International Journal of Clinical Pharmacology and Therapeutics, Vol. 50 No.

3/2012 (157-161)

Oral tetracyclines may not be effective in treating acne: dominance of the placebo effect
Win L. Chiou

Review
2012 Dustri-Verlag Dr. K. Feistle ISSN 0946-1965 DOI 10.5414/CP201650 e-pub: February 27, 2012

Chiou Consulting Inc., Burr Ridge, IL, USA

Key words oral tetracyclines acne treatment placebo effect

Abstract. Oral tetracyclines (tetracycline, doxycycline, minocycline and lymecycline) have been commonly regarded as effective for treating both inflammatory and non-inflammatory acne. An excellent review of efficacy of 57 clinical trials by Simonart et al. in 2007 concluded that compared to the baseline before treatment, efficacies of all tetracyclines were similar (mean reduction in inflammatory lesions and non-inflammatory lesion being 54.3 1.4% and 45 2.6%, respectively) and were not affected by dosage amount (40 1,000 mg per day) and treatment period (4 24 weeks). These interesting findings may be pharmacodynamically rationalized by weak intrinsic anti-acne activity of tetracyclines and strong placebo effects. This hypothesis was supported by published data indicating that during weeks or months of daily administration, the placebo effect approached the effect of minocycline or doxycline in reducing acne lesions in three separate studies. The present work suggests the importance of considering placebo effects in the evaluation of anti-acne products. The treatability of acne was discussed in view of the slow and weak intrinsic anti-acne property of oral tetracyclines, and the reported fast (within days) elimination (curing) of severe acne by intralesional corticosteroids and antibiotics. The subantimicrobial or non-antimicrobial doses (e.g., only a small fraction) of various oral tetracyclines may be much lower than those commonly recognized.

ders [17, 18, 19, 20, 21, 22]. In an excellent review and correlation analysis of 57 clinical studies published between 1962 and 2007, Simonart et al. [1], reported that the efficacies of various tetracyclines (tetracycline, doxycycline, minocycline and lymecycline) as a class in reducing inflammatory lesions (papules and pustules) and non-inflammatory lesions (whiteheads and blackheads) from initial baseline were all similar, i.e., no evidence of superiority of one tetracycline over another. Furthermore, their efficacies were independent of the dosage amount (40 1,000 mg per day) and treatment period (4 24 weeks). Their findings are summarized in Table 1 for quick comparison. It is remarkable that for each of four tetracyclines, their mean reductions in inflammatory lesions were essentially identical, and their mean reductions in non-inflammatory lesions were very similar. The mean reductions for all four tetracyclines were 54 1.4% and 45 7.6%, respectively, for inflammatory and non-inflammatory lesions (Table 1). In the above cited 57 studies, no data from placebo controls was considered in their analysis. In other words, in these studies treatment efficacy was assumed proven when reduction in lesion count at the end of study compared to the baseline value was shown to be statistically significant. The main purpose of this communication is to present a rationale suggesting that contrary to earlier reports and common notion, oral tetracyclines in acne treatments may not be effective in treating non-inflammatory lesions and may be only slightly better than placebo in treating inflammatory lesions. Two commonly prescribed tetracyclines, doxycline and minocycline, will be used for illustration. New perspectives on the treat-

Received August 26, 2011; accepted December 18, 2011 Correspondence to Win L. Chiou, PhD Chiou Consulting Inc., 80 Burr Ridge Parkway, PO Box 187, Burr Ridge, IL 60527, USA win@chiouconsulting. com

Introduction
Oral tetracyclines have been extensively used in the treatment of acne for more than half a century because of their known antimicrobial properties [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. More recently, the subantimicrobial doses of oral tetracyclines have been found to be useful in treating acne, rosacea and other disor-

Chiou

158

Table 1. Relevant data* from the tetracycline studies reported by Simonart et al. [1]. Drug Tetracxcline Doxycycline Minocycline Lymecycline Mean SD Dosage range (mg/day) 500 1,000 40 200 50 200 150 300 Period range (weeks) 8 24 4 24 6 20 all 12 Mean SD reduction in inflammatory lesions (%) 54.3 19.5 (n = 14) 52.5 17.7 (n = 6) 56.0 12.2 (n = 12) 54.3 8.2 (n = 4) 54.3 1.4 Mean SD reduction in noninflammatory lesions (%) 44.7 13.5 (n = 8) 47.0 5.8 (n = 4) 41.4 17.6 (n = 10) 46.7 6.5 (n = 3) 45.0 2.6

*Data obtained from their Tables 2 4, n = number of studies.

ability of acne will also be presented. It is hoped that the present work may stimulate discussions on this and related subjects.

