J Clin Oncol: Clinical calculators based on molecular and clinical pathological characteristics predict the risk of recurrence after stage I-III colon cancer

Clinical calculators and nomograms have been endorsed by the United Cancer Council (AJCC) for providing patients with the most personalized and accurate prognostic estimates.
using molecular and clinical pathological variables, Weiser et al. established a third-generation clinical calculator to predict the risk of recurrence after stage I-III colon cancer removal.
researchers looked forward to collecting data on 1,095 patients who undersecision at Memorial Sloan-Kettering Cancer Center between 2007 and 2014 to develop clinical calculators.
use consistency indices to measure discernment and calibration curves to assess the variability of individual predictions.
, the clinical calculator was externally validated in data from a group of patients at Washington University's Sittman Cancer Center in St. Louis.
The clinical calculator consists of six variables: microsatellite genomic esology, AJCC T classification, number of tumor-tired lymph nodes, presence of high-risk pathological characteristics (e.g. intravenous, lymphatic tube or nerve immersion), presence of tumor-incussive lymphocytes, and use of complementary chemotherapy.
the clinical calculator's consistency index is 0.792 (95% CI, 0.749 to 0.837), while the 5th and 8th editions of the AJCC Manual phased program consistency index Only 0.708 (95% CI, 0.671 to 0.745) and 0.757 (0.715 to 0.799), respectively.
, external validation confirmed this robust performance, with a consistency index of 0.738 (95% CI, 0.703-0.811) and a calibration graph of predicted probability and observed events approaching a 45-degree verse.
In summary, the third generation clinical calculator that predicts tumor recurrence after genomic colonectomy successfully combines the presence of microsatellite genomic esoteric and tumor-immersive lymphocytes, thus improving resolution and predictive accuracy.
same time, the study is an example of the development of clinical calculators, which, as widely validated and accepted in the oncology community, can maintain relevance and improve predictive performance by constantly incorporating emerging variables.