Comparisons of Efficacy and Safety of Different Doses of Doxycycline for the Treatment of Moderate‐to‐Severe Acne Vulgaris: A Systematic Review and Network Meta‐Analysis

Background: The evidence is insufficient for the administration of subantimicrobial doses of doxycycline in the treatment of acne vulgaris, and the results from the limited studies were inconsistent. Objectives: This study aims to comprehensively compare the efficacy and safety of different doses of doxycycline for moderate‐to‐severe acne vulgaris. Methods: A systematic review and network meta‐analysis of randomized controlled trials were carried out. Literature was searched from PubMed (from inception to May 31, 2025), EMBASE (from inception to May 31, 2025), Cochrane Central Register of Controlled Trials (CENTRAL, from inception to May 31, 2025), Wanfang Data (from inception to May 11, 2025), and CNKI (from inception to May 11, 2025).The RCTs comparing the efficacy and safety of different doses of doxycycline in the treatment of acne vulgaris were included if they fulfilled the following inclusion criteria: (1) participants were adolescents or adults aged 12–60 years with moderate‐to‐severe acne vulgaris; (2) both interventions and controls were limited to different doses (0–200 mg/d) of doxycycline for different durations with or without the use of other systemic or topical drugs; and (3) outcomes at least included the changes in inflammatory skin lesions (papules, pustules, cysts, nodules, and so on) and the incidence of adverse events. Exclusion criteria were as follows: (1) duplicate records; (2) reviews, comments, meta‐analysis, guideline/consensus, proceeding abstract, case reports, and thesis; (3) studies with unavailable outcomes or incomplete data; (4) studies whose participants, controls, design, outcomes, and intervention failed to meet the inclusion criteria; and (5) not in English or Chinese language. Risk of bias was independently evaluated by two reviewers separately using the version 2.0 of the Cochrane risk‐of‐bias tool for randomized trials (RoB 2.0). The pooled effects on continuous variables were summarized as standardized mean differences (SMDs), and those on the dichotomous variable as odds ratio (OR). Standard pairwise comparisons using both fixed and random effect models as well as network meta‐analysis were carried out. Results: Totally, 635 records were potentially relevant based on initial screening. After excluding duplicates and publications that did not meet the inclusion criteria, four English articles describing four RCTs with a total of 1070 participants were finally included. All the four included studies administered 40 mg/d of doxycycline, three studies administered placebo, two studies administered 100 mg/d of doxycycline, and one study administered 80 and 160 mg/d of doxycycline. As the network meta‐analysis showed, for reduction of the count of inflammatory acne lesions, 40 mg/d of doxycycline was significantly more effective than that of placebo ( p = 0.03) but was insignificantly different from that of 80 mg/d ( p = 0.22), 100 mg/d ( p = 0.07), and 160 mg/d of doxycycline ( p = 0.08). As for the incidence of adverse events, compared with the 40 mg/d doxycycline group, the placebo group ( p = 0.46), 80 mg/d ( p = 0.53) and 160 mg/d doxycycline group ( p = 0.79) showed no significant difference, but 100 mg/d doxycycline group showed a significant increase (OR = 4.70, p < 0.01). There were some limitations in our analyses: some essential data were deduced according to the literature; since the courses of treatments differed among the included trials, the data at different time points were combined for meta‐analysis; the sample size was relatively small. Conclusions: The subantimicrobial dose (40 mg/d) of doxycycline showed similar efficacy and similar or more favorable safety profile in the reduction of inflammatory lesions of acne vulgaris compared with 80–160 mg/d of doxycycline.