NEURAL NETWORKS AS A PROGNOSTIC TOOL FOR PATIENTS WITH NONSMALL CELL-CARCINOMA OF THE LUNG

Patients with non-small cell carcinoma of the lung (NSCLC) have a poor prognosis (64 and 41% survival rates in Stages I and II), It is curre ntly not possible to predict which patients with Stage I or II NSCLC w ill survive the disease, Sixty-seven patients with NSCLC, including 49 patients with Stage I NSCLC and 18 with Stage II disease (11 with squ amous cell carcinomas, 35 with adenocarcinomas, and 21 with large cell carcinomas) were treated with lobectomy and followed for a minimum of 5 years, The tumors were studied with DNA now cytometry and quantitat ive immunocytochemical studies for proliferation cell nuclear antigen, p53 protein, and MIB-1, The data were analyzed with backpropagation n eural networks, univariate analysis of variance, the Kaplan-Meier surv ival method, and Cox proportional hazards model, The dependent variabl es were ''free of disease'' and ''recurrence or dead from disease.'' T wenty neural network models were trained, using all cases but one, aft er 1883 to 2000 training cycles, At 5 years, 30 patients were free of disease and 37 were dead or had recurrence. Proliferating cell nuclear antigen was the only statistically significant prognostic factor by u nivariate analysis of variance and Cox proportional hazards analysis. The S phase was statistically significant by univariate analysis of va riance (P <.05). All of the 20 models classified the test cases correc tly, Study with backpropagation neural networks using multiple prognos tic features from patients with NSCLC suggests that this technology mi ght be useful for prediction of survival, This preliminary study must be validated with data from a larger group of patients with NSCLC befo re its clinical adequacy is established.