- Email: [email protected]
Newer glycopeptide antibiotics for treatment of complicated skin and soft tissue infections: systematic review, network meta-analysis and cost analysis
Accepted Manuscript Newer glycopeptide antibiotics for treatment of complicated skin and soft tissue infections: a systematic review, network meta-analysis and cost analysis Rajender Agarwal, Sarah M. Bartsch, Brendan J. Kelly, Malinda Prewitt, Yulun Liu, Yong Chen, Craig A. Umscheid PII:
S1198-743X(17)30491-3
DOI:
10.1016/j.cmi.2017.08.028
Reference:
CMI 1060
To appear in:
Clinical Microbiology and Infection
Received Date: 13 June 2017 Revised Date:
29 August 2017
Accepted Date: 30 August 2017
Please cite this article as: Agarwal R, Bartsch SM, Kelly BJ, Prewitt M, Liu Y, Chen Y, Umscheid CA, Newer glycopeptide antibiotics for treatment of complicated skin and soft tissue infections: a systematic review, network meta-analysis and cost analysis, Clinical Microbiology and Infection (2017), doi: 10.1016/j.cmi.2017.08.028. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Newer glycopeptide antibiotics for treatment of complicated skin and soft tissue infections: a systematic review, network meta-analysis and cost analysis
RI PT
Rajender Agarwal 1,2, Sarah M. Bartsch 3,4, Brendan J. Kelly 5, Malinda Prewitt 6, Yulun Liu7, Yong Chen7, Craig A. Umscheid 7,8 1)
Division of Hospital Medicine, North Mississippi Medical Center; 2) Kelley School of
SC
Business, Indiana University; 3) Public Health Computational and Operations Research, Johns Hopkins Bloomberg School of Public Health; 4) Global Obesity Prevention Center, Johns
M AN U
Hopkins Bloomberg School of Public Health; 5) Division of Infectious Diseases, Perelman School of Medicine, University of Pennsylvania; 6) Division of Infectious Diseases, North Mississippi Medical Center; 7) Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine, University of Pennsylvania; 8) Center for Evidence-based Practice,
TE D
University of Pennsylvania Health System and the Division of General Internal Medicine, Perelman School of Medicine, University of Pennsylvania
EP
Category: Systematic review Running title: Newer antibiotics for skin infections Corresponding author:
AC C
Rajender Agarwal, MD MPH
Medical Director, Division of Hospital Medicine, North Mississippi Medical Center 830 South Gloster, Tupelo, MS 38801 email: [email protected]; phone: 215-873-1820; fax: 662-377-2920
1
ACCEPTED MANUSCRIPT
ABSTRACT Objectives. Skin and soft tissue infections (SSTIs) carry significant economic burden, as well as
RI PT
morbidity and mortality, especially when caused by methicillin-resistant Staphylococcus aureus (MRSA). Several new MRSA-active antibiotics have been developed, including semisynthetic glycopeptides (telavancin, dalbavancin and oritavancin). Of these, dalbavancin and oritavancin
SC
offer extended dosing intervals.
Methods. We performed a systematic review, network meta-analysis and cost analysis to
M AN U
compare the newer glycopeptides to standard care and each other for the treatment of complicated SSTIs (cSSTI). A search for randomized controlled trials (RCTs) was conducted in MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials. We also developed a model to evaluate the costs associated with dalbavancin and oritavancin from the
TE D
third-party payer perspective.
Results. Seven RCTs met inclusion criteria. Network meta-analyses suggested that the clinical response to telavancin, dalbavancin and oritavancin was similar to standard care (odds ratio [OR]
EP
1.09, 95% confidence interval [CI] 0.90-1.33; OR 0.78, CI 0.52-1.18; and OR 1.06, CI 0.85-1.33, respectively). Head-to-head comparisons showed no difference in clinical response between
AC C
oritavancin and dalbavancin (OR 1.36; CI 0.85-2.18), oritavancin and telavancin (OR 0.98; CI 0.72-1.31) or dalbavancin and telavancin (OR 0.72; CI 0.45-1.13). Telavancin had a higher incidence of overall adverse events compared to standard care (OR 1.33; CI 1.10-1.61). Compared to telavancin, there were fewer overall adverse events with dalbavancin (OR 0.58; CI 0.45-0.76) and oritavancin (OR 0.71; CI 0.55-0.92). Studies were of high quality overall. Our cost analyses demonstrated that dalbavancin and oritavancin were less costly compared to 2
ACCEPTED MANUSCRIPT
standard care under baseline assumptions and many scenarios evaluated. The use of dalbavancin could save third-party payers $1,442 to $4,803 per cSSTI, while the use of oritavancin could
RI PT
save $3,571 to $6,932 per cSSTI. Conclusions. Dalbavancin and oritavancin demonstrate efficacy and safety comparable to
standard care in well-designed RCTs and result in cost savings when standard care is treatment
AC C
EP
TE D
M AN U
SC
that covers MRSA.
3
ACCEPTED MANUSCRIPT
INTRODUCTION Skin and soft tissue infections (SSTIs) carry a significant economic burden, mainly driven by the high cost of hospitalization and inpatient treatment with intravenous (IV) antibiotics (1, 2).
RI PT
SSTIs that require complex management (e.g., surgical procedures) or that occur in the setting of significant comorbidities are broadly referred to as complicated SSTIs (cSSTIs) (3-5). In 2013, the Food and Drug Administration (FDA) introduced a new definition for such skin infections
SC
termed “acute bacterial skin and skin structure infections” (ABSSSI). These infections include cellulitis, erysipelas, major skin abscesses and wound infections with a minimum lesion surface
M AN U
area of 75 cm2 and accompanied by signs of systemic inflammation or significant medical comorbidities (6). While uncomplicated SSTIs respond well to oral antibiotic therapy, cSSTIs typically require IV antibiotic therapy (7).
Additionally, there has recently been a sharp rise in the incidence of infections caused by
TE D
methicillin-resistant Staphylococcus aureus (MRSA) (8, 9). This has resulted in increased use of vancomycin and the emergence of vancomycin resistant strains of S. aureus (9, 10). As a consequence, there has been an emergence of new antibiotics to combat the potential increase in
EP
morbidity and mortality from gram-positive (e.g., Staphylococcus, Streptococcus,
AC C
Corynebacterium, Clostridium, Listeria) SSTIs (11). Vancomycin is the most frequently prescribed treatment for MRSA infections overall and
is currently the second most common antibiotic used in hospitals (12). Continued efforts to develop safe and effective alternatives to vancomycin have led to the development of more potent semisynthetic glycopeptides – telavancin, dalbavancin and oritavancin (13). Of these, dalbavancin and oritavancin have long half-lives and extensive tissue distribution, thus allowing 4
ACCEPTED MANUSCRIPT
the benefit of extended dosing intervals (13-16). Current national guidelines on treatment of SSTIs do not incorporate evidence on
RI PT
semisynthetic glycopeptides, which may offer the advantage of outpatient treatment given their novel dosing regimens (5). In this paper, we compare the efficacy and safety of the newer
glycopeptide antibiotics with standard care and each other for the treatment of cSSTI using the technique of network meta-analysis. For the purposes of this review, we defined standard care as
SC
vancomycin or its traditional alternatives (such as linezolid, tedizolid, daptomycin, clindamycin,
M AN U
trimethoprim-sulfamethoxazole, doxycycline, oxacillin, cefazolin, ceftaroline and tigecycline) used for treatment of gram-positive SSTIs. We then perform a cost analysis to compare the costs of standard care to treatment with dalbavancin and oritavancin and each other, given their novel dosing regimens. Our cost analysis is relevant for cases in which the empirical treatment of
METHODS
EP
Study design
TE D
choice is a drug that covers MRSA.
A systematic review and network meta-analysis of randomized controlled trials (RCTs) was
AC C
performed in accordance with the Preferred Reporting Items for Systematic Reviews and Metaanalyses (PRISMA) guidelines (17). Network meta-analysis (also known as mixed-treatment comparisons) allows the synthesis of direct and indirect evidence across a network of randomized trials to infer the relative effectiveness of multiple treatments or interventions. In this method, evidence on the relative performance of treatments A and B is obtained from trials that directly compare them (known as direct evidence) and also from trials that compare them 5
ACCEPTED MANUSCRIPT
through a common comparator (known as indirect evidence). By combining direct and indirect evidence, more precise estimates on relative performance can be obtained. Moreover, relative performance on treatments that have not been directly compared in trials can be estimated (18,
RI PT
19).
