Treatment Patterns Among Metastatic Esophageal Cancer Patients in the US: A Retrospective Analysis of Health Insurance Claims Data

Abstract

Background: Esophageal cancer is a common cancer with over 19,000 new cases in the United States estimated for the year 2021. The National Comprehensive Cancer Network guidelines recommend ablation or surgery for patients that have early stages of the cancer. For unresectable or metastatic esophageal cancer, a variety of systemic treatments are recommended, including chemotherapies, monoclonal antibodies, and targeted therapies. Several recent treatment pattern studies have been published but none are specific to esophageal cancer and without restrictive exclusion criteria. Objective: Characterize treatment patterns among patients diagnosed with metastatic esophageal cancer during the years 2017, 2018, and 2019. Analyze time to change in treatment for the first line of therapy and probability of treatment initiation by year. Methods: The source data was compiled from the IBM MarketScan Commercial databases. Patients were included in the analysis if they had at least 2 outpatient diagnoses or 1 inpatient diagnosis for metastatic esophageal cancer. Patients were required to have continuous enrollment for 6 months prior to the index date and for 1 month after the index date. The index date was the second outpatient diagnosis or the first inpatient diagnosis of esophageal cancer. Analysis of treatment patterns and lines of therapy was performed by combining claims for cancer treatments for each patient within a specified timeframe and determining if the treatment was contained in the NCCN guidelines. Kaplan-Meier survival methods were used to determine treatment persistence, and a logistic regression was used to analyze the odds of treatment initiation by year. Results: In total, 190 patients were identified who matched eligibility criteria. The treatment pattern analysis resulted in 52 different treatment combinations being identified across 4 lines of therapy, and 120 patients received at least a one line of therapy. The most common treatments were a combination of fluorouracil and oxaliplatin and a combination of paclitaxel and carboplatin. First line treatment persistence based on the Kaplan-Meier analysis showed that the median duration of treatment for patients who received 2 or more lines of therapy was 106 days, and the median for patients who received only one line therapy was 49 days. The results of the logistic regression for treatment initiation by year showed no significant association between year of initiation and odds of starting treatment. Conclusion: The results of the treatment pattern analysis clearly show that a combination of fluorouracil and oxaliplatin is the most common first line treatment in metastatic esophageal cancer. The treatment persistence curve demonstrated a considerably longer duration of median first line therapy for patients who received additional lines of therapy beyond the first. The mean for all patients based on first line of therapy duration was greater than the median suggesting a left skew with a few patients having long first line of therapy times. Despite changes in esophageal cancer guidelines and the inclusion of new therapies over time, the logistic regression showed no significant findings; indicating that recommendations and availability of medications had limited effect on the odds of treatment initiation.