(A-324) DARWIN EU® - A multi-national network cohort study of the effect of doxycycline versus active comparators on the risk of suicidality in individuals with acne

Background: There have been several case reports of a potential association between use of doxycycline and suicide, as well as other psychiatric symptoms in observational studies.

Objectives: The aim was to compare the risk of composite suicide-related events, anxiety and depression after doxycycline use compared to active comparators in individuals with acne.

Methods: This active comparator cohort study used primary care data from the UK (CPRD GOLD), the Netherlands (IPCI), Spain (SIDIAP, with hospital discharge linkage) over the period 2010-2023. Individuals were included based on new use of doxycycline, erythromycin, or isotretinoin, with a condition record of acne within 180 days prior to the first date of treatment and with 365 days of prior observation. Individuals prescribed doxycycline were matched to those prescribed the comparator drugs by propensity-score and followed up until occurrence of the outcome, death, data end-date, or end of the treatment episode plus seven-days. Cox-proportional hazards regression was used to estimate hazard ratios for the association between doxycycline and composite suicide-related events (ideation, attempt, completion, self-harm), anxiety, and depression against the comparator drugs.

Results: Across the data sources, we included 31,097 individuals prescribed doxycycline, 48,473 prescribed erythromycin, and 13,948 prescribed isotretinoin. In SIDIAP, an increased risk of suicide-related events was observed in individuals prescribed doxycycline compared to erythromycin (HR 3.77, 95% CI [1·03-17·80]), and in CPRD GOLD, there was a non-statistically significant association (HR 1.71, 95% CI [0.74-4.07]). Meta-analysed together without IPCI, which remained blinded, the association remained (HR 2.11, (95% CI [1.01-4.39]). No association for suicide-related events was observed when using isotretinoin as comparator drug. In IPCI, an association with anxiety in doxycycline users was observed compared to isotretinoin users (HR 1.94, 95%CI [1.28-2.94]). No associations were found between doxycycline use and anxiety compared to erythromycin in CPRD GOLD and SIDIAP, and between doxycycline and depression compared to either comparators.

Conclusions: We identified a two-fold association of suicide-related events in individuals who use doxycycline compared to individuals who use erythromycin in CPRD GOLD and SIDIAP. We also identified a two-fold association of anxiety with doxycycline use compared to isotretinoin use in IPCI. Further results from a self-controlled case series analysis and within other indications did not confirm these associations. All results were included in the full regulatory assessment and no regulatory actions were taken by the European Medicines Agency for doxycycline.