Literature review
SC
We searched MEDLINE (including articles “in process” and non-indexed citations) using the search strategy described in the Supplementary material (Table S1), the Cochrane Central
M AN U
Register of Controlled Trials and EMBASE using the keywords telavancin, dalbavancin and oritavancin. We searched clinicaltrials.gov for any relevant completed studies pending results. Searches were completed in May 2017. Screening of titles and abstracts was done by one author (RA) which was followed by full-text screening by two independent authors (RA and MP). We
TE D
included in our review all RCTs that: (1) included adults with cSSTIs and (2) compared one of the newer glycopeptide antibiotics to standard care for gram positive SSTIs. The primary outcome was the clinical response one to two weeks after end of therapy,
EP
broadly defined as an improvement in the signs and symptoms associated with the cSSTI. Secondary efficacy outcomes included the clinical response among patients infected with S.
AC C
aureus (both methicillin-sensitive and resistant), and the clinical response at 48-72 hours among patients with ABSSSI. We also examined the overall and serious adverse events reported in the RCTs.
We extracted data on patient population, details of the intervention and comparison, size of the study sample and our outcome measures. We evaluated the quality of included RCTs by 6
ACCEPTED MANUSCRIPT
examining the randomization and blinding procedures and the extent of attrition using the Cochrane risk of bias tool (20).
RI PT
Meta-analysis We performed two types of meta-analysis. First, we performed standard pairwise meta-analyses in RevMan version 5.3 using the Mantel-Haenszel procedure that assumes a fixed effect size
SC
(21). We measured between-study heterogeneity using the I2 statistic (22). We defined
substantial heterogeneity as an I2 statistic ≥ 50%, which would imply that real differences exist
M AN U
between study results that would not be explained by chance alone. We also evaluated the significance of between-study heterogeneity using the likelihood ratio test statistic comparing fixed effects and random-effects meta-analysis models. In the presence of substantial heterogeneity, meta-analysis using the random-effects model was considered (23). Second, we
TE D
performed a fixed-effects network meta-analysis using the “netmeta” package in R version 3.4.1. The estimates used in our meta-analyses were odds ratios (OR) and 95% confidence intervals (CI). Results from network meta-analyses were reported in the text, unless the results from the
reported.
EP
network and pairwise meta-analyses were inconsistent. In such cases, both estimates were
AC C
A fundamental assumption for network meta-analysis is transitivity. This means that one
can learn about treatment A versus B via treatment C (i.e., the common comparator C is transitive). An interpretation of transitivity is that the common comparator C (e.g., the standard care in our analysis) is similar when it appears in A versus C and B versus C trials (e.g., dalbavancin vs. standard care RCTs; and oritavancin vs. standard care RCTs). An equivalent way to assess transitivity is that studies which directly compare A with C and B with C would have 7
ACCEPTED MANUSCRIPT
similar distributions of effect modifiers (19). We assessed our transitivity assumption conceptually and if our network contained a closed loop, we assessed the consistency assumption (i.e., agreement between direct and indirect evidence) statistically by comparing direct and
RI PT
indirect summary effects (18).
Cost analysis
SC
Using TreeAge Pro 2016 (TreeAge Software Inc., Williamstown, MA), we developed a
computer simulation model to compare the costs associated with dalbavancin and oritavancin
M AN U
from the third-party payer perspective (i.e., direct costs). We focused on these two drugs (and not telavancin) based on our hypothesis that despite their high requisition costs, outpatient treatment of cSSTI would help offset the hospitalization costs associated with the use of standard treatment. Adult cSSTI patients enter the model and receive either one of the new drugs or one of
TE D
the commonly used vancomycin based regimens (vancomycin only or vancomycin followed by either linezolid, trimethoprim-sulfamethoxazole [TMP-SMX], doxycycline, or clindamycin). A patient’s drug treatment regimen determined the location of treatment, where newer antibiotics
EP
for cSSTI are given in an outpatient setting or freestanding infusion center, while standard antibiotics for cSSTI require initial hospitalization.
AC C
The dosing and treatment regimen for all drugs followed those described in the literature
(5, 24, 25). Patients treated with dalbavancin received 1 g over 30 minutes on the first day and 500 mg over 30 minutes 7 days later (i.e., two doses over the course of one week, each requiring an outpatient visit) (24). Patients treated with oritavancin received 1200 mg once over 3 hours (25). Those treated with only vancomycin received 15 mg/kg every 12 hours for 7 to 10 days. Patients treated with vancomycin followed by another drug received 15 mg/kg of vancomycin 8
ACCEPTED MANUSCRIPT
every 12 hours for 3 days, followed by linezolid (600 mg), TMP-SMX (two double strength 800 mg-160 mg tablets twice a day), doxycycline (100 mg twice daily), or clindamycin (300 mg four times daily) for the remainder of their 10 to 14-day treatment regimen. Patients receiving
RI PT
dalbavancin or oritavancin incurred the costs of the drug, evaluation and management (i.e.,
outpatient visit cost), infusion supplies, and infusion services for each hour of infusion and the
costs of the drugs as well as hospitalization.
SC
therapy duration, while those treated with any of the standard treatment approaches incurred the
M AN U
Table 1 shows the model input parameters, values, and sources. All parameters are agespecific where available (age distribution of patients based on SSTI discharges from the Healthcare Cost and Utilization Project [HCUP]) (26). We used mean hospitalization costs per patient for a primary discharge diagnosis of SSTI (ICD-9 code 682.9). All drug costs are based on average wholesale unit price for all available products, including generics and repackages
TE D
(27). All costs are presented in 2017 US dollars, converted using a 3% discount rate (28). Each simulation ran 1000 adult cSSTI patients through the model 1000 times for a total
EP
of 1 million trials with unique outcomes. Monte Carlo simulation (i.e., probabilistic sensitivity analysis) varied each parameter throughout the ranges in Table 1. As the economic value of
AC C
outpatient therapy may change with the cost of the new drug and/or changes in hospitalization cost, additional sensitivity analyses varied the cost of hospitalization for all ages (using values from $4,000 to $8,000, as well as the median value of $5,466), the cost of dalbavancin per 500 mg (0.5 to 2 times the current cost), and the cost of oritavancin per 400 mg (0.5 to 2 times the current cost). Lastly, we performed analyses to estimate the additional days of treatment avoided with use of dalbavancin and oritavancin. 9
ACCEPTED MANUSCRIPT
RESULTS Study selection
RI PT
Our searches yielded a total of 573 citations, of which 15 were retrieved for full-text review after title and abstract screening. Seven studies met inclusion criteria (24, 25, 29-33) (Figure 1).
Among the excluded studies was an early RCT comparing dalbavancin to standard care that did
SC
not have a defined treatment regimen in the comparison group (it was determined by the
M AN U
clinician prior to randomization) (34).
Study characteristics
Combined, the 7 studies included 6398 patients. Study characteristics are detailed in Table S2. All study participants had cSSTI caused by a suspected or confirmed gram-positive organism
TE D
and defined by the presence of a major abscess, infected burn, extensive cellulitis, wound infection or ulcer, along with systemic and local signs of infection. Of note, the definitions of cSSTI used by the included studies varied, but all incorporated the principles of the 2013 FDA
EP
definition (e.g., accompanying signs of systemic inflammation or significant medical comorbidities). Mean age in all studies was less than 50 years, and patients were predominantly
AC C
white and male. Cellulitis, abscess and wound infection were the most common infection types. The interventions used were IV telavancin in 3 studies (29-31), IV dalbavancin (administered as 1 g on day 1, followed by 500 mg on day 8) in 2 studies (24, 32), and IV oritavancin (given as a single 1200 mg dose) in 2 studies (25, 33). In all studies, standard therapy was administered and dosing adjustments were made according to standard practice by unblinded personnel. Other drugs (e.g., aztreonam and metronidazole) were permitted for polymicrobial infections. All 10
ACCEPTED MANUSCRIPT
studies were funded by the pharmaceutical company that manufactured the respective study drugs. One or more authors reported a conflict of interest in all studies.