Efficacy or inefficacy of tetracyclines to treat noninflammatory acne

The reported similar ability of the four tetracyclines to moderately reduce noninflammatory lesions (Table 1) independent of doses (over a 25-fold difference between 40 mg and 1,000 mg) and treatment periods (over a 6-fold difference) appears to be very interesting. It is postulated that such unique pharmacodynamic phenomena may be rationalized by a weak, intrinsic activity of the tetracyclines to treat non-inflammatory acne, and the observed efficacy may be mostly attributed to the placebo effect. When the dose is reduced to zero (i.e., placebo), one should expect existence of a substantial anti-acne effect based on their does-effect correlation [1]. The results from two excellent largescale studies on extended-release minocycline [17, 18] tend to support the above hypothesis. These studies were approved by the United States Food and Drug Administration (FDA) and involved multi-centers with a 12week, randomized, double-blinded, placebocontrolled protocol. They were considered as Phase II and Phase III studies in preparation of a new drug application [9, 17, 18]. In the first study [17], the mean reductions of non-inflammatory lesion counts at 12 weeks were 18.0% and 15.7%, following daily administration of 1 mg/kg dose (n = 59) and placebo control (n = 55), respectively. The above nearly identical results apparently indicate an almost total lack of efficacy of

minocycline in treating non-inflammatory acne. The above negative finding was also confirmed in a second combined study [18] with a mean reduction of 14.9% (n = 615) in the treatment group and that of 6.3% in the placebo group (n = 309): no statistically significant difference between them. With the above results, the FDA did not approve the extended-release product for treating non-inflammatory acne. A total lack of efficacy was also found at 2 mg/kg and 3 mg/kg dose [17]. The above lack of efficacy of minocycline in treating non-inflammatory acne could not be expected from earlier 10 clinical studies [1] in which there were no placebo controls employed. It is also of interest to note that after 12 weeks of drug treatment the mean reduction of 14.9% from baseline from the two pooled studies [18, Table 6] was much lower than the mean of 41.4% from the previous 10 minocycline studies (Table 1) or the mean of 45% from the previous 25 studies of the four various tetracyclines (Table 1). The reason for such a huge (~ 3-fold) difference is not known. Perhaps it may partly relate to a more rigid criteria employed in the FDAapproved protocol. An earlier study by Skidmore et al. [19] reported the success of using subantimicrobial dose (20 mg twice a day) of doxycycline to treat inflammatory and non-inflammatory lesions; this study was included in the analysis by Simonart et al. [1]. After 6 months of daily treatment, the mean reduction in comedones (non-inflammatory lesions) from the treatment group (n = 21), 53.6%, was significantly (p < 0.01) higher (~ 5-fold) than from the placebo group (n = 19), 10.6%. However, the above encouraging result seemed not consistent with other parameters reported in the same article [19]. For example, at the end of the study on a scale of 7.0, the mean

Efficacy of oral tetracyclines in acne treatment

159
(39.4%) and the 2 mg/kg (49.3%) or 3 mg/kg (46.6%) group. When 1 mg/kg data from the Phase II and Phase III studies were combined [18], mean reductions from the treatment (n = 674) and placebo (n = 364) groups were 45.5% and 32.4%, respectively; the difference being only 13.1%; the placebo can be calculated to contribute 71% to the total drug effect (100 32.4/45.5). Viewed differently, the mean inflammatory lesions remaining on the face after 12 weeks of daily administration from the treatment group and the placebo group were 54.5% and 67.6%, respectively. Although such a small difference was concluded [18] to be statistically significant (p < 0.001), its clinical significance may be subject to debate. This is because these small differences may not be readily noticed by patients themselves or by others. It seems worthwhile to note that at 12 weeks 8.7% of patients on placebo alone had skin cleared or almost cleared of all acne lesions, while only 16.6% of patients on 1 mg/kg achieved this [18]. The difference was only less than 8%. The weak and slow-acting nature of doxycline in treating inflammatory lesions seemed also evident from the previous study [19]. For example, at 2, 4 and 6 months doxycline at 20 mg twice a day reduced mean inflammatory lesion counts by 24%, 36% and 50.1%, respectively, in 21 patients, while the placebo reduced by 9%, 29% and 30.2%, respectively, in 19 patients. The differences attributed to doxycline alone were only 15%, 5.8% and 19.9%, respectively, at 2, 4 and 6 months reflecting low and slow intrinsic activity of the drug. Furthermore, the reported [19] great similarity in clinicians global assessment scores and patient self-assessment scores between the treatment and placebo group at 6 months as mentioned above also appeared to support the notion that doxycline is not that efficient or effective in treating inflammatory acne, and the observed apparent efficacy, especially, in the first 4 months, may be largely attributed to the placebo effect. It has been reported that in acne studies 6 8 weeks is an appropriate time to assess response [10]. In a recent extensive review on clinical applications of non-antimicrobial tetracyclines in dermatology [22], clinical data from a paper by Tossi et al. [20] was cited as evidence to show the efficacy of sub-antimicro-

