ABSTRACT
Several fusion genes are directly involved in the initiation and progression of cancers. Numerous bioinformatics tools have been developed to detect fusion events, but they are mainly based on RNA-seq data. The whole-exome sequencing (WES) represents a powerful technology that is widely used for disease-related DNA variant detection. In this study, we build a novel analysis pipeline called Fuseq-WES to detect fusion genes at DNA level based on the WES data. The same method applies also for targeted panel sequencing data. We assess the method to real datasets of acute myeloid leukemia (AML) and prostate cancer patients. The result shows that two of the main AML fusion genes discovered in RNA-seq data, PML-RARA and CBFB-MYH11, are detected in the WES data in 36 and 63% of the available samples, respectively. For the targeted deep-sequencing of prostate cancer patients, detection of the TMPRSS2-ERG fusion, which is the most frequent chimeric alteration in prostate cancer, is 91% concordant with a manually curated procedure based on four other methods. In summary, the overall results indicate that it is challenging to detect fusion genes in WES data with a standard coverage of â¼ 15-30x, where fusion candidates discovered in the RNA-seq data are often not detected in the WES data and vice versa. A subsampling study of the prostate data suggests that a coverage of at least 75x is necessary to achieve high accuracy.
ABSTRACT
BACKGROUND: A new generation of risk calculators (RCs) for prostate cancer (PCa) incorporating magnetic resonance imaging (MRI) data have been introduced. However, these have not been validated externally, and their clinical benefit compared with alternative approaches remains unclear. OBJECTIVE: To assess previously published PCa RCs incorporating MRI data, and compare their performance with traditional RCs (European Randomized Study of Screening for Prostate Cancer [ERSPC] 3/4 and Prostate Biopsy Collaborative Group [PBCG]) and the blood-based Stockholm3 test. DESIGN, SETTING, AND PARTICIPANTS: RCs were tested in a prospective multicenter cohort including 532 men aged 45-74 yr participating in the Stockholm3-MRI study between 2016 and 2017. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: The probabilities of detection of clinically significant PCa (csPCa) defined as Gleason score ≥3 + 4 were calculated for each patient. For each RC and the Stockholm3 test, discrimination was assessed by area under the curve (AUC), calibration by numerical and graphical summaries, and clinical usefulness by decision curve analysis (DCA). RESULTS AND LIMITATIONS: The discriminative ability of MRI RCs 1-4 for the detection of csPCa was superior (AUC 0.81-0.87) to the traditional RCs (AUC 0.76-0.80). The observed prevalence of csPCa in the cohort was 37%, but calibration-in-the-large predictions varied from 14% to 63% across models. DCA identified only one model including MRI data as clinically useful at a threshold probability of 10%. The Stockholm3 test achieved equivalent performance for discrimination (AUC 0.86) and DCA, but was underpredicting the actual risk. CONCLUSIONS: Although MRI RCs discriminated csPCa better than traditional RCs, their predicted probabilities were variable in accuracy, and DCA identified only one model as clinically useful. PATIENT SUMMARY: Novel risk calculators (RCs) incorporating imaging improved the ability to discriminate clinically significant prostate cancer compared with traditional tools. However, all but one predicted divergent compared with actual risks, suggesting that regional modifications be implemented before usage. The Stockholm3 test achieved performance comparable with the best MRI RC without utilization of imaging.
