Figure 3 LASSO regression plot. (A) Plot of partial likelihood deviance of CSS, (B) plot of LASSO coefficient profiles of CSS, Each curve represents the LASSO coefficient profile of a feature against the log (lambda) sequence. when the optimal lambda value was 0.02, retentiong variables were screen; (C) Plot of partial likelihood deviance of OS, (D) plot of LASSO coefficient profiles of OS, Each curve represents the LASSO coefficient profile of a feature against the log (lambda) sequence, when the optimal lambda value was 0.04, retentiong variables were screen.
Based on multivariate analysis results and LASSO regression analysis, eight prognostic factors including marital status, grade, T stage, N stage, CEA, tumor size, surgery and chemotherapy were used to establish the nomogram of CSS (Figure 4A), And three prognostic factors including grade, chemotherapy and months from diagnosis to treatment were used to establish the nomogram of OS (Figure 4B), These nomograms were adopted to predict the 1-, 3-, and 5-year survival rates of training cohort. In a nomogram, the score of each value level of each variable was assigned according to its contribution to the outcome; then, each score was added to obtain the total score; finally, the predicted value of each individual outcome event was calculated through the function conversion relation between the total score and the survival probability.