Treatment options for acne

It has been difficult to derive definitive treatment guidelines for acne.7^,8 A key reason for this is that methodological differences between trials of treatment options, particularly with respect to outcome measures and grading of disease severity, make it hard to compare studies directly. Also, few studies have compared different types of treatment. In mild to moderate non-inflammatory acne, topical benzoyl peroxide or a topical retinoid is recommended.1^,3 In mild to moderate inflammatory acne, topical azelaic acid, or a topical antibacterial are further options.1^,3 Oral antibacterials are usually reserved for more severe inflammatory acne; for patients who have not responded adequately to topical therapy; for those at greater risk of scarring; or where lesions predominantly on the trunk make topical therapy impractical.3 Oral co-cyprindiol (cyproterone acetate and ethinylestradiol) is licensed for the treatment of women with severe acne who have not responded to prolonged oral antibacterial therapy.2 It may be helpful if contraception is also required but the risk of venous thromboembolism is higher than with a low-dose oral contraceptive.2 Oral isotretinoin should be reserved for patients with severe acne which has not responded to other therapies.9

Is minocycline more effective?

A systematic review identified 27 randomised trials that compared oral minocycline with placebo or other treatment in a total of 3,031 patients with inflammatory acne.10 Comparators included topical clindamycin, erythromycin and zinc, or fusidic acid; oral doxycycline, lymecycline, oxytetracycline or tetracycline; oral co-cyprindiol; and oral isotretinoin. The trials were generally small and of poor quality, and no meta-analysis of data was attempted due to lack of common outcome measures and unavailability of some primary data. In only two studies was minocycline more effective than the comparator (tetracycline or oxytetracycline). However, both were non-blinded studies of poor quality, which makes their results questionable. Overall, minocycline appeared to be an effective treatment for acne, but there was no convincing evidence that it was superior to other oral antibacterials.

In a later randomised controlled trial, involving 649 patients with mild to moderate inflammatory acne of the face, modified-release minocycline 100mg daily for 18 weeks was no more effective than oral oxytetracycline 500mg twice daily; topical benzoyl peroxide 5% twice daily; topical benzoyl peroxide 5% plus erythromycin 3% (Benzamycin) twice daily; or topical erythromycin 2% applied in the morning and 5% benzoyl peroxide in the evening.11 Moderate or greater improvement was reported by half to two-thirds of patients with any of the treatments.

Two other randomised controlled trials provide conflicting views of the effectiveness of minocycline compared with lymecycline.12^,13 In one study, involving 136 patients with moderate to severe acne, modified-release minocycline 100mg daily achieved the same mean reduction in the number of inflammatory lesions (the primary outcome measure) as lymecycline 300mg daily.12 In the other study, involving 86 patients, minocycline 50mg twice daily for 4 weeks then 50mg once daily for 8 weeks was more effective in reducing lesions than lymecycline 300mg once daily.13 However, there were key deficiencies in the reporting of the trial that make it difficult to interpret the trial's findings. For instance, it is unclear whether there was a power calculation for the study; data were reported for only 68 patients (79% of total); there was only graphical, rather than numerical, presentation of results; and no confidence intervals were included.

What about resistant bacteria?

Propionibacteria, especially P. acnes and P. granulosum are implicated in the pathogenesis of acne.14 The proportion of patients carrying strains resistant to tetracyclines appears to be increasing.14 Skin colonisation with such strains has been shown to reduce response to minocycline and oxytetracycline.11 However, the practical implication of this is unclear since propionibacterial susceptibility is rarely tested in clinical practice. Also, there may be other reasons why treatment with an oral antibacterial is unsuccessful, such as poor adherence, inadequate dose, very greasy skin or development of acute folliculitis.15 We can find no evidence to support the claim that minocycline offers less likelihood of propionibacterial resistance, nor that changing to minocycline from another tetracycline improves response.

Is minocycline easier to take?

Once-daily administration of minocycline, whether as the immediate- or modified-release preparation, provides a theoretical advantage over oxytetracycline and tetracycline, which need to be taken at least twice daily.2 In addition, minocycline does not have to be taken on an empty stomach, unlike these older tetracyclines. However, lymecycline and doxycycline are also administered once daily, and their absorption is also not affected by food.2

What about safety?

Minocycline, as with all tetracyclines, is contraindicated in pregnancy and children aged under 12 years. It appears to be little different to other tetracyclines in its ability to cause unwanted effects such as gastrointestinal upset, vaginal candidiasis, photosensitivity, hypersensitivity reactions, and benign intracranial hypertension.2^,10 However, minocycline appears to be the only tetracycline to cause slate-grey hyperpigmentation of the skin, which may be irreversible.16 In addition, lupus-like syndrome, less specific arthropathies, autoimmune hepatitis and vasculitis which are associated with the production of a range of autoantibodies, have been reported with minocycline.17–20 These abnormalities tend to occur after prolonged therapy (often longer than 12 months).21

While it is well known that many antibacterials (and other groups of drugs) can induce lupus-like syndromes, the likelihood with minocycline appears to be higher. A case-control study using data from the UK General Practice Research Database identified 29 cases of lupus-like syndromes among 27,688 patients with acne.18 The likelihood of this problem was 8.5 times higher (95% CI 2.1-35) with minocycline and 1.7 times higher (95% CI 0.4-8.1) with other tetracyclines compared with the risk for non-users and past users of all tetracyclines.The overall risk was significantly higher in females (14-times that in males). The pattern of antibodies induced by minocycline is unusual. Most patients with drug-induced lupus have antinuclear antibodies directed against histones; in minocycline-induced disease, these antibodies are unusual, and there is often a perinuclear antineutrophil cytoplasmic antibody (pANCA), which appears to be commonly associated with development of symptoms.19

The risk of developing minocycline-induced hepatitis is unknown. However, a UK retrospective review suggested two types of liver injury may occur.21 Firstly, there may be an acute hypersensitivity reaction with eosinophilia and cutaneous involvement (usually within 30 days). Much later (at a median of 1 year in women and 2 years in men), patients may develop a delayed autoimmune pattern with production of autoantibodies and, sometimes, clinical lupus.21 This pattern occurs more often in females. Summaries of product characteristics recommend that if minocycline is to be continued for over 6 months, patients should be monitored at least 3-monthly for symptoms and signs of hepatitis, systemic lupus erythematosus or unusual pigmentation.

Minocycline, like doxycycline and unlike other tetracyclines, can be used in patients with renal impairment. However, dose reduction and monitoring of renal fuction may be needed in patients with severe impairment.

Cost

At the recommended dose of 100mg daily, the approximate cost to the NHS of 6 months' treatment with minocycline 100mg capsules is £96; for modified-release capsules, the cost is £69. By comparison, the cost of oxytetracycline 500mg twice daily is £43; tetracycline 500mg twice daily £79; doxycycline 50mg once daily £34; and lymecycline 408mg once daily £46.

Conclusion

Treatment options for acne vulgaris include oral antibacterials, particularly the tetracyclines. Minocycline is often used instead of other tetracyclines. However, we can find no convincing evidence to support such a preference. Also, minocycline is associated with serious unwanted effects. Given the availability of tetracyclines that are as effective, safer and less expensive, we believe it is difficult to justify use of minocycline in the management of patients with acne