Sr. Potter et al., Genetically engineered neural networks for predicting prostate cancer progression after radical prostatectomy, UROLOGY, 54(5), 1999, pp. 791-795
Objectives, To use pathologic, morphometric, DNA ploidy, and clinical data
to develop and test a genetically engineered neural network (GENN) for the
prediction of biochemical (prostate-specific antigen [PSA]) progression aft
er radical prostatectomy in a select group of men with clinically localized
prostate cancer.
Methods. Two hundred fourteen men who underwent anatomic radical retropubic
prostatectomy for clinically localized prostate cancer were selected on th
e basis of adequate follow-up, pathologic criteria indicating an intermedia
te risk of progression, and availability of archival tissue, The median age
was 58.9 years (range 40 to 87), Men with Gleason score 5 to 7 and clinica
l Stage T1b-T2c tumors were included. Follow-up was a median of 9.5 years.
Three GENNs were developed using pathologic findings (Gleason score, extrap
rostatic extension, surgical margin status), age, quantitative nuclear grad
e (QNG), and DNA ploidy, These networks were developed using three randomly
selected training (n = 136) and testing (n = 35) sets. Different variable
subsets were compared for the ability to maximize prediction of progression
. Both standard logistic regression and Cox regression analyses were used c
oncurrently to calculate progression risk.
Results. Biochemical (PSA) progression occurred in 84 men (40%), with a med
ian time to progression of 48 months (range 1 to 168), GENN models were tra
ined using inputs consisting of (a) pathologic features and patient age; (b
) QNG and DNA ploidy; and (c) all variables combined. These GENN models ach
ieved an average accuracy of 74.4%, 63.1%, and 73.5%, respectively, for the
prediction of progression in the training sets. In the testing sets, the t
hree GENN models had an accuracy of 74.3%, 80.0%, and 78.1%, respectively.
Conclusions. The GENN models developed show promise in predicting progressi
on in select groups of men after radical prostatectomy. Neural networks usi
ng QNG and DNA ploidy as input variables performed as well as networks usin
g Gleason score and staging information. All GENN models were superior to l
ogistic regression modeling and to Cox regression analysis in prediction of
PSA progression. The development of models using improved input variables
and imaging systems in larger, well-characterized patient groups with long-
term follow-up is ongoing. UROLOGY 54: 791-795, 1999. (C) 1999, Elsevier Sc
ience Inc.