difference in clinicians global assessment scores (also taking into consideration of improvement in inflammatory lesion counts) between the treatment and placebo group was only 0.7 (4.4 vs. 5.1) or only 10% of the total score; barely statistically significant [19]) and that in patient self-assessment scores only 0.5 (4.8 vs. 5.3; statistically not significant). Interestingly, at 2 and 4 months there were virtually no differences in patient self-assessment scores between the treatment and placebo group (4.9 vs. 5.1 at 2 months; 5.0 vs. 5.4 at 4 months). Similar results were also found in clinicians global assessment scores. Also, at 4 months the reductions in non-inflammatory lesion count from both groups were very similar, namely 32% for the treatment and 24% for the placebo. Thus, a daily treatment for four months resulted in no significant improvement in non-inflammatory lesions over the placebo control. The above review suggests that like minocycline, the doxycycline in that study was very weak and slow in combating non-inflammatory acne, and the often reported or observed clinical efficacy of doxycycline in treating non-inflammatory acne may also be largely attributed to the placebo effect. It is noted that the flawed design of the study [19] has been pointed out [20]. Clearly, more studies are needed to support the present hypothesis of weak intrinsic activity of oral tetracyclines in treating non-inflammatory acne.

Efficacy or inefficacy of tetracyclines in treating inflammatory acne

The weak and slow-acting property of minocycline in treating inflammatory lesions was also evident in the large-scale Phase II [17] and Phase III [18] studies. For example, in the Phase II study involving 233 patients at 4 and 8 weeks there were no statistically significant differences in percent reduction of lesion counts between patients (n = 55) receiving placebo (~ 30% reduction at 4 weeks and 40% reduction at 8 weeks) and patients receiving 1 mg/kg (n = 59), 2 mg/kg (n = 59) or 3 mg/kg (n = 60) of the minocycline. Similarly, after 12 weeks of treatment, there were no statistically significant differences in lesion reduction between the placebo group

Chiou

160
effects. Obviously such a hypothesis needs to be confirmed clinically.

bial doxycyline (20 mg twice daily) for treating acne in 50 patients. At 1, 2 and 3 months their mean reduction in inflammatory lesion count were 49, 88 and 92%, respectively (based on their Table 3 [20]). Practically identical results were also obtained using 100 mg dose in 50 patients. These high efficacy results were quite different from data reported by others (Table 1). The reason for this huge difference is not known. It may be possible that the placebo effect in their patients might be unusually strong.

References
[1] Simonart T, Dramaix M, De Maertelaer V. Efficacy of tetracyclines in the treatment of acne vulgaris: a review. Br J Dermatol. 2008; 158: 208-216. PubMed Del Rosso JQ, Kim G. Optimizing use of oral antibiotics in acne vulgaris. Dermatol Clin. 2009; 27: 33-42. doi:10.1016/j.det.2008.07.006 PubMed Ingram JR, Grindlay DJC, Williams HC. Management of acne vulgaris: an evidence-based update. Clin Exp Dermatol. 2010; 35: 351-354. doi:1 0.1 111/j.1365-2230.2009.03683.x PubMed Habif TP. 2010 Clinical Dermatology. 5th Ed. Mosby Elsevier. p. 217-255. Gannon M, Underhill M, Wellik KE, Strickland CG. Clinical inquiries. Which oral antibiotics are best for acne? J Fam Pract. 2011; 60: 290-292. PubMed Hayashi N, Kawashima M. Efficacy of oral antibiotics on acne vulgaris and their effects on quality of life: a multicenter randomized controlled trial using minocycline, roxithromycin and faropenem. J Dermatol. 2011; 38: 111-119. doi:10.1 11 1/j.1346-8138.2010.00969.x PubMed Ochsendorf F. Minocycline in acne vulgaris: benefits and risks. Am J Clin Dermatol. 2010; 11: 327-341. doi:10.2165/11319280-000000000-00 0 0 0 PubMed Garner SE, Eady A, Popescu CM, Newton J, Wan Po A. Summary of minocycline for acne vulgaris: efficacy and safety. Evid-Based Child Health. 2011; 6: 545-546. Leyden JJ, Del Ross JQ. Oral antibiotic therapy for acne vulgaris: pharmacokinetics and pharmacodynamic perspectives. J Clin Aesthetic Dermatol. 2011; 4: 40-47. Williams HC, Dellovalle RP, Garner S., Acne vulgaris. Published online Aug 30. 2011. D01:10,1016/ 50140-6736(11) 60321-8. Thiboutot D. Dermatologists do not yet fully understand the clinical significance of antibiotic use and bacterial resistance in patients with acne: comment on Antibiotics, acne, and Staphylococcus aureus colonization. Arch Dermatol. 2011; 147: 921-922. doi:10.1001/archdermatol.20 11.2 01 PubMed Sathish D, Shayeda D, Rao YM. Acne and its treatment options a review. Curr Drug Deliv. 2011; 6: 634-639. doi:10.2174/1567 201117 976 35 540 PubMed Dohil M. Clinical update on acne. Johns Hopkins Adv Stud Med. 2008; 8: 24-29. Newman MD, Bowe WP, Heughebaert C, Shalita AR. Therapeutic considerations for severe nodular acne. Am J Clin Dermatol. 2011; 12: 7-14. doi:10.2165/11532280-000000000-00000 PubMed Skidmore R, Kovach R, Walker C, Thomas J, Bradshaw M, Leyden J, Powala C, Ashley R. Effects of subantimicrobial-dose doxycycline in the treatment of moderate acne. Arch Dermatol. 2003; 139: 459-464. doi:10.1001/archderm.13 9.4 . 459 PubMed

