Prediction of postoperative prostatic cancer stage on the basis of systematic biopsies using two types of artificial neural networks

Objective: The choice of therapy for prostatic cancer should depend on a ra tional preoperative estimate of tumor stage. Artificial neural networks wer e used to predict postoperative staging of prostatic cancer from sextant bi opsies and routinely available preoperative data. Methods: In group I (97 cases), nonorgan confinement (tumor stage greater t han or equal to pT3a) was predicted on the basis of age and six histopathol ogical variables from sextant biopsies. In group II (77 cases), nonorgan co nfinement and extraprostatic organ infiltration (tumor classification great er than or equal to pT3b) were predicted from age, four histopathological v ariables, the preoperative PSA level, and the total prostate volume estimat ed by preoperative ultrasonography. Learning vector quantization (LVQ) netw orks were applied for this purpose and compared to multilayer perceptrons ( MLP) and linear discriminant analysis (LDA). Results: Nonorgan confinement could be predicted correctly in 90% of newly presented cases from sextant biopsy histopathology alone. A similar accurac y of predicting nonorgan confinement (83%) was obtained by combining preope rative biopsy histology with clinical data. Extraprostatic organ infiltrati on could be predicted correctly in 82%. The best results were obtained by L VQ networks, followed by MLP networks and LDA. Conclusion: The postoperative tumor stage of prostatic cancer can be estima ted with high accuracy, sensitivity and specificity from preoperative routi ne parameters using artificial neural networks, especially LVQ networks. Th e results suggest that this methodology should be evaluated in a larger pro spective study. Copyright (C) 2001 S. Karger AG, Basel.