Development and Validation of Multicenter Predictive Nomograms for Locally Advanced Pancreatic Cancer after Chemoradiotherapy

Abstract

Background: Due to common practice of hypofractionated radiotherapy in pancreatic cancer and heterogeneous chemotherapy regimens in previous studies, modified nomograms are required. Therefore, we aim to develop and validate prognostic nomograms for locally advanced pancreatic cancer (LAPC) after stereotactic body radiation therapy (SBRT) and chemotherapy.Methods: The development cohort comprised 925 patients with LAPC receiving SBRT and gemcitabine-based chemotherapy in our center, while 297 patients from another two centers formed the validation cohort. Nomograms were created from COX models and internally validated by bootstrap. Model discriminations were evaluated by calibration plots and concordance index (C-index). A decision curve analysis (DCA) was performed to evaluate clinical benefits of nomograms. Additionally, recursive partitioning analysis (RPA) was used for stratifications of survival probability based on the total score of each patient calculated by nomograms.Results: Weight loss, tumor diameter, radiation dose, CA19-9 kinetics after treatment and surgical resection were included in the nomogram for overall survival (OS), while the five factors plus performance status formed the nomogram for progression free survival (PFS). The corrected C-indexes for estimated 1-year and 2-year OS of the development cohort were 0.88 (95% CI: 0.85-0.91) and 0.86 (95% CI: 0.83-0.90). For those of the validation cohort, it was 0.88 (95% CI: 0.82-0.94) and 0.83 (95% CI: 0.74-0.91). Additionally, the corrected C-index for predicted 1-year PFS in the development and validation cohort was 0.83 (95% CI: 0.81-0.86) and 0.82 (95% CI: 0.78-0.87), respectively. The calibration plots showed good agreement of 1- and 2-year OS and 1-year PFS between the estimations and actual observations. Potential clinical benefits were demonstrated with DCA. Additionally, for 1- and 2-year OS and 1-year PFS, patients were stratified into four groups with different survival probability by RPA.Conclusion: The validated nomograms provided useful predictions of OS and PFS for LAPC with chemoradiotherapy.

Construction of a cuproptosis‐related lncRNA signature for predicting prognosis and immune landscape in osteosarcoma patients

“…It is challenging to put a single indicator into clinical practice, even though the risk score shows the potential to be regarded as a promising predictor of clinical outcomes in osteosarcoma patients 55 . Predictions made using the nomogram are more accurate and useful since they can take into account the relative importance of each variable 56,57 . Therefore, we integrated the risk score and other clinicopathological parameters to develop a nomogram.…”

“…55 Predictions made using the nomogram are more accurate and useful since they can take into account the relative importance of each variable. 56,57 Therefore, we integrated the risk score and other clinicopathological parameters to develop a nomogram. The results indicated that the AUC values of ROC curves were significantly higher compared to the single-factor values (age, risk score, gender, and metastasis).…”