RI PT
Study quality Appropriate performance of randomization and allocation concealment was reported in all
studies. All studies were described as double-blind, although details about participant blinding
SC
were not reported in one study (31). Outcome assessment was reported as blinded in all studies. Attrition was greater than acceptable limits in one study (32). Intention-to-treat analysis was
M AN U
performed in all studies. Study quality is depicted in Figure 2.
Outcomes
The network of eligible comparisons for our outcome ‘clinical response – all patients’ is depicted in Figure 3. The studies form a star shape network, where node size represents the treatment
TE D
option, and line thickness is proportional to the number of direct comparisons. Similar networks were noted for all other outcomes examined. As depicted in Figure 3, the network does not
EP
contain a closed loop (i.e., triangular loop). In such cases, the assumption of consistency cannot be evaluated statistically. However, the transitivity assumption can be assessed conceptually.
AC C
Although not identical, the populations, comparators, outcomes and key effect modifiers were clinically similar across our included studies, which supports the transitivity assumption. Results of the direct comparisons and network estimates for the clinical response and safety outcomes are shown in Table 2. For our primary analysis, we used data from intention-to-treat analyses reported in the studies when available. We also performed additional sensitivity analyses using only per-protocol data reported in the studies. Forest plots of the pairwise meta-analyses are 11
ACCEPTED MANUSCRIPT
included in Figures S1-S5 in the Supplementary material. Clinical response
RI PT
All included studies reported data on clinical response. Further, all studies except one (32) reported clinical response among patients infected with S. aureus and MRSA. Pairwise metaanalyses of the included RCTs showed that the clinical response to televancin, dalbavancin and
SC
oritavancin was similar to standard care. These results were consistent with those of the network meta-analyses. Telavancin, dalbavancin and oritavancin had similar clinical response in head-to-
M AN U
head comparisons in the network meta-analysis. Among patients with MRSA, telavancin was found to have a superior clinical response compared to standard care (OR 1.62; 95% CI 1.002.61) in pairwise and network meta-analysis. Dalbavancin and oritavancin were similar to standard care. Network meta-analysis suggested that oritavancin and dalbavancin were similar to
TE D
telavancin, and oritavancin was similar to dalbavancin in head-to-head comparisons. Three studies evaluated early clinical response at 48-72 hours among patients with ABSSSI (24, 25, 33). Network meta-analysis of these studies showed no differences between oritavancin and
EP
dalbavancin, oritavancin and standard care, or dalbavancin and standard care. Studies reported intention-to-treat and per-protocol data separately for the outcome of ‘clinical response – all
AC C
patients’. Pairwise and network meta-analyses using per-protocol data showed that the clinical response to televancin, dalbavancin and oritavancin was similar to standard care (Table S3 and Figure S3 in Supplementary material). Network meta-analysis showed that telavancin, dalbavancin and oritavancin had similar clinical response in head-to-head comparisons using per-protocol data. Adverse events 12
ACCEPTED MANUSCRIPT
All included studies reported data on overall and serious adverse events. Table S4 in the Supplementary material describes the serious adverse events that were deemed to be treatment related. Dalbavancin was found to have fewer overall adverse events compared to standard care
RI PT
(OR 0.77; 95% CI 0.64-0.93) in pairwise and network meta-analyses, but similar odds of serious adverse events. Telavancin had higher overall adverse events (OR 1.33; 95% CI 1.10-1.61) and serious adverse events (OR 1.54; 95% CI 1.07-2.21) compared to standard care in pairwise and
SC
network meta-analyses. Oritavancin had fewer overall adverse events compared to telavancin (OR 0.71; 95% CI 0.55-0.92) but the odds of serious adverse events were similar. Dalbavancin
M AN U
had fewer overall adverse events (OR 0.58; 95% CI 0.45-0.76) and serious adverse events (OR 0.50; 95% CI 0.29-0.86) compared to telavancin. There were no differences in the overall and serious adverse events between oritavancin and dalbavancin. Intention-to-treat and per-protocol adverse event data were not separately reported in the included studies.(23) Transient elevations
TE D
in liver enzymes were noted with dalbavancin and oritavancin (25, 32, 33), but these differences were not reported as significant when compared with the comparison group.
EP
Cost analysis
Table 3 shows the average cost of treatment for cSSTI patients for the evaluated treatment
AC C
regimens. With baseline assumptions (age-specific hospitalization costs and current drug costs), the use of each of the new drugs was less costly compared with vancomycin based regimens. The use of dalbavancin could save third-party payers $1,442 to $4,803 per cSSTI patient treated, while the use of oritavancin could save $3,571 to $6,932 per cSSTI patient treated. Compared to standard of care, dalbavancin could save between 6.5-10.0 days of treatment and oritavancin could save between 7.5-11.0 days of treatment. 13
ACCEPTED MANUSCRIPT
When all parameters were held at baseline, dalbavancin was the least costly treatment option up to a cost of $1,132, at which point oritavancin became the least costly treatment option. At dalbavancin costs ≥ $2,600, all other treatment strategies cost less except for the
RI PT
strategy of intravenous vancomycin followed by oral linezolid for completion of the antibiotic course. When all parameters were held at baseline, oritavancin was the least costly treatment option up to a cost of $1,905, at which point dalbavancin became the least costly option.
SC
Oritavancin remained less costly than vancomycin based treatments at twice its current cost
M AN U
($2,390 per 400 mg).
When hospitalization costs were decreased to $4,000 per patient, treatment with dalbavancin was less costly than treatment with vancomycin followed by linezolid, but was more expensive than the other vancomycin based regimens. In probabilistic sensitivity analysis, 100% of iterations favored treatment with oritavancin (i.e., it was the least costly option) for all
TE D
scenarios explored except when hospitalization cost $4,000 and oritavancin cost $2,390 per 400 mg (twice the current cost), at which point 100% of iterations favored treatment with
DISCUSSION
EP
vancomycin followed by doxycycline.
AC C
Our systematic review and meta-analysis suggests that newer glycopeptide antibiotics are similar to standard therapy for treatment of cSSTIs caused by a suspected or confirmed gram positive organism. This finding was also true for cSSTIs caused specifically by MRSA with the exception of telavancin, which demonstrated increased odds of clinical response compared to standard care. Head-to-head comparisons suggested no differences in clinical response between the new agents. There was increased odds of adverse events with telavancin. Compared with standard treatment, 14
ACCEPTED MANUSCRIPT
dalbavancin and oritavancin were less costly compared to standard of care under baseline assumptions and many of the scenarios evaluated. Even reducing the total cost of hospitalization for cSSTI patients to $4,000 was not enough to overcome the potential savings of outpatient
RI PT
infusion therapy at oritavancin’s current price point. Importantly, our cost analysis is relevant for cases in which the empirical treatment of choice is a drug that covers MRSA.
In the studies that evaluated telavancin, it was administered intravenously for 4-14 days
SC
in two studies (29, 30) and 7-14 days in one study (31). We found that the clinical response to
M AN U
telavancin was superior compared to standard care for patients infected with MRSA. However, the duration of treatment with telavancin is similar to conventional agents. Further, we found increased odds of adverse events with telavancin, which was supported by a past meta-analysis that compared telavancin with vancomycin among patients with cSSTI and hospital acquired pneumonia, and demonstrated an increased risk of serious adverse events and acute renal failure
TE D
among telavancin recipients (35). In light of this, telavancin may have little net advantage over vancomycin for the treatment of cSSTIs, especially in the presence of other vancomycin alternatives, such as linezolid. A recent large meta-analysis demonstrated that linezolid was
EP
associated with a significantly higher clinical and microbiological cure rate as well as a reduction
AC C
in the length of hospital stay and cost of hospital charges compared with vancomycin among patients with SSTI (36).
