A nomogram for predicting severe myelosuppression in small cell lung cancer patients following the first-line chemotherapy

Abstract

AbstractThis study aimed at establishing and validating a nomogram to predict the probability of severe myelosuppression in small cell lung cancer (SCLC) patients following the first-line chemotherapy. A total of 179 SCLC cases were screened as the training group and another 124 patients were used for the validation group. Predictors were determined by the smallest Akaike’s information criterion (AIC) in multivariate logistic regression analysis, leading to a new nomogram. The nomogram was validated in both training and validation groups and the predicting value was evaluated by area under the receiver operating characteristics (ROC) curve (AUC), calibration curve, and decision curve analysis (DCA). Age and tumor staging were extracted as predictors to establish a nomogram, which displayed the AUC values as 0.725 and 0.727 in the training and validation groups, respectively. This nomogram exhibited acceptable calibration curves in the two groups and its prediction added more net benefits than the treat-all scheme and treat-none scheme if the range of threshold probability in the DCA was between 15 and 60% in the training and validation groups. Therefore, the nomogram objectively and accurately predict the occurrence of severe myelosuppression in SCLC patients following the first-line chemotherapy.