Antimicrob Agents Chemother. 2015 Aug;59(8):4510-20.

Systematic Review and Meta-Analysis To Estimate Antibacterial Treatment Effect in Acute Bacterial Skin and Skin Structure Infection.

Cates JE1, Mitrani-Gold FS2, Li G3, Mundy LM3.
  • 1University of North Carolina, Chapel Hill, North Carolina, USA.
  • 2GlaxoSmithKline (GSK), Research Triangle Park, North Carolina, USA
  • 3GSK, Collegeville, Pennsylvania, USA.



A systematic literature review and meta-analysis were conducted to inform on the non-inferiority trial design in acute bacterial skin and skin structure infection (ABSSSI). The primary endpoints included an early clinical treatment response (ECTR) defined as cessation of lesion spread at 48 to 72 h post-randomization and the test-of-cure (TOC) response defined as total resolution of the infection at 7 to 14 days post-treatment.  The systematic review identified no placebo-controlled trials in ABSSSI, 4 placebo-controlled trials in uncomplicated skin and soft tissue infection as a proxy for placebo in ABSSSI, 12 linezolid trials in ABSSSI, 3 ceftaroline trials in ABSSSI, and 2 trials for non-antibacterial treatment.  The ECTR rates at 48 to 72 h and corresponding 95% confidence intervals (CI) were 78.7% (95% CI, 61.1 to 96.3%) for linezolid, 74.0% (95% CI, 69.7 to 78.3%) for ceftaroline, and 59.0% (95% CI, 52.8 to 65.3%) for non-antibacterial treatment.  The early clinical treatment effect could not be estimated, given no available placebo or proxy-for-placebo data for this endpoint.  Clinical, methodological, and statistical heterogeneity influenced the selection of trials for the meta-analysis of the TOC treatment effect estimation.  The pooled estimates of the TOC treatment response were 31.0% (95% CI, 6.2 to 55.9%) for the proxy for placebo, 88.1% (95% CI, 81.0 to 95.1%) for linezolid, and 86.1% (95% CI, 83.7 to 88.6%) for ceftaroline.  The TOC clinical treatment effect estimation was 25.1% for linezolid and 27.8% for ceftaroline.  The antibacterial treatment effect estimation at TOC will inform on the design and analysis of future non-inferiority ABSSSI clinical trials.1

PMID: 25987628


Supplement (word count 1,293)

Acute bacterial skin and skin structure infection (ABSSSI) defines a heterogeneous group of moderate to severe types of skin and soft tissue infection. The estimated annual incidence (5%) of these infection types is a recognized clinical and economic healthcare burden in the United States (US).2 The 2013 regulatory guidance for clinical trials in ABSSSI from the US Food and Drug Administration (FDA) recommends revised inclusion criteria for infection type by lesion type and size as well as a primary endpoint for early clinical treatment response (ECTR).3 The purpose of our systematic review and meta-analysis was to determine the ABSSSI treatment effect estimation for two well-established antibacterial agents, linezolid and ceftaroline, at the recently recommended ECTR endpoint and the previously well-established test-of-cure (TOC) endpoint.1

Prior to the study initiation, linezolid and ceftaroline were identified as the active comparators for ABSSSI treatment effect estimation in a clinical trial design that would include subjects with methicillin-susceptible Staphylococcus aureus (MSSA) and methicillin-resistant S. aureus (MRSA) infection.  The systematic review utilized the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations for identification, screening, and analysis of trial data, when available, in ABSSSI for linezolid, ceftaroline, placebo, and non-antibacterial treatment responses.4  Given the lack of data for placebo treatment in infections meeting the 2013 guidance criteria for ABSSSI, proxy-for-placebo treatment response was estimated from placebo treatment in uncomplicated skin infections.  When data were available, the endpoints after treatment exposure were categorized as ECTR and TOC responses.  The ECTR was defined as cessation of lesion spread at 48 to 72 hours post-randomization, and the TOC response was defined as total resolution of the infection at 7 to 14 days post-treatment.   Subsequently, these estimations of treatment response were the basis of the treatment effect estimations for linezolid and ceftaroline.  Aligned with the 2013 FDA guidance for ABSSSI, we extracted information from each eligible study on the proportion of subjects who had resolution of spread in the skin infection, or reduction in size, after 48 to 72 hours of taking the drug or placebo (ECTR), and the proportion of subjects deemed to be cured of the infection after one to two weeks of treatment or placebo (TOC).  Treatment responses were pooled across studies for each drug or placebo and at each of the two endpoints. To estimate the antibacterial treatment effect of linezolid and of ceftaroline, we calculated the difference between the lower bound of the 95% confidence interval (CI) of the treatment response for each drug and the upper bound of the 95% CI for the placebo treatment in studies of uncomplicated skin infection.  The latter data were assumed as the proxy-for-placebo given the lack of data or evidence for placebo treatment effect for complicated or serious skin infections aligned with the new ABSSSI definition.

