DEVELOPMENT AND VALIDATION OF A NOMOGRAM TO INDIVIDUALLY PREDICT SURVIVAL OF OLD ATHLETIC PATIENTS WITH COLON ADENOCARCINOMA AFTER SURGERY

Abstract

Objective: Nomograms can be used in quantitative and intuitive methods to predict the survival rate of athletic patients. This study was designed to identify prognosis markers that affected overall survival (OS) in old colon adenocarcinoma (CAC) athletic patients after surgery and constructed a nomogram to predict postoperative survival.

Methods:  We selected qualified athletic patients diagnosed with CAC from 2010 to 2015 from the surveillance, epidemiology, and end results (SEER) database. We divided athletic patients diagnosed from 2010 to 2013 into a validation and training cohorts according to 3: 7. Athletic Patients from 2014 to 2015 were regarded as a test cohort. We used the Cox regression model to construct a new nomogram, and test the nomogram performance through validation and test cohorts.

Results:  The multivariate Cox regression analysis identified age, sex, brain metastasis, lymph node metastasis, carcinoembryonic antigen (CEA), liver metastasis, surgery style, tumor lymph node ratio (LNR), distant metastasis, and lung metastasis as independent prognosis markers associated with OS. The constructed nomogram had good calibration and discriminative ability. The concordance indexes (c-indexes) of validation and test cohorts were 0.767, 0.775.

Conclusions:  We constructed and validated a convenient and effective nomogram, which individually predicted OS for old athletic patients with CAC after surgery the based on clinical information available.