Ania Gawlik's article featured in the Best of the 2016 AUA

Tuesday, November 15, 2016 - 11:43

Ania Gawlik's article was featured in a paper describing the best of AUA. The quoted text is as follows.

"In the 1970s there was great interest in prostate cancer cytology (based on needle aspiration) for prostate cancer diagnoses. It was supplanted after the pioneering work of Gleason demonstrated that the architecture of prostate cancer histology provided very significant prognostic information. History was reassessed in the report from Gawlik and coworkers,93 who utilized computer image analysis of Feulgen and hematoxylin-eosin (H&E) nuclear features to predict BCR in men following RP; 69 patients (20 BCR and 49 nonrecurrences) were assessed with mean BCR-free survival time of 6.6 years and follow-up to 14 years. A total of 242 quantitative histomorphometric (QH) features describing nuclear shape, architecture, and disorder were calculated from the H&E and Feulgen-stained tissue microarray core images in each patient. The top 10 ranked features for each stain type were selected.

Gleason score did not discriminate between those who did or did not have BCR predictions using QH features extracted from Feulgen and H&E images and revealed statistically different outcomes. Combining Gleason score, H&E, and Feulgen together showed the highest classification accuracy (0.75; P < .001). Although this is a very small study, if validated, this may provide a fruitful arena for development—perhaps there is something new (again) under the sun."

 

The full paper can be found at 

Rev Urol. 2016; 18(3): 159–173.
Highlights From the 2016 American Urological Association Annual Meeting, May 6–10, 2016, San Diego, CA

Citation of CCIPD's paper is 

Gawlik A, Lee G, Whitney J, et al. Computer extracted nuclear features from Feulgen and H&E images predict biochemical recurrence in prostate cancer patients following radical prostatectomy [AUA abstract MP02-17] J Urol. 2016;195(suppl 4):e16–e17.