The RCTs that evaluated oritavancin and dalbavancin demonstrated that patients with
complicated SSTIs can be treated successfully with one or two doses of these agents without an increase in adverse events. Further, a recent RCT demonstrated that a single 1500 mg IV dose of dalbavanvin was non-inferior to the two-dose regimen (37) and this dosing has now been 15
ACCEPTED MANUSCRIPT
approved by the FDA. These agents represent an advance in the treatment of these infections by eliminating the need for inpatient or prolonged outpatient parenteral therapy, thus potentially decreasing resource utilization, improving compliance and preventing potential complications
RI PT
from peripherally inserted central catheters. Further, no monitoring for blood concentrations is required in contrast to vancomycin (13, 15, 38). However, use of these agents may be limited by their high acquisition costs (39), as well as unexpected adverse events. For example, the SOLO II
SC
trial reported that osteomyelitis occurred in 5 patients on oritavancin and none in the vancomycin
M AN U
arm. This finding, along with data from earlier studies, led to a warning issued by the FDA (40). Our review is strengthened by the fact that all included studies were described as doubleblind and had low risk of bias in our quality evaluation using established criteria. A key limitation of the review is that most of the studies were published prior to the FDA guidance on designing RCTs to evaluate drugs for cSSTIs (6), with the result that the definition of cSSTI
TE D
varied between studies. The comparator regimens were also different as no formal guidance currently exists on empiric therapy and duration of treatment for cSSTIs.
EP
Although our cost analysis focuses on the cost of cSSTI treatments in the United States, the direction of our results is likely generalizable to other countries, particularly those with
AC C
similar standards of care. In addition, we examined non-monetary outcomes to compare the “value” of each approach, such as number of treatment days avoided, which may be less sensitive to differences in direct drug costs between countries. Another potential limitation of the evidence-base is that all RCTs were sponsored by the
pharmaceutical companies that produced the antibiotics. In addition, our findings are limited to cases of cSSTI where a gram-positive organism is suspected or confirmed. Future research 16
ACCEPTED MANUSCRIPT
should test the comparative effectiveness of these newer drugs in patients with cSSTI that meet the criteria defined by the FDA.
RI PT
CONCLUSION Dalbavancin and oritavancin demonstrate efficacy and safety comparable to vancomycin alone or followed by linezolid when used to treat cSSTI where a gram-positive organism is suspected
SC
or confirmed. Oritavancin yielded greater cost savings compared to dalbavancin. Treatment with these new antibiotics consists of one or two doses and offers cost savings over standard therapies
M AN U
for cSSTI when standard care is treatment that covers MRSA. These drugs are potential options for the outpatient management of complicated SSTIs, defined as SSTIs with systemic signs of infection or significant medical comorbidities.
Funding: none
TE D
Acknowledgements: none
Transparency declaration: The authors have no conflicts of interest and no sources of industry
AC C
EP
support to declare
17
ACCEPTED MANUSCRIPT
REFERENCES 1.
Jenkins TC, Sabel AL, Sarcone EE, Price CS, Mehler PS, Burman WJ. Skin and soft-
antimicrobial stewardship. Clin Infect Dis. 2010;51(8):895-903. 2.
RI PT
tissue infections requiring hospitalization at an academic medical center: opportunities for
Lee BY, Singh A, David MZ, Bartsch SM, Slayton RB, Huang SS, et al. The economic
SC
burden of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA). Clin Microbiol Infect. 2013;19(6):528-36.
Miller LG, Eisenberg DF, Liu H, Chang CL, Wang Y, Luthra R, et al. Incidence of skin
M AN U
3.
and soft tissue infections in ambulatory and inpatient settings, 2005-2010. BMC Infect Dis. 2015;15:362. 4.
Russo A, Concia E, Cristini F, De Rosa FG, Esposito S, Menichetti F, et al. Current and
future trends in antibiotic therapy of acute bacterial skin and skin-structure infections. Clin
5.
TE D
Microbiol Infect. 2016;22 Suppl 2:S27-36.
Stevens DL, Bisno AL, Chambers HF, Dellinger EP, Goldstein EJ, Gorbach SL, et al.
Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014
US Food and Drug Administration CfDEaR. Guidance for industry. Acute bacterial skin
AC C
6.
EP
update by the Infectious Diseases Society of America. Clin Infect Dis. 2014;59(2):e10-52.
and skin-structure infections: developing drugs for treatment 2010 [Available from: http://www.fda.gov/downloads/Drugs/.../Guidances/ucm071185.pdf. 7.
Miller LG, Daum RS, Creech CB, Young D, Downing MD, Eells SJ, et al. Clindamycin
versus trimethoprim-sulfamethoxazole for uncomplicated skin infections. N Engl J Med. 2015;372(12):1093-103. 18
ACCEPTED MANUSCRIPT
8.
Chambers HF. The changing epidemiology of Staphylococcus aureus? Emerg Infect Dis.
2001;7(2):178-82. 9.
Daum RS. Clinical practice. Skin and soft-tissue infections caused by methicillin-
10.
Cosgrove SE, Carroll KC, Perl TM. Staphylococcus aureus with reduced susceptibility to
vancomycin. Clin Infect Dis. 2004;39(4):539-45.
Tsoulas C, Nathwani D. Review of meta-analyses of vancomycin compared with new
SC
11.
RI PT
resistant Staphylococcus aureus. N Engl J Med. 2007;357(4):380-90.
Antimicrob Agents. 2015;46(1):1-7. 12.
M AN U
treatments for Gram-positive skin and soft-tissue infections: Are we any clearer? Int J
Cardona AF, Wilson SE. Skin and soft-tissue infections: a critical review and the role of
telavancin in their treatment. Clin Infect Dis. 2015;61 Suppl 2:S69-78. 13.
Klinker KP, Borgert SJ. Beyond Vancomycin: The Tail of the Lipoglycopeptides. Clin
14.
TE D
Ther. 2015;37(12):2619-36.
Lin G, Credito K, Ednie LM, Appelbaum PC. Antistaphylococcal activity of dalbavancin,
an experimental glycopeptide. Antimicrob Agents Chemother. 2005;49(2):770-2. Saravolatz LD, Stein GE. Oritavancin: A Long-Half-Life Lipoglycopeptide. Clin Infect
EP
15.
Dis. 2015;61(4):627-32.
Zhanel GG, Schweizer F, Karlowsky JA. Oritavancin: mechanism of action. Clin Infect
AC C
16.
Dis. 2012;54 Suppl 3:S214-9. 17.
Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The
PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339:b2700. 18.
Cipriani A, Higgins JP, Geddes JR, Salanti G. Conceptual and technical challenges in 19
ACCEPTED MANUSCRIPT
network meta-analysis. Ann Intern Med. 2013;159(2):130-7. 19.
Salanti G. Indirect and mixed-treatment comparison, network, or multiple-treatments
synthesis tool. Res Synth Methods. 2012;3(2):80-97. 20.
Collaboration TC. Cochrane handbook for systematic reviews of interventions (Version
5.1.0) March 2011 [Available from: http://cochrane.handbook.org.
Collins R, Gray R, Godwin J, Peto R. Avoidance of large biases and large random errors
SC
21.
RI PT
meta-analysis: many names, many benefits, many concerns for the next generation evidence
1987;6(3):245-54. 22.
Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-
analyses. BMJ. 2003;327(7414):557-60. 23.
M AN U
in the assessment of moderate treatment effects: the need for systematic overviews. Stat Med.
DerSimonian R, Kacker R. Random-effects model for meta-analysis of clinical trials: an
24.
TE D
update. Contemp Clin Trials. 2007;28(2):105-14.
Boucher HW, Wilcox M, Talbot GH, Puttagunta S, Das AF, Dunne MW. Once-weekly
dalbavancin versus daily conventional therapy for skin infection. N Engl J Med.
25.
EP
2014;370(23):2169-79.
Corey GR, Good S, Jiang H, Moeck G, Wikler M, Green S, et al. Single-dose oritavancin
AC C
versus 7-10 days of vancomycin in the treatment of gram-positive acute bacterial skin and skin structure infections: the SOLO II noninferiority study. Clin Infect Dis. 2015;60(2):254-62. 26.
United States Department of Health & Human Services. HCUP facts and figures:
statistics on hospital-based care in the United States Rockville, MD: AHRQ: Agency for Healthcare Research and Quality; 2013 [Available from: http://hcupnet.ahrq.gov/HCUPnet.jsp. 27.
