Exploration of Prolonged Remission and the Natural Course of Cluster Headache: An Interview-Based Cohort Study

The aim of this study was to gain rare insight into prolonged cluster headache (CH) remission by (1) identifying patterns and factors associated with and (2) phenotypical changes before prolonged remission. The results can help patients better understand their disease course and uncover mechanisms behind spontaneous remission.

In this cross-sectional cohort study, all participants with a history of (probable) CH from the Leiden University Medical Center cohort were invited to complete a screening survey. Participants in prolonged remission were invited for a telephone interview. Prolonged remission was defined as (1) no current prophylactic treatment and (2) an attack-free period of ≥5 years and/or twice the mean between-episode time. Main outcomes are average age at prolonged remission onset and disease duration. Data were collected between April 10 and August 9, 2024, and analyzed using descriptive and survival statistics.

Of those invited, 43.2% (778/1,801) responded; 625 were included in the survey analysis, and 125 (20%) met prolonged remission criteria during interview. The median age at inclusion was 58 years (interquartile range [IQR] 48-67) with 32% female. Remission occurred on average at age 55 (IQR 48-63) after a disease duration of 23 (15-33) years. In 62% (N = 78), remission occurred abruptly. Of those with gradual remission (38%, N = 47), attack frequency (65%) and intensity (59%) decreased and between-episode intervals increased (52%) before remission. Probability of prolonged remission was higher in those with episodic CH (hazard ratio [HR] 6.60, 95% CI 3.55-12.31), who had quit smoking (HR 2.53, 95% CI 1.66-3.86), who had a higher attack intensity (HR 1.28, 95% CI 1.08-1.52), and who had a higher age at disease onset (HR 1.05, 95% CI 1.03-1.06).

This cohort offers rare insight into prolonged CH remission, typically starting around the mid-50s after 25 years of active disease. Prolonged remission is not tied to a single factor such as disease duration. Remission onset does not peak at a specific age, and disease duration varies widely between patients with remission. Remission probability is higher in the episodic form despite a longer disease duration compared with the chronic form. The association between quitting smoking and prolonged remission supports a causal link with smoking and disease activity. These preliminary retrospective results require confirmation in future studies.