Prediction tools evaluated in large retrospective cohort study

Using a Northern California Kaiser Permanente cohort linked with local cancer registries to determine incidence of EAC, Dr. Rubenstein and colleagues evaluated the accuracy of four published prediction tools for EAC. The Kunzmann algorithm was found to be most effective (AUC = 0.73) and substantially better than using GERD symptoms alone to identify high-risk persons. Some of the limitations include the relatively small number of EAC outcomes (168), the need to impute multiple variables and records, and the lack of information on changes to participants' risk profiles (e.g., obesity, GERD symptoms, etc.) over the very long follow up time. Many of these limitations are conservative in nature, i.e., resulting in an underestimate of predictive ability. Also lacking, due to the source of the data, is information on other potential risk and preventive factors which have been shown to predict subsequent EAC. For example, a risk calculator (IC-RISC) that also includes race, family history and use of NSAIDs exhibits an AUC of 0.81 (although it has not been independently validated.) Nevertheless, the authors make the important point that any of the tested tools are vastly superior to using only GERD in defining "high-risk" persons for screening, and should be implemented in clinical practice.

Am J Gastroenterol. 2021 May 1;116(5):949-957. doi: 10.14309/ajg.0000000000001255.

Validation of Tools for Predicting Incident Adenocarcinoma of the Esophagus or Esophagogastric Junction

Joel H Rubenstein 1 2, Trivellore Raghunathan 3 4, Cecilia Doan 5, Jennifer Schneider 5, Wei Zhao 5, Valbona Metko 2, Kimberly Nofz 1, Maryam Khodadost 1, Douglas A Corley 5

PMID: 33852454 DOI: 10.14309/ajg.0000000000001255


Introduction: Guidelines suggest screening of individuals who are at increased risk of esophageal adenocarcinoma (EAC). Tools for identifying patients at risk of Barrett's esophagus have been validated. Here, we aimed to compare and validate the tools for the primary outcomes of interest: EAC and esophagogastric junction adenocarcinoma (EGJAC).

Methods: Retrospective longitudinal analysis of the Kaiser Permanente Northern California Multiphasic Health Checkup Cohort, a community-based cohort including 206,974 patients enrolled between 1964 and 1973 followed through 2016. Baseline questionnaires and anthropometrics classified predictor variables for each tool and were linked to cancer registry outcomes. Analyses used logistic regression, Cox proportional hazards regression, and Kaplan-Meier survival curves.

Results: We identified 168 incident EAC cases and 151 EGJAC cases at a mean of 32 years after enrollment (mean follow-up among controls 26 years). Gastroesophageal reflux disease (GERD) symptoms predicted incident EAC (hazard ratio 2.66; 95% confidence interval 1.01, 7.00), but not EGJAC. The Nord-Trøndelag Health Study tool, Kunzmann tool, and Michigan Barrett's Esophagus pREdiction Tool were more accurate than GERD for predicting EAC, with individuals in the fourth quartile of Kunzmann having 17-fold the risk of those in the 1st quartile (hazard ratio = 16.7, 95% confidence interval = 4.72, 58.8). Each tool also predicted incident EGJAC with smaller magnitudes of effect.

Discussion: The study independently validated 4 tools for predicting incident EAC and EGJAC in a large community-based population. The Kunzmann tool appears best calibrated; all appear preferable to using GERD alone for risk stratification. Future studies should determine how best to implement such tools into clinical practice.

Copyright © 2021 by The American College of Gastroenterology.

This website contains a curated and opinionated look at recent literature regarding the epidemiology and prevention of esophageal cancer, with an emphasis on esophageal adenocarcinoma. It is developed by Thomas L Vaughan MD, MPH ©2021
This website should not be considered, or used as a substitute for, medical advice, diagnosis or treatment. This site does not constitute the practice of any medical or other professional health care advice, diagnosis or treatment. The information on this website represent the views solely of Dr. Vaughan.
Cookies are used solely to quantify website use for resource planning. Complete IP addresses are not collected (i.e., will be recorded as 203.0.113.??) Feel free to opt out of cookies using add-ins of your choice without affecting website function. No user information is collected by the risk calculator (IC-RISC).