[2] [3]

Treatability of acne
The weak- and slow-acting nature of oral tetracyclines in treating acne discussed above may be more evident or better appreciated if one considers that severe inflammatory nodulocystic acne (conventionally, treatable only after months of dosing with potent oral isotretinon) can be completely flattened (i.e., counted as elimination of lesions) with an intralesional injection of triamcinolone acetonide and lincomycin within 1 week [23]. Also, the effect of such intralesional steroid injection to treat papules and pustules is immediate and gratifying [4, 23]. There is a widely accepted notion that acne treatments work by preventing new lesions rather than treating existing ones [10]. This author believes that the primary goal of treatments should be to treat existing acne lesions. In view of the great success of intralesional corticosteroids [4, 14, 23], it is postulated that an ideal anti-acne preparation, either oral or topical, should be able to clear most of existing acne lesions in 1 2 weeks. The dominance of placebo effects in topical acne treatment will be the subject of next communication.

[4] [5]

[6]

[7]

[8]

[9]

[10] [11]

[12]

Subantimicrobial dose range

Presently, doses of 40 50 mg per day of doxycycline have been tested for non-inflammatory clinical applications in acne or other disorders [22]. In view of the above marked dose independency in anti-acne effect, it is possible that only a very small fraction of antimicrobial doses of various tetracyclines may be sufficient to produce similar clinical

[13] [14]

[15]

Efficacy of oral tetracyclines in acne treatment [16] Miller ST, Stevermer JJ. Low-dose doxycycline moderately effective for acne. J Fam Pract. 2003; 52: 594-597. PubMed [17] Stewart DM, Torok HM, Weiss JS, Plott RT; Solodyn Phase 2 Study Group. Dose-ranging efficacy of new once-daily extended-release minocycline for acne vulgaris. Cutis. 2006; 78 (Suppl): 11-20. PubMed [18] Fleischer AB Jr, Dinehart S, Stough D, Plott RT; Solodyn Phase 2 Study Group; Solodyn Phase 3 Study Group. Safety and efficacy of a new extended-release formulation of minocycline. Cutis. 2006; 78 (Suppl): 21-31. PubMed [19] Kircik LH. Doxycycline and minocycline for the management of acne: a review of efficacy and safety with emphasis on clinical implications. J Drugs Dermatol. 2010; 9: 1407-1411. PubMed [20] Toossi P, Farshchian M, Malekzad F, Mohtasham N, Kimyai-Asadi A. Subantimicrobial-dose doxycycline in the treatment of moderate facial acne. J Drugs Dermatol. 2008; 7: 1149-1152. PubMed [21] Griffin MO, Ceballos G, Villarreal FJ. Tetracycline compounds with non-antimicrobial organ protective properties: possible mechanisms of action. Pharmacol Res. 2011; 63: 102-107. doi:10.1016/j.phrs.2010.10.004 PubMed [22] Monk E, Shalita A, Siegel DM. Clinical applications of non-antimicrobial tetracyclines in dermatology. Pharmacol Res. 2011; 63: 130-145. doi:1 0.1016/j.phrs.2010.10.007 PubMed [23] Mahajan BB, Garg G. Therapeutic efficacy of intralesional triamcinolone acetonide versus intralesional triamcinolone acetonide plus lincomycin in the treatment of nodulocystic acne. Indian J Dermatol Venereol Leprol. 2003; 69: 217-219. PubMed

161

Copyright of International Journal of Clinical Pharmacology & Therapeutics is the property of Dustri-Verlag Dr. Karl Feistle GmbH & Co., KG and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.