Red Book Online [Internet]. Truven Health Analytics. 2016 [cited October 24, 2016]. 20
ACCEPTED MANUSCRIPT
28.
M.R. Gold JES, L.B. Russell, and M.C. Weinstein. New York: Oxford University Press;
1996. 29.
Stryjewski ME, O'Riordan WD, Lau WK, Pien FD, Dunbar LM, Vallee M, et al.
due to gram-positive bacteria. Clin Infect Dis. 2005;40(11):1601-7. 30.
RI PT
Telavancin versus standard therapy for treatment of complicated skin and soft-tissue infections
Stryjewski ME, Chu VH, O'Riordan WD, Warren BL, Dunbar LM, Young DM, et al.
SC
Telavancin versus standard therapy for treatment of complicated skin and skin structure
infections caused by gram-positive bacteria: FAST 2 study. Antimicrob Agents Chemother.
31.
M AN U
2006;50(3):862-7.
Stryjewski ME, Graham DR, Wilson SE, O'Riordan W, Young D, Lentnek A, et al.
Telavancin versus vancomycin for the treatment of complicated skin and skin-structure infections caused by gram-positive organisms. Clin Infect Dis. 2008;46(11):1683-93. Jauregui LE, Babazadeh S, Seltzer E, Goldberg L, Krievins D, Frederick M, et al.
TE D
32.
Randomized, double-blind comparison of once-weekly dalbavancin versus twice-daily linezolid therapy for the treatment of complicated skin and skin structure infections. Clin Infect Dis.
33.
EP
2005;41(10):1407-15.
Corey GR, Kabler H, Mehra P, Gupta S, Overcash JS, Porwal A, et al. Single-dose
90. 34.
AC C
oritavancin in the treatment of acute bacterial skin infections. N Engl J Med. 2014;370(23):2180-
Seltzer E, Dorr MB, Goldstein BP, Perry M, Dowell JA, Henkel T, et al. Once-weekly
dalbavancin versus standard-of-care antimicrobial regimens for treatment of skin and soft-tissue infections. Clin Infect Dis. 2003;37(10):1298-303. 35.
Polyzos KA, Mavros MN, Vardakas KZ, Makris MC, Rafailidis PI, Falagas ME. Efficacy 21
ACCEPTED MANUSCRIPT
and safety of telavancin in clinical trials: a systematic review and meta-analysis. PLoS One. 2012;7(8):e41870. 36.
Yue J, Dong BR, Yang M, Chen X, Wu T, Liu GJ. Linezolid versus vancomycin for skin
37.
RI PT
and soft tissue infections. Cochrane Database Syst Rev. 2016(1):CD008056.
Dunne MW, Puttagunta S, Giordano P, Krievins D, Zelasky M, Baldassarre J. A
Randomized Clinical Trial of Single-Dose Versus Weekly Dalbavancin for Treatment of Acute
38.
SC
Bacterial Skin and Skin Structure Infection. Clin Infect Dis. 2016;62(5):545-51.
Nathwani D, Eckmann C, Lawson W, Stephens JM, Macahilig C, Solem CT, et al. Pan-
M AN U
European early switch/early discharge opportunities exist for hospitalized patients with methicillin-resistant Staphylococcus aureus complicated skin and soft tissue infections. Clin Microbiol Infect. 2014;20(10):993-1000.
Drugs for MRSA skin and soft-tissue infections. JAMA. 2014;312(15):1583-4.
40.
Oritavancin: Highlights of prescribing information 2014 [Available from:
TE D
39.
https://www.accessdata.fda.gov/drugsatfda_docs/label/2014/206334s000lbl.pdf. 41.
Centers for Medicare & Medicaid Services. Physician Fee Schedule Baltimore, MD:
EP
Centers for Medicare & Medicaid; 2016 [updated October 18. Available from: https://www.cms.gov/apps/physician-fee-schedule/overview.aspx. Centers for medicare & Medicaid Services. Durable Medical Equipment, Prosthetics,
AC C
42.
Orthotics, and Supplies (DMEPOS) Fee Schedule Baltimore, MD: Centers for medicare & Medicaid 2016 [updated January. Available from: https://www.cms.gov/medicare/medicare-feefor-service-payment/DMEPOSFeeSched/DMEPOS-Fee-Schedule.html. 43.
U.S. Department of Health and Human Services, Centers for Disease Control and
Prevention, National Center for Health Statistics. Anthropometric Reference Data for Children 22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
and Adults: United States, 2011-2014. 2016.
23
ACCEPTED MANUSCRIPT Tables Table 1. Model input parameters, values, and sources
Table 2: Results from pairwise and network meta-analysis
Figure Legends: Figure 1: Selection process for included studies Figure 2: Cochrane risk of bias table
AC C
EP
TE D
M AN U
SC
Figure 3: Network of treatment comparisons for primary outcome
RI PT
Table 3: Average total cost in dollars (95% uncertainty interval) of treatment regimens
24
ACCEPTED MANUSCRIPT Table 1. Model input parameters, values, and sources Parameter
Mean or Median
Standard Deviation or Range
Source
1,194.8 1,841.64 0.015 190.05 3.16
0.008 23.69 0.85
(27) (27) (27) (27) (27)
Costs (2017 $US)
RI PT
Oritavancin (per 400mg) Dalbavancin (per 500mg) Vancomycin (per mg) Linezolid (per 600mg) Trimethoprim-sulfamethoxazole (per 800mg-160mg) Doxycycline (per 100mg)
3.31
1.85
(27)
Clindamycin (per 300mg)
4.45
0.54
(27)
5,318.03 8,113.79 7,973.10 8,657.41 71.91 19.55 111.37 45.07
565.01 818.24 956.68 970.19 0.93
(26) (26) (26) (26) (41) (41) (41) (42)
3 0.5 -
7 – 10 10 – 14
(25) (24) (25) (24)
-
10-14
(5)
21.3 34.7 36.6 7.5
-
(26) (26) (26) (26)
Hospitalization
¶
M AN U
≥ 85 years First hour of infusion Additional hours of infusion Office visit for infusion (evaluation and management)
SC
18 to 44 years 45 to 64 years 65 to 84 years
TE D
Infusion supplies Duration Oritavancin (hours) Dalbavancin (hours) Vancomycin only (days) Vancomycin followed by linezolid (days)* Vancomycin followed by trimethoprim-sulfamethoxazole, doxycycline or clindamycin (days)* Patient Characteristics Age distribution of patients (%) 18 to 44 years 45 to 64 years 65 to 84 years ≥ 85 years
AC C
EP
Average Weight (kg) 18 to 44 years 82.8 73.5 – 91.5 (43) 45 to 64 years 84.5 76.6 – 91.5 (43) 65 to 84 years 78.6 64.4 – 90.6 (43) ≥ 85 years 71.8 (43) ¶ Mean and standard deviation of hospitalization costs for patients with a discharge diagnosis of skin and soft-tissue infection (ICD-9 code 682.9) *Total treatment duration represented three days of intravenous vancomycin followed by the oral antibiotic to complete therapy
ACCEPTED MANUSCRIPT
Table 2: Results from pairwise and network meta-analysis Treatment comparison
Number of studies
Pairwise meta-analysis Odds ratio (95% CI)
AC C
EP
TE D
M AN U
SC
RI PT
Clinical response – all patients Oritavancin vs. Standard care 2 1.06 (0.85-1.33) Dalbavancin vs. Standard care 2 0.79 (0.53-1.20) Telavancin vs. Standard care 3 1.09 (0.90-1.33) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Clinical response – patients infected with methicillin-resistant Staphylococcus aureus Oritavancin vs. Standard care 2 0.95 (0.56-1.61) Dalbavancin vs. Standard care 1 0.73 (0.06-8.33) Telavancin vs. Standard care 3 1.62 (1.01-2.61) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Early clinical response at 48-72 hours among patients with acute bacterial skin and skin-structure infection (ABSSSI) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Standard care 2 1.02 (0.80-1.27) Dalbavancin vs. Standard care 1 0.99 (0.76-1.30) Overall adverse events Oritavancin vs. Standard care 2 0.94 (0.79-1.13) Dalbavancin vs. Standard care 2 0.77 (0.64-0.93) Telavancin vs. Standard care 3 1.33 (1.10-1.60) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Serious adverse events Oritavancin vs. Standard care 2 0.99 (0.68-1.45) Dalbavancin vs. Standard care 2 0.77 (0.52-1.14) Telavancin vs. Standard care 3 1.52 (1.07-2.17) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Abbreviations: CI – confidence interval; NA – not available
