Xavier, Debora

Abstract Background Cyclin-dependent kinases (CDK) 4/6 inhibitors have significantly improved outcomes for patients with ER+/HER2− breast cancer. Nevertheless, they differ from each other in terms of chemical, biological, and pharmacological features, as well as toxicity profiles. We aim to determine whether QTc prolongation is caused by CDK4/6i in general or if it is associated with ribociclib only. Methods We systematically searched PubMed, Embase, and Cochrane Library for randomized controlled trials (RCTs) comparing the prevalence of QTc prolongation as an adverse event in HR+ breast cancer patients treated with CDK4/6i vs those without CDK4/6i. We pooled relative risk (RR) and mean difference (MD) with 95% confidence interval (CI) for the binary endpoint of QT prolongation. Results We included 14 RCTs comprising 16 196 patients, of whom 8576 underwent therapy with CDK4/6i. An increased risk of QTc prolongation was associated with the use of CDK4/6i (RR = 2.35, 95% CI = 1.67 to 3.29, P < .001; I2 = 44%). Subgroup analyses revealed a significant increase in the QTc interval for the ribociclib and palbociclib cohorts. The ribociclib subgroup showed a relative risk of 3.12 (95% CI = 2.09 to 4.65, P < .001; I2 = 12%), whereas the palbociclib subgroup had a relative risk of 1.51 (95% CI = 1.05 to 2.15, P = .025; I2 = 0%). Conclusion Palbociclib was associated with QTc prolongation; however, the relative risk for any grade QTc was quantitively twice with ribociclib. Furthermore, grade 3 QTc prolongations were observed exclusively with ribociclib. These results are important for guiding clinical decision-making and provide reassurance regarding the overall safety profile of this drug class.

Objective Deep learning (DL) has shown promising results for improving mammographic breast cancer diagnosis. However, the impact of artificial intelligence (AI) on the breast cancer screening process has not yet been fully elucidated in terms of potential workload reduction. We aim to assess if AI-based triaging of breast cancer screening mammograms could reduce the radiologist's workload with non-inferior sensitivity. Methods PubMed, EMBASE, Cochrane Central, and Web of Science databases were systematically searched for studies that evaluated AI algorithms on computer-aided triage of breast cancer screening mammograms. We extracted data from homogenous studies and performed a proportion meta-analysis with a random-effects model to examine the radiologist's workload reduction (proportion of low-risk mammograms that could be theoretically ruled out from human's assessment) and the software's sensitivity to breast cancer detection. Results Thirteen studies were selected for full review, and three studies that used the same commercially available DL algorithm were included in the meta-analysis. In the 156,852 examinations included, the threshold of 7 was identified as optimal. With these parameters, radiologist workload decreased by 68.3% (95%CI 0.655–0.711, I² = 98.76%, p < 0.001), while achieving a sensitivity of 93.1% (95%CI 0.882–0.979, I² = 83.86%, p = 0.002) and a specificity of 68.7% (95% CI 0.684–0.723, I² = 97.5%, p < 0.01). Conclusions The deployment of DL computer-aided triage of breast cancer screening mammograms reduces the radiology workload while maintaining high sensitivity. Although the implementation of AI remains complex and heterogeneous, it is a promising tool to optimize healthcare resources.