Through the eyes of Janus: Remission and disease activity in cluster headache

Cluster headache is among the most painful conditions known and affects approximately 0.12% of the population, being similar in prevalence to Parkinson's disease. However, the natural course and long-term prognosis remain poorly understood. As a result, a frequently asked question by patients in clinical practice – “When will my cluster bouts eventually stop over time?” – cannot yet be answered and therefore the consultation room momentarily becomes Janus’ temple of uncertainty.

In ancient Roman mythology, Janus – the God of beginnings, endings and transition – was depicted with two faces: one looking to the past and one to the future. Similarly, when addressing a patients’ prognosis, the physician takes on the role of the Janus: one face turned toward the patient's past, the other toward an uncertain future. Fortunately, the new prospective study by Lee et al. ¹ strengthens this forward gaze, offering clearer guidance on what lies ahead.

Predicting a prognosis in cluster headache remains a complex endeavor. While the classification in subtypes provides a framework, individual disease trajectories often vary. To offer patients presenting during their first cluster bout in the consultation room a meaningful answer, we must examine the three paths cluster headache can follow: (1) long-term remission, in which the patient remains free of further bouts; (2) chronification, where the bout continues for at least a year resulting in chronic cluster headache; or (3) bout recurrence, relapsing episodic cluster headache characterized by the recurrence of bouts after a period of remission. Lee et al. ¹ aimed to shed more light on the prognosis of cluster headache with their prospective study that comprehensively recorded the ongoing “index” bout of patients at baseline, before following them for five years. They especially focused on relapse rates (rate of episode recurrence) and chronification in patients with episodic cluster headache.

For some patients, Janus gazes hopefully into the future – those patients develop long-term remission. This path has been investigated in a few retrospective studies including a previous study by Lee et al.^2–5 More than one-quarter of patients did not have bout recurrence after their first-ever cluster headache bout. ⁶ In the other patients cluster headache can last for decades, making research on long-term remission challenging, as this requires a long-term observation (often more than years, ideally more than 15 years), which is less feasible for a prospective setting. Consequently, most investigations into the long-term disease course of cluster headache have been done retrospectively, by contacting patients who had visited clinics years to decades prior, enabling identification of long-term remission, but limited by recall bias.^2–6

For other patients, the journey leads to chronification. This is a path more frequently studied, yet still unpredictable. Like Janus at the threshold, chronification often feels like the crossing of an invisible line: an unclear unpredictable tipping point is reached, which often only becomes fully apparent in hindsight. Chronic cluster headache can develop directly at first-onset (“primary chronic cluster headache”) or result from subtype transition from episodic (“secondary chronic cluster headache”). ⁷ The chronic form is more prevalent in European and North-American populations (∼15%) compared to Asian populations (2.8–7.5%). ⁸ Transition to secondary chronic cluster headache has largely been studied retrospectively, with reported rates between 2.8% and 12.9%.^2,3,6 One European study found a one-year risk of secondary chronification of 4.0% and a five-year risk of 12.3%.^7,9 Lee et al. ¹ now provide prospective data on chronification in the Asian population, reporting five-year transition rates of 3.8% of first-onset cluster headache to primary and 1.4% to secondary chronic cluster headache.

The most common trajectory falls between remission and chronification: recurrence of cluster headache bouts. For these patients, Janus never fully turns away – glancing back at past bouts while anticipating the next. Remission may last for years, but recurrence is always a possibility. This third path has received limited attention, with earlier studies (based on retrospective self-reports) estimating bout frequency at one or two per year.^10,11 Lee et al. ¹ provide new prospective data showing that roughly one in three patients relapses each year (relapse rate of 0.29 in overall cluster headache). Notably, this is lower than previously reported, including Lee and colleagues' own analysis using the same patients, most likely reflecting recall bias in previous retrospective studies. ¹¹

Disease activity as expressed in bout relapse followed a dynamic arc, as observed in this prospective study. First-onset cluster headache patients faced an 11% risk, whereas this steeply increased to 33% in patients with recurrent bouts, suggesting that disease activity increases over time. Yet this escalation does not continue indefinitely. After reaching a peak, relapse rates declined with advancing age and longer disease duration – suggesting that disease activity decreases as the condition matures. Of note, the use of preventive medication should always carefully be taken into account when studying the natural disease course of cluster headache. Patients may be free of attacks for months or years while still needing to use preventive therapies such as verapamil. This implies that despite a seemingly beneficial clinical situation, the disease itself is not truly in remission.

As clinicians, we play Janus daily: listening to the past, studying the present and cautiously peering ahead. We are tasked not with prophecy. Alternatively, we prepare patients for their future by helping patients understand the spectrum of possible futures and equipping them to face each one. Based upon the findings of Lee et al., ¹ we can tell patients that disease activity of episodic cluster headache tends to gradually increase after onset and then regress over time. While we cannot promise which door will open, either remission, chronicity or recurrence, we can prepare our patients for each possibility.