Network meta-analysis Odds ratio (95% CI) 1.06 (0.85-1.33) 0.78 (0.52-1.18) 1.09 (0.90-1.33) 1.36 (0.85-2.18) 0.98 (0.72-1.31) 0.72 (0.45-1.13) 0.95 (0.56-1.61) 0.73 (0.06-8.33) 1.62 (1.00-2.61) 1.29 (0.11-15.48) 0.59 (0.29-1.19) 0.45 (0.04-5.39) 1.02 (0.72-1.46) 1.02 (0.80-1.27) 0.99 (0.76-1.30) 0.94 (0.79-1.13) 0.77 (0.64-0.93) 1.33 (1.10-1.61) 1.23 (0.94-1.58) 0.71 (0.55-0.92) 0.58 (0.45-0.76) 0.99 (0.68-1.45) 0.77 (0.52-1.15) 1.54 (1.07-2.21) 1.28 (0.74-2.22) 0.64 (0.38-1.09) 0.50 (0.29-0.86)
ACCEPTED MANUSCRIPT
Table 3: Average total cost in dollars (95% uncertainty interval) of treatment regimens Total Cost for Treatment in Outpatient Settings
Total Cost for Treatment Requiring Hospitalization Median ($5,466) SSTI Hospitalization Cost
$8,000 SSTI Hospitalization Cost
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
7,569 (4,425 – 8,958)
5,782 (5,744 – 5,806)
4,318 (4,283 – 4,339)
10,784 (7,626 – 12,200)
8,998 (8,955 – 9,041)
7,534 (7,491 – 7,577)
8,316 (8,279 – 8,339) 11,532 (11,490 – 11,580)
7,478 (4,329 – 8,890)
5,692 (5,678 – 5,700)
4,226 (4,213 – 4,234)
8,225 (8,212 – 8,234)
7,524 (4,376 – 8,935)
5,738 (5,724 – 5,747)
4,272 (4,260 – 4,281)
8,272 (8,259 – 8,280)
5,637 (5,624 – 5,646)
4,172 (4,159 – 4,180)
8,171 (8,158 – 8,179)
RI PT
Age-Specific SSTI Hospitalization Cost -
SC
-
M AN U
-
7,423 (4,274 – 8,834)
AC C
EP
Dalbavancin ($1,842 per 500 5,982 (5,982 – 5,982) mg)* Dalbavancin ($921 per 500 mg)^ 3,219 (3,219 – 3,219) Dalbavancin ($3,684 per 500 11,507 (11,507 – 11,507) mg)† Oritavancin ($1,195 per 400 mg)* 3,852 (3,852 – 3,852) Oritavancin ($598 per 400 mg)^ 2,060 (2,060 – 2,060) Oritavancin ($2,390 per 400 mg) 7,436 (7,436 – 7,436) † Vancomycin Vancomycin followed by linezolid Vancomycin followed by trimethoprim-sulfamethoxazole Vancomycin followed by clindamycin Vancomycin followed by doxycycline *Current cost ^Half the current cost †Two times the current cost Abbreviations: SSTI – skin and soft tissue infection
$4,000 SSTI Hospitalization Cost
TE D
Treatment Regimen
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S1198-743X(17)30491-3
DOI:
10.1016/j.cmi.2017.08.028
Reference:
CMI 1060
To appear in:
Clinical Microbiology and Infection
Received Date: 13 June 2017 Revised Date:
29 August 2017
Accepted Date: 30 August 2017
Please cite this article as: Agarwal R, Bartsch SM, Kelly BJ, Prewitt M, Liu Y, Chen Y, Umscheid CA, Newer glycopeptide antibiotics for treatment of complicated skin and soft tissue infections: a systematic review, network meta-analysis and cost analysis, Clinical Microbiology and Infection (2017), doi: 10.1016/j.cmi.2017.08.028. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Newer glycopeptide antibiotics for treatment of complicated skin and soft tissue infections: a systematic review, network meta-analysis and cost analysis
RI PT
Rajender Agarwal 1,2, Sarah M. Bartsch 3,4, Brendan J. Kelly 5, Malinda Prewitt 6, Yulun Liu7, Yong Chen7, Craig A. Umscheid 7,8 1)
Division of Hospital Medicine, North Mississippi Medical Center; 2) Kelley School of
SC
Business, Indiana University; 3) Public Health Computational and Operations Research, Johns Hopkins Bloomberg School of Public Health; 4) Global Obesity Prevention Center, Johns
M AN U
Hopkins Bloomberg School of Public Health; 5) Division of Infectious Diseases, Perelman School of Medicine, University of Pennsylvania; 6) Division of Infectious Diseases, North Mississippi Medical Center; 7) Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine, University of Pennsylvania; 8) Center for Evidence-based Practice,
TE D
University of Pennsylvania Health System and the Division of General Internal Medicine, Perelman School of Medicine, University of Pennsylvania
EP
Category: Systematic review Running title: Newer antibiotics for skin infections Corresponding author:
AC C
Rajender Agarwal, MD MPH
Medical Director, Division of Hospital Medicine, North Mississippi Medical Center 830 South Gloster, Tupelo, MS 38801 email: [email protected]; phone: 215-873-1820; fax: 662-377-2920
1
ACCEPTED MANUSCRIPT
ABSTRACT Objectives. Skin and soft tissue infections (SSTIs) carry significant economic burden, as well as
RI PT
morbidity and mortality, especially when caused by methicillin-resistant Staphylococcus aureus (MRSA). Several new MRSA-active antibiotics have been developed, including semisynthetic glycopeptides (telavancin, dalbavancin and oritavancin). Of these, dalbavancin and oritavancin
SC
offer extended dosing intervals.
Methods. We performed a systematic review, network meta-analysis and cost analysis to
M AN U
compare the newer glycopeptides to standard care and each other for the treatment of complicated SSTIs (cSSTI). A search for randomized controlled trials (RCTs) was conducted in MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials. We also developed a model to evaluate the costs associated with dalbavancin and oritavancin from the
TE D
third-party payer perspective.
Results. Seven RCTs met inclusion criteria. Network meta-analyses suggested that the clinical response to telavancin, dalbavancin and oritavancin was similar to standard care (odds ratio [OR]
EP
1.09, 95% confidence interval [CI] 0.90-1.33; OR 0.78, CI 0.52-1.18; and OR 1.06, CI 0.85-1.33, respectively). Head-to-head comparisons showed no difference in clinical response between
AC C
oritavancin and dalbavancin (OR 1.36; CI 0.85-2.18), oritavancin and telavancin (OR 0.98; CI 0.72-1.31) or dalbavancin and telavancin (OR 0.72; CI 0.45-1.13). Telavancin had a higher incidence of overall adverse events compared to standard care (OR 1.33; CI 1.10-1.61). Compared to telavancin, there were fewer overall adverse events with dalbavancin (OR 0.58; CI 0.45-0.76) and oritavancin (OR 0.71; CI 0.55-0.92). Studies were of high quality overall. Our cost analyses demonstrated that dalbavancin and oritavancin were less costly compared to 2
ACCEPTED MANUSCRIPT
standard care under baseline assumptions and many scenarios evaluated. The use of dalbavancin could save third-party payers $1,442 to $4,803 per cSSTI, while the use of oritavancin could
RI PT
save $3,571 to $6,932 per cSSTI. Conclusions. Dalbavancin and oritavancin demonstrate efficacy and safety comparable to
standard care in well-designed RCTs and result in cost savings when standard care is treatment
AC C
EP
TE D
M AN U
SC
that covers MRSA.
3
ACCEPTED MANUSCRIPT
INTRODUCTION Skin and soft tissue infections (SSTIs) carry a significant economic burden, mainly driven by the high cost of hospitalization and inpatient treatment with intravenous (IV) antibiotics (1, 2).
