Biochemical profiling and symptomatology of androgen deficiency in males with cluster headache: A prospective...

Cluster headache (CH) has historically been considered a male-dominated disorder, suggesting a role for androgens in its pathophysiology. We aimed to evaluate hormone profiles and symptoms of androgen deficiency in CH.

In this prospective, matched case-control study with a nested repeated-measures component, fasted morning blood samples were collected in 30 males with chronic CH (CCH), 33 males with episodic CH (ECH; both during episode and remission), and in 30 age-matched male headache-free controls (HCs). Surveys assessing symptoms of androgen deficiency (Quantitative Androgen Deficiency of Aging Males [QADAM], Aging Males' Symptoms [AMS]) were collected simultaneously. Steroid hormone profiles were compared among groups. The primary outcome was total testosterone (tT) including biochemical hypogonadism (tT ≤ 10.5). Secondary outcomes were hormonal pathways and individual hormones. Data were collected between October 2022 and April 2025.

Biochemical hypogonadism was observed in 20% of participants with CCH (n = 6), in 9% with ECH (n = 3), and in 7% of HCs (n = 2). None of the HCs had severe hypogonadism, in contrast to those with CH (CCH, n = 3/6; ECH, n = 1/3). AMS and QADAM scores indicated more symptoms of androgen deficiency in participants with hypogonadism, in ECH (AMS: B = 9.59 [95% CI = 5.69-13.50], p < 0.001; QADAM: B = -5.25 [95% CI = -7.39 to -3.11], p < 0.001) and CCH (AMS: B = 8.78 [95% CI = 4.45-13.12], p < 0.001; QADAM: B = -4.59 [95% CI = -6.97 to -2.22], p < 0.001) compared to HCs. In the unadjusted model, we observed a lower, albeit non-statistically significant, mean tT in participants with CCH versus HCs (CCH: B = -2.35 nmol/L [95% CI = -4.79 to 0.09], p = 0.059). In the adjusted model, the differences attenuated, suggesting a strong effect of body mass index and age on the mean tT (CCH: B = -0.93 nmol/L [95% CI = -3.81 to 1.95], p = 0.524; ECH: B = 0.55 nmol/L [95% CI = -2.05 to 3.15], p = 0.674). Measures of hormonal pathways and individual hormones did not differ among groups.

Biochemical hypogonadism was observed in one out of five males with CCH versus one out of 20 in the ECH and HCs. Apart from one case, gonadotrophin concentrations were not increased, suggesting a central origin. In the crude data we observed a lower, albeit non-statistically significant, mean tT in males with CCH, but this appeared to be mainly driven by body mass index and age, because it attenuated after correction. Clinical symptoms of androgen deficiency were more prevalent in CH compared to HCs, independent of hypogonadism. Physicians should be aware of potential increased risk and symptoms of (central) hypogonadism in CCH. Further investigation should explore shared underlying mechanisms and testosterone supplementation in CH.