Background: Previous studies have shown that immune-mediated colitis is a common side effect in patients with cancer who are treated with ICIs. In this study, using claims data, we used 2 methods to examine the patterns of care after colitis diagnosis in ICI-treated patients with melanoma.
Objectives: (1) Use a Sankey chart to describe medication treatment patterns after colitis diagnosis among melanoma patients treated with ICIs; (2) Use the Kaplan-Meier (K-M) method to examine the occurrence of medication discontinuation/switching among these patients.
Methods: This retrospective cohort study used the IQVIA PharMetrics® Plus claims database. Adult patients with melanoma who received an ICI and had evidence of a cancer diagnosis within 5 days prior to their first ICI treatment (index date) from 01/2016 through 12/2020 were identified. Patients were required to have a year of continuous enrollment prior to the index date (baseline period) and no codes for colitis during that time. Colitis was defined as having at least one claim code for colitis (K50.xx, K51.xx, and K52.xx codes, except codes for polyps, radiation, and allergic/dietetic colitis) and at least one prescription for a colitis medication within one month after a claim for colitis. Prescriptions were noted post-colitis until the end of the patient’s enrollment. A Sankey chart was used to depict treatment sequences. The K-M method was used to describe discontinuation/switching of colitis medications and to calculate the mean and median (95% confidence interval) time to discontinuation/switching.
Results: After applying the selection criteria, 3,169 patients ICI-treated patients with melanoma were identified, of whom 294 (9.3%) experienced colitis after ICI treatment initiation. Of these 294 patients, 280 were initially treated with corticosteroids, 13 with gastrointestinal agents (e.g., infliximab, vedolizumab, mesalamine), and 1 with an anorectal. Of the 280 (95.2%) patients with colitis who were initially treated with corticosteroids, 255 (91.1%) discontinued corticosteroids and, among these, 65 (25.5%) switched to a different medication. Mean time to discontinuation/switching was 6.8 months; median time was 2.26 months.
Conclusions: This observational study using secondary data analysis demonstrated that the vast majority of post-ICI patients who develop colitis are treated initially with steroids. As the majority of these patients are able to discontinue all colitis medications within a few months, corticosteroids appear to be a rapid and effective treatment for this common side effect.