RI PT
SSTIs that require complex management (e.g., surgical procedures) or that occur in the setting of significant comorbidities are broadly referred to as complicated SSTIs (cSSTIs) (3-5). In 2013, the Food and Drug Administration (FDA) introduced a new definition for such skin infections
SC
termed “acute bacterial skin and skin structure infections” (ABSSSI). These infections include cellulitis, erysipelas, major skin abscesses and wound infections with a minimum lesion surface
M AN U
area of 75 cm2 and accompanied by signs of systemic inflammation or significant medical comorbidities (6). While uncomplicated SSTIs respond well to oral antibiotic therapy, cSSTIs typically require IV antibiotic therapy (7).
Additionally, there has recently been a sharp rise in the incidence of infections caused by
TE D
methicillin-resistant Staphylococcus aureus (MRSA) (8, 9). This has resulted in increased use of vancomycin and the emergence of vancomycin resistant strains of S. aureus (9, 10). As a consequence, there has been an emergence of new antibiotics to combat the potential increase in
EP
morbidity and mortality from gram-positive (e.g., Staphylococcus, Streptococcus,
AC C
Corynebacterium, Clostridium, Listeria) SSTIs (11). Vancomycin is the most frequently prescribed treatment for MRSA infections overall and
is currently the second most common antibiotic used in hospitals (12). Continued efforts to develop safe and effective alternatives to vancomycin have led to the development of more potent semisynthetic glycopeptides – telavancin, dalbavancin and oritavancin (13). Of these, dalbavancin and oritavancin have long half-lives and extensive tissue distribution, thus allowing 4
ACCEPTED MANUSCRIPT
the benefit of extended dosing intervals (13-16). Current national guidelines on treatment of SSTIs do not incorporate evidence on
RI PT
semisynthetic glycopeptides, which may offer the advantage of outpatient treatment given their novel dosing regimens (5). In this paper, we compare the efficacy and safety of the newer
glycopeptide antibiotics with standard care and each other for the treatment of cSSTI using the technique of network meta-analysis. For the purposes of this review, we defined standard care as
SC
vancomycin or its traditional alternatives (such as linezolid, tedizolid, daptomycin, clindamycin,
M AN U
trimethoprim-sulfamethoxazole, doxycycline, oxacillin, cefazolin, ceftaroline and tigecycline) used for treatment of gram-positive SSTIs. We then perform a cost analysis to compare the costs of standard care to treatment with dalbavancin and oritavancin and each other, given their novel dosing regimens. Our cost analysis is relevant for cases in which the empirical treatment of
METHODS
EP
Study design
TE D
choice is a drug that covers MRSA.
A systematic review and network meta-analysis of randomized controlled trials (RCTs) was
AC C
performed in accordance with the Preferred Reporting Items for Systematic Reviews and Metaanalyses (PRISMA) guidelines (17). Network meta-analysis (also known as mixed-treatment comparisons) allows the synthesis of direct and indirect evidence across a network of randomized trials to infer the relative effectiveness of multiple treatments or interventions. In this method, evidence on the relative performance of treatments A and B is obtained from trials that directly compare them (known as direct evidence) and also from trials that compare them 5
ACCEPTED MANUSCRIPT
through a common comparator (known as indirect evidence). By combining direct and indirect evidence, more precise estimates on relative performance can be obtained. Moreover, relative performance on treatments that have not been directly compared in trials can be estimated (18,
RI PT
19).
Literature review
SC
We searched MEDLINE (including articles “in process” and non-indexed citations) using the search strategy described in the Supplementary material (Table S1), the Cochrane Central
M AN U
Register of Controlled Trials and EMBASE using the keywords telavancin, dalbavancin and oritavancin. We searched clinicaltrials.gov for any relevant completed studies pending results. Searches were completed in May 2017. Screening of titles and abstracts was done by one author (RA) which was followed by full-text screening by two independent authors (RA and MP). We
TE D
included in our review all RCTs that: (1) included adults with cSSTIs and (2) compared one of the newer glycopeptide antibiotics to standard care for gram positive SSTIs. The primary outcome was the clinical response one to two weeks after end of therapy,
EP
broadly defined as an improvement in the signs and symptoms associated with the cSSTI. Secondary efficacy outcomes included the clinical response among patients infected with S.
AC C
aureus (both methicillin-sensitive and resistant), and the clinical response at 48-72 hours among patients with ABSSSI. We also examined the overall and serious adverse events reported in the RCTs.
We extracted data on patient population, details of the intervention and comparison, size of the study sample and our outcome measures. We evaluated the quality of included RCTs by 6
ACCEPTED MANUSCRIPT
examining the randomization and blinding procedures and the extent of attrition using the Cochrane risk of bias tool (20).
RI PT
Meta-analysis We performed two types of meta-analysis. First, we performed standard pairwise meta-analyses in RevMan version 5.3 using the Mantel-Haenszel procedure that assumes a fixed effect size
SC
(21). We measured between-study heterogeneity using the I2 statistic (22). We defined
substantial heterogeneity as an I2 statistic ≥ 50%, which would imply that real differences exist
M AN U
between study results that would not be explained by chance alone. We also evaluated the significance of between-study heterogeneity using the likelihood ratio test statistic comparing fixed effects and random-effects meta-analysis models. In the presence of substantial heterogeneity, meta-analysis using the random-effects model was considered (23). Second, we
TE D
performed a fixed-effects network meta-analysis using the “netmeta” package in R version 3.4.1. The estimates used in our meta-analyses were odds ratios (OR) and 95% confidence intervals (CI). Results from network meta-analyses were reported in the text, unless the results from the
reported.
EP
network and pairwise meta-analyses were inconsistent. In such cases, both estimates were
AC C
A fundamental assumption for network meta-analysis is transitivity. This means that one
can learn about treatment A versus B via treatment C (i.e., the common comparator C is transitive). An interpretation of transitivity is that the common comparator C (e.g., the standard care in our analysis) is similar when it appears in A versus C and B versus C trials (e.g., dalbavancin vs. standard care RCTs; and oritavancin vs. standard care RCTs). An equivalent way to assess transitivity is that studies which directly compare A with C and B with C would have 7
ACCEPTED MANUSCRIPT
similar distributions of effect modifiers (19). We assessed our transitivity assumption conceptually and if our network contained a closed loop, we assessed the consistency assumption (i.e., agreement between direct and indirect evidence) statistically by comparing direct and
RI PT
indirect summary effects (18).
Cost analysis
SC
Using TreeAge Pro 2016 (TreeAge Software Inc., Williamstown, MA), we developed a
computer simulation model to compare the costs associated with dalbavancin and oritavancin
M AN U
from the third-party payer perspective (i.e., direct costs). We focused on these two drugs (and not telavancin) based on our hypothesis that despite their high requisition costs, outpatient treatment of cSSTI would help offset the hospitalization costs associated with the use of standard treatment. Adult cSSTI patients enter the model and receive either one of the new drugs or one of
TE D
the commonly used vancomycin based regimens (vancomycin only or vancomycin followed by either linezolid, trimethoprim-sulfamethoxazole [TMP-SMX], doxycycline, or clindamycin). A patient’s drug treatment regimen determined the location of treatment, where newer antibiotics
EP
for cSSTI are given in an outpatient setting or freestanding infusion center, while standard antibiotics for cSSTI require initial hospitalization.
AC C
The dosing and treatment regimen for all drugs followed those described in the literature
(5, 24, 25). Patients treated with dalbavancin received 1 g over 30 minutes on the first day and 500 mg over 30 minutes 7 days later (i.e., two doses over the course of one week, each requiring an outpatient visit) (24). Patients treated with oritavancin received 1200 mg once over 3 hours (25). Those treated with only vancomycin received 15 mg/kg every 12 hours for 7 to 10 days. Patients treated with vancomycin followed by another drug received 15 mg/kg of vancomycin 8
ACCEPTED MANUSCRIPT
every 12 hours for 3 days, followed by linezolid (600 mg), TMP-SMX (two double strength 800 mg-160 mg tablets twice a day), doxycycline (100 mg twice daily), or clindamycin (300 mg four times daily) for the remainder of their 10 to 14-day treatment regimen. Patients receiving
RI PT
dalbavancin or oritavancin incurred the costs of the drug, evaluation and management (i.e.,
outpatient visit cost), infusion supplies, and infusion services for each hour of infusion and the
costs of the drugs as well as hospitalization.