In our review, there were 12 trials that reported response rates for linezolid, three studies that reported response rates for ceftaroline, four studies that reported placebo response rates in uncomplicated SSTI, and two studies that reported non-antibacterial (ultraviolet) treatment response rates.  The ECTR, at 48 to 72 hours post-treatment, was 79% (95% CI, 61-96%) for linezolid, 74% (95% CI, 70- 78%) for ceftaroline, and 59.0% (95% CI, 52.8-65.3%) for ultraviolet (non-antibacterial) treatment.  There were no available data to estimate a placebo response at this ECTR endpoint based on proxy-for-placebo data in uncomplicated SSTI trials.  In contrast, the treatment response estimates at the TOC endpoint were 88% (95% CI, 81-95%) for linezolid , 86% (95% CI, 84-89%) for ceftaroline, and 31%, (95% CI, 6-56%) for placebo.  Together, these response estimates permitted a calculated 25.1% TOC treatment effect for linezolid and 27.8% TOC treatment effect for ceftaroline.

The systematic review, with meta-analysis, informed on the treatment response and treatment effect estimates of linezolid and ceftaroline in ABSSSI.  The findings are particularly useful in the design of future clinical trials that utilize either linezolid or ceftaroline as an active comparator and when the trial design and methodology incorporates the 2013 FDA ABSSSI trial recommendations.  More broadly, our methodological framework for the systematic review and the selection of trials for the meta-analysis can be modified or adopted to estimate treatment effect of other antibacterial agents under consideration as active comparators in future ABSSSI trials. There are some key similarities and distinctions in our systematic review and meta-analysis when compared to another ABSSSI systematic review that was published in the same year.5  Thom, et al. reported a comparison of treatment efficacy for oritavancin relative to nine other antimicrobial agents:  vancomycin, linezolid, tigecycline, dalbavancin, daptomycin, ceftaroline, teicoplanin, clindamycin, and telavancin.5  A network meta-analysis of the systematic review provided a framework for the comparison of the 10 antimicrobial agents. Although the two systematic reviews had different objectives, each systematic review evaluated the ECTR and TOC endpoints and included treatment response estimates for linezolid and ceftaroline.1,5  Additionally, both reviews acknowledged clinical and methodological heterogeneity in the identified trials, with potential inference of benefit for further standardization of ABSSSI trial design and analyses.  Greater standardization in future ABSSSI clinical trials will optimize comparative effectiveness research, clinical care, and health economics evaluation of ABSSSI treatment.

One key distinction in the content of the two systematic reviews was the consideration of placebo treatment response in ABSSSI treatment efficacy and treatment effect estimation. In our review, there was a pre-hoc assumption that ultraviolet light exposure in skin infection was not a proxy-for-placebo treatment response. We categorized data from trials of ultraviolet light exposure as a separate non-antibacterial treatment response; these data were available for only the ECTR endpoint.1 It is noteworthy that disease understanding of skin exposure to ultraviolet light remains incomplete and that this exposure has established evidence in both efficacy and safety. In 1903, the Nobel Prize in Physiology or Medicine was awarded to Niels Ryberg Finsen in recognition of his use of concentrated light therapy in the treatment of lupus vulgaris, a type of cutaneous tuberculosis.6,7 Data from an exploratory study in 2005 suggest that Finsen’s use of light in lupus vulgaris may mechanistically involve the production of singlet oxygen through radiation of porphyrins with light.7 It is, therefore, perhaps not surprising that two trials of ultraviolet light exposure in erysipelas and cellulitis from the late 1930s had a pooled ECTR of 59.0% (95% CI, 52.8-65.3%).1,8,9 This pooled ECTR for ultraviolet light seems implausibly high for an estimate of proxy-for-placebo treatment response in ABSSSI, especially when the placebo treatment response at TOC was 31% (95% CI, 6-56%) based on placebo treatment in four trials of uncomplicated skin infection.

