Artificial neural network (ANN) models have a strong self-learning ability and can deal with complex biological information, but there is no ANN model for predicting the benefits of adjuvant chemotherapy in patients with gastric cancer (GC).
The clinicopathological data of patients who underwent radical resection of GC from January 2010 to September 2014 were analyzed retrospectively. Patients who underwent surgery combined with adjuvant chemotherapy were randomly divided into a training cohort (70%) and a validation cohort (30%). An ANN model (CT-benefit-ANN) was established, and its ability to predict the benefit of chemotherapy was evaluated by the C-index. The prognostic prediction and stratification ability of CT-benefit-ANN and the 8th AJCC staging system were compared by ROC curves and Kaplan-Meier curves.
In the training and validation cohort, CT-benefit-ANN both shows good prediction accuracy for adjuvant chemotherapy benefit. The ROC curve showed that the prediction accuracy of CT-benefit-ANN was better than that of the 8th AJCC staging system in all groups. The calibration plots showed that the predicted prognosis of CT-benefit-ANN was highly consistent with the actual value. The survival curves showed that CT-benefit-ANN could stratify prognosis well for all groups and performed significantly better than the 8th AJCC staging system.
The CT-benefit-ANN model developed in this study can accurately predict the benefits of adjuvant chemotherapy in patients with stage II/III GC. The benefit score based on CT-benefit-ANN can predict the long-term prognosis of patients with adjuvant chemotherapy and has good prognostic stratification ability.
The authors.
Scientific and Technological Innovation Joint Capital Projects of Fujian Province.
All authors have declared no conflicts of interest.