SC
therapy duration, while those treated with any of the standard treatment approaches incurred the
M AN U
Table 1 shows the model input parameters, values, and sources. All parameters are agespecific where available (age distribution of patients based on SSTI discharges from the Healthcare Cost and Utilization Project [HCUP]) (26). We used mean hospitalization costs per patient for a primary discharge diagnosis of SSTI (ICD-9 code 682.9). All drug costs are based on average wholesale unit price for all available products, including generics and repackages
TE D
(27). All costs are presented in 2017 US dollars, converted using a 3% discount rate (28). Each simulation ran 1000 adult cSSTI patients through the model 1000 times for a total
EP
of 1 million trials with unique outcomes. Monte Carlo simulation (i.e., probabilistic sensitivity analysis) varied each parameter throughout the ranges in Table 1. As the economic value of
AC C
outpatient therapy may change with the cost of the new drug and/or changes in hospitalization cost, additional sensitivity analyses varied the cost of hospitalization for all ages (using values from $4,000 to $8,000, as well as the median value of $5,466), the cost of dalbavancin per 500 mg (0.5 to 2 times the current cost), and the cost of oritavancin per 400 mg (0.5 to 2 times the current cost). Lastly, we performed analyses to estimate the additional days of treatment avoided with use of dalbavancin and oritavancin. 9
ACCEPTED MANUSCRIPT
RESULTS Study selection
RI PT
Our searches yielded a total of 573 citations, of which 15 were retrieved for full-text review after title and abstract screening. Seven studies met inclusion criteria (24, 25, 29-33) (Figure 1).
Among the excluded studies was an early RCT comparing dalbavancin to standard care that did
SC
not have a defined treatment regimen in the comparison group (it was determined by the
M AN U
clinician prior to randomization) (34).
Study characteristics
Combined, the 7 studies included 6398 patients. Study characteristics are detailed in Table S2. All study participants had cSSTI caused by a suspected or confirmed gram-positive organism
TE D
and defined by the presence of a major abscess, infected burn, extensive cellulitis, wound infection or ulcer, along with systemic and local signs of infection. Of note, the definitions of cSSTI used by the included studies varied, but all incorporated the principles of the 2013 FDA
EP
definition (e.g., accompanying signs of systemic inflammation or significant medical comorbidities). Mean age in all studies was less than 50 years, and patients were predominantly
AC C
white and male. Cellulitis, abscess and wound infection were the most common infection types. The interventions used were IV telavancin in 3 studies (29-31), IV dalbavancin (administered as 1 g on day 1, followed by 500 mg on day 8) in 2 studies (24, 32), and IV oritavancin (given as a single 1200 mg dose) in 2 studies (25, 33). In all studies, standard therapy was administered and dosing adjustments were made according to standard practice by unblinded personnel. Other drugs (e.g., aztreonam and metronidazole) were permitted for polymicrobial infections. All 10
ACCEPTED MANUSCRIPT
studies were funded by the pharmaceutical company that manufactured the respective study drugs. One or more authors reported a conflict of interest in all studies.
RI PT
Study quality Appropriate performance of randomization and allocation concealment was reported in all
studies. All studies were described as double-blind, although details about participant blinding
SC
were not reported in one study (31). Outcome assessment was reported as blinded in all studies. Attrition was greater than acceptable limits in one study (32). Intention-to-treat analysis was
M AN U
performed in all studies. Study quality is depicted in Figure 2.
Outcomes
The network of eligible comparisons for our outcome ‘clinical response – all patients’ is depicted in Figure 3. The studies form a star shape network, where node size represents the treatment
TE D
option, and line thickness is proportional to the number of direct comparisons. Similar networks were noted for all other outcomes examined. As depicted in Figure 3, the network does not
EP
contain a closed loop (i.e., triangular loop). In such cases, the assumption of consistency cannot be evaluated statistically. However, the transitivity assumption can be assessed conceptually.
AC C
Although not identical, the populations, comparators, outcomes and key effect modifiers were clinically similar across our included studies, which supports the transitivity assumption. Results of the direct comparisons and network estimates for the clinical response and safety outcomes are shown in Table 2. For our primary analysis, we used data from intention-to-treat analyses reported in the studies when available. We also performed additional sensitivity analyses using only per-protocol data reported in the studies. Forest plots of the pairwise meta-analyses are 11
ACCEPTED MANUSCRIPT
included in Figures S1-S5 in the Supplementary material. Clinical response
RI PT
All included studies reported data on clinical response. Further, all studies except one (32) reported clinical response among patients infected with S. aureus and MRSA. Pairwise metaanalyses of the included RCTs showed that the clinical response to televancin, dalbavancin and
SC
oritavancin was similar to standard care. These results were consistent with those of the network meta-analyses. Telavancin, dalbavancin and oritavancin had similar clinical response in head-to-
M AN U
head comparisons in the network meta-analysis. Among patients with MRSA, telavancin was found to have a superior clinical response compared to standard care (OR 1.62; 95% CI 1.002.61) in pairwise and network meta-analysis. Dalbavancin and oritavancin were similar to standard care. Network meta-analysis suggested that oritavancin and dalbavancin were similar to
TE D
telavancin, and oritavancin was similar to dalbavancin in head-to-head comparisons. Three studies evaluated early clinical response at 48-72 hours among patients with ABSSSI (24, 25, 33). Network meta-analysis of these studies showed no differences between oritavancin and
EP
dalbavancin, oritavancin and standard care, or dalbavancin and standard care. Studies reported intention-to-treat and per-protocol data separately for the outcome of ‘clinical response – all
AC C
patients’. Pairwise and network meta-analyses using per-protocol data showed that the clinical response to televancin, dalbavancin and oritavancin was similar to standard care (Table S3 and Figure S3 in Supplementary material). Network meta-analysis showed that telavancin, dalbavancin and oritavancin had similar clinical response in head-to-head comparisons using per-protocol data. Adverse events 12
ACCEPTED MANUSCRIPT
All included studies reported data on overall and serious adverse events. Table S4 in the Supplementary material describes the serious adverse events that were deemed to be treatment related. Dalbavancin was found to have fewer overall adverse events compared to standard care
RI PT
(OR 0.77; 95% CI 0.64-0.93) in pairwise and network meta-analyses, but similar odds of serious adverse events. Telavancin had higher overall adverse events (OR 1.33; 95% CI 1.10-1.61) and serious adverse events (OR 1.54; 95% CI 1.07-2.21) compared to standard care in pairwise and
SC
network meta-analyses. Oritavancin had fewer overall adverse events compared to telavancin (OR 0.71; 95% CI 0.55-0.92) but the odds of serious adverse events were similar. Dalbavancin
M AN U
had fewer overall adverse events (OR 0.58; 95% CI 0.45-0.76) and serious adverse events (OR 0.50; 95% CI 0.29-0.86) compared to telavancin. There were no differences in the overall and serious adverse events between oritavancin and dalbavancin. Intention-to-treat and per-protocol adverse event data were not separately reported in the included studies.(23) Transient elevations
TE D
in liver enzymes were noted with dalbavancin and oritavancin (25, 32, 33), but these differences were not reported as significant when compared with the comparison group.
EP
Cost analysis
Table 3 shows the average cost of treatment for cSSTI patients for the evaluated treatment
AC C
regimens. With baseline assumptions (age-specific hospitalization costs and current drug costs), the use of each of the new drugs was less costly compared with vancomycin based regimens. The use of dalbavancin could save third-party payers $1,442 to $4,803 per cSSTI patient treated, while the use of oritavancin could save $3,571 to $6,932 per cSSTI patient treated. Compared to standard of care, dalbavancin could save between 6.5-10.0 days of treatment and oritavancin could save between 7.5-11.0 days of treatment. 13
ACCEPTED MANUSCRIPT
When all parameters were held at baseline, dalbavancin was the least costly treatment option up to a cost of $1,132, at which