The role of ultralight light therapy in current medical care continues to evolve.  While excess sun exposure has been associated with sunburn, melanoma, and other skin cancers, specific types of ultraviolet light therapy have reported treatment benefit in several conditions. Specific types of phototherapy, defined as ultraviolet A light, ultraviolet B light, psorelan ultraviolet A (PUVA), and pulsed dye laser have attributable benefits, if not also specific societal endorsements for use in specific patient populations (Table). These conditions include, but are not limited to, atopic dermatitis, morphea, mycosis fungoides, psoriasis, Sezary syndrome, and vitiligo.10-14, Future collaborative research efforts in phototherapy and photochemotherapy in disease-specific conditions have the potential to accelerate decision framework analyses and targeted treatment regimens.15, 16

In summary, the totality of evidence from our systematic review and meta-analysis in ABSSSI has potential significance for future regulatory reviews, approvals, uptake, and use of antibacterial agents for this indication. It is important to note that the studies used to calculate these treatment effects were restricted to adult populations and our findings may not be directly generalized to pediatric or other special populations.  However, since the natural history and clinical presentations of skin infections are similar in immunocompetent adult and pediatric populations, there may be similar antibacterial treatment responses and treatment effect estimations in adults and in children.


Table.  Medical conditions associated with treatment benefit from specific phototherapy.*


*Not necessarily representative of all associations of medical condition and treatment benefit

NB ultraviolet B light = Narrowband ultraviolet B light

PUVA = Psorelan ultraviolet A



  1. Cates J, Mitrani-Gold F, LI G, Mundy LM.  Systematic review and meta-analysis to estimate antibacterial treatment effect in acute bacterial skin and skin structure infection.  Antimicrob Agents Chemother  2015;59:4510-20.
  2. Ray GT, Suaya JA, Baxter R.  Incidence, microbiology, and patient characteristics of skin and soft tissue infections in a U.S. population:  a retrospective population-based study.  BMC Infect Dis 2013;13:252.
  3. United States Food and Drug Administration. Guidance for Industry. Acute bacterial skin and skin structure infections: developing drugs for treatment. Silver Spring, MD: USA 2013.  Available from: /Drugs/GuidanceComplianceRegulatoryInformation/Guidances /UCM071185.pdf
  4. Moher D, Liberati A, Tetzlaff J, Altman DG.  2009.  Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement.  Ann Intern Med  151(4):264–269.
  5. Thom H, Thompson JC, Scott DA, Halfpenny N, Sulham K, Corey GR.  Comparative efficacy of antibiotics for the treatment of acute bacterial skin and skin structure infections (ABSSSI):  a systematic review and network meta-analysis.  Curr Med Res Opin 2015;31:1539-51.
  6. The Nobel Prize in Physiology or Medicine 1903. Media AB 2014. Web. 11 Apr 2016. <>
  7. Møller KI, Kongshoj B, Philipsen PA, Thomsen VO, Wulf HC.  How Finsen’s light cured lupus vulgaris.  Photodermatol Photoimmunol Photomed 2005;21:118-24.
  8. Snodgrass WR, Anderson T.   Prontosil in erysipelas.  Brit Med J 1937; 2(3993):101–4.
  9. Snodgrass WR, Anderson T.  Sulphanilamide in the treatment of erysipelas.  Brit Med J 1937; 2(4014):1156–9.
  10. Taïeb A, Picardo M. Vitiligo.   N Engl J Med 2009; 360:160-9.
  11. Ling TC, Clayton TH, Crawley J, Exton LS, Goulden V. Ibbotson S. McKenna K, Mohd Mustapa MF, Rhodes LE, Sarkany R. Dawe RS. British Association of Dermatologists and British Photodermatology Group guidelines for the safe and effective use of psoralen–ultraviolet A therapy.  British Journal of Dermatology 2016;174:24–552015
  12. Olsen EA, Hodak E, Anderson T, Carter JB, Henderson M, Cooper K, Lim HW. Guidelines for phototherapy of mycosis fungoides and Sezary syndrome: A consensus statement of the United States Cutaneous Lymphoma Consortium.  J Am Acad Dermatol 2016;74:27-58.
  13. Garritsen FM, Brouwer MW, Limpens J, Spuls PI. Photo(chemo)therapy in the management ofatopic dermatitis: an updated systematic review with implications for practice and research. Br J Dermatol. 2014;170:501-13.
  14. Erceg A, de Jong EM, van de Kerkhof PC, Seyger MM. The efficacy of pulsed dye laser treatment for inflammatory skin diseases: a systematic review. J Am Acad Dermatol. 2013;69:609-15.
  15. Schneider LA, Hinrichs R, Scharffetter-Kochanek K.  Phototherapy and photochemotherapy. Clin Dermatol. 2008 Sep-Oct;26:464-76.
  16. Hamblin MR, Chiang LY, Lakshmanan S, Huang YY, Garcia-Diaz M, Karimi M, de Souza Rastelli AN, Chandran R.  Nanotechnology for photodynamic therapy: a perspective from the Laboratory of Dr. Michael R. Hamblin in the Wellman Center for Photomedicine at Massachusetts General Hospital and Harvard Medical School. Nanotechnol Rev 2015;4:359-372.



Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann