogram that combined clinicopathological and radiomics features was developed. The models were evaluated based on the area under the receiver operating characteristic curve (AUC), calibration, and clinical usefulness.

RESULTS

Age at diagnosis, family history of BC, personal history of other BRCA-related cancers, and human epidermal growth factor receptor 2 status were independent predictors of the clinicopathological model. The AUC of the imaging radiomics model combining intratumoral and peritumoral 3 mm areas in the validation set was 0.783 (95% confidence interval [CI]: 0.702-0.862), which showed the best performance among three imaging models. The nomogram yielded better performance than the clinicopathological model in validation sets (AUC: 0.824 [0.755-0.894] versus 0.659 [0.563-0.755], p = 0.007).

CONCLUSION

The nomogram based on ultrasound images and clinicopathological factors performs well in predicting gBRCA mutations in BC patients and may help to improve clinical decisions about genetic testing.

A review of genetic variant databases and machine learning tools for predicting the pathogenicity of breast cancer

Studies continue to uncover contributing risk factors for breast cancer (BC) development including genetic variants. Advances in machine learning and big data generated from genetic sequencing can now be used for predicting BC pathogenicity. However, it is unclear which tool developed for pathogenicity prediction is most suited for predicting the impact and pathogenicity of variant effects. A significant challenge is to determine the most suitable data source for each tool since different tools can yield different prediction results with different data inputs. To this end, this work reviews genetic variant databases and tools used specifically for the prediction of BC pathogenicity. We provide a description of existing genetic variants databases and, where appropriate, the diseases for which they have been established. Through example, we illustrate how they can be used for prediction of BC pathogenicity and discuss their associated advantages and disadvantages. We conclude that the tools that are specialized by training on multiple diverse datasets from different databases for the same disease have enhanced accuracy and specificity and are thereby more helpful to the clinicians in predicting and diagnosing BC as early as possible.

Validation of the BOADICEA model for predicting the likelihood of carrying pathogenic variants in eight breast and ovarian cancer susceptibility genes

BOADICEA is a comprehensive risk prediction model for breast and/or ovarian cancer (BC/OC) and for carrying pathogenic variants (PVs) in cancer susceptibility genes. In addition to BRCA1 and BRCA2, BOADICEA version 6 includes PALB2, CHEK2, ATM, BARD1, RAD51C and RAD51D. To validate its predictions for these genes, we conducted a retrospective study including 2033 individuals counselled at clinical genetics departments in Denmark. All counselees underwent comprehensive genetic testing by next generation sequencing on suspicion of hereditary susceptibility to BC/OC. Likelihoods of PVs were predicted from information about diagnosis, family history and tumour pathology. Calibration was examined using the observed-to-expected ratio (O/E) and discrimination using the area under the receiver operating characteristics curve (AUC). The O/E was 1.11 (95% CI 0.97-1.26) for all genes combined. At sub-categories of predicted likelihood, the model performed well with limited misestimation at the extremes of predicted likelihood. Discrimination was acceptable with an AUC of 0.70 (95% CI 0.66-0.74), although discrimination was better for BRCA1 and BRCA2 than for the other genes in the model. This suggests that BOADICEA remains a valid decision-making aid for determining which individuals to offer comprehensive genetic testing for hereditary susceptibility to BC/OC despite suboptimal calibration for individual genes in this population.

Clinical application of artificial neural network (ANN) modeling to predict BRCA1/2 germline deleterious variants in Chinese bilateral primary breast cancer patients

BACKGROUND

Bilateral breast cancer (BBC), as well as ovarian cancer, are significantly associated with germline deleterious variants in BRCA1/2, while BRCA1/2 germline deleterious variants carriers can exquisitely benefit from poly (ADP-ribose) polymerase (PARP) inhibitors. However, formal genetic testing could not be carried out for all patients due to extensive use of healthcare resources, which in turn results in high medical costs. To date, existing BRCA1/2 deleterious variants prediction models have been developed in women of European or other descent who are quite genetically different from Asian population. Therefore, there is an urgent clinical need for tools to predict the frequency of BRCA1/2 deleterious variants in Asian BBC patients balancing the increased demand for and cost of cancer genetics services.

METHODS

The entire coding region of BRCA1/2 was screened for the presence of germline deleterious variants by the next generation sequencing in 123 Chinese BBC patients. Chi-square test, univariate and multivariate logistic regression were used to assess the relationship between BRCA1/2 germline deleterious variants and clinicopathological characteristics. The R software was utilized to develop artificial neural network (ANN) and nomogram modeling for BRCA1/2 germline deleterious variants prediction.

RESULTS

Among 123 BBC patients, we identified a total of 20 deleterious variants in BRCA1 (8; 6.5%) and BRCA2 (12; 9.8%). c.5485del in BRCA1 is novel frameshift deleterious variant. Deleterious variants carriers were younger at first diagnosis (P = 0.0003), with longer interval between two tumors (P = 0.015), at least one medullary carcinoma (P = 0.001), and more likely to be hormone receptor negative (P = 0.006) and HER2 negative (P = 0.001). Area under the receiver operating characteristic curve was 0.903 in ANN and 0.828 in nomogram modeling individually (P = 0.02).

CONCLUSION

This study shows the spectrum of the BRCA1/2 germline deleterious variants in Chinese BBC patients and indicates that the ANN can accurately predict BRCA deleterious variants than conventional statistical linear approach, which confirms the BRCA1/2 deleterious variants carriers at the lowest costs without adding any additional examinations.