Non muscle invasive bladder cancers recur frequently and identification of biomarkers for predicting recurrence are necessary. The present study evaluated the individual and synergistic effects of tumor suppressor (p53/p21waf1) and angiogenesis [vascular endothelial growth factor (VEGF)/endoglin (CD105)] markers. The study included 90 cases of non muscle invasive bladder cancer. Cell spots were stained with primary antibodies and Flourescein isothiocyanate (FITC). Slides were observed under confocal laser scanning microscope for protein expression. The association between the markers individually and synergistically with recurrence were assessed by a χ2 and Fisher's Exact test. Survival analysis was performed to predict recurrence and test for significant difference in recurrence free survival probability. Recurrence [overall:39(43.3%) and low grade(LG):26(54.2%)] was significant with p53 and VEGF expression and the profiles p53/VEGF, p53/CD105, VEGF/CD105, p53/p21/CD105, p53/VEGF/CD105 and all four were significantly associated with recurrence in both groups. In the multivariable model the [HR(95%CI),p: overall and LG] profiles p21/VEGF [2.195(1.052-4.582),0.036; 3.425(1.332-8.811),0.011], VEGF/CD105[2.624(1.274-5.403),0.009 and 3.380(1.348-8.472),0.009], p53/p21/CD105 [2.000(0.993-4.027),0.052 and 2.539(1.047-6.157),0.039], p53/VEGF/CD105 [2.360(1.148-4.849),0.020 and 2.738(1.104-6.788),0.030], p21/VEGF/CD105 [2.611(1.189-5.731),0.017 and 3.946(1.530-10.182),0.005] and all four [2.382(1.021-5.556),0.045 and 3.572(1.287-9.911),0.014] significantly predicted the recurrence along with significant log rank. In the pTa subset (n = 33) the profiles p53/p21, p53/CD105, p21/VEGF, VEGF/CD105, p53/VEGF/CD105, p53/p21/CD105 and p21/VEGF/CD105, significantly predicted hazard for recurrence. The present study emphasizes an underlying association between tumor suppressor (p21waf1) and angiogenesis (VEGF/CD105) biomarkers. In addition combination profiles appeared to indicate an aggressive nature with high propensity for recurrence in LG and pTa tumours.

Objectives This study aimed to develop and validate a deep learning algorithm to screen digitized acid fast–stained (AFS) slides for mycobacteria within tissue sections. Methods A total of 441 whole-slide images (WSIs) of AFS tissue material were used to develop a deep learning algorithm. Regions of interest with possible acid-fast bacilli (AFBs) were displayed in a web-based gallery format alongside corresponding WSIs for pathologist review. Artificial intelligence (AI)–assisted analysis of another 138 AFS slides was compared to manual light microscopy and WSI evaluation without AI support. Results Algorithm performance showed an area under the curve of 0.960 at the image patch level. More AI-assisted reviews identified AFBs than manual microscopy or WSI examination (P < .001). Sensitivity, negative predictive value, and accuracy were highest for AI-assisted reviews. AI-assisted reviews also had the highest rate of matching the original sign-out diagnosis, were less time-consuming, and were much easier for pathologists to perform (P < .001). Conclusions This study reports the successful development and clinical validation of an AI-based digital pathology system to screen for AFBs in anatomic pathology material. AI assistance proved to be more sensitive and accurate, took pathologists less time to screen cases, and was easier to use than either manual microscopy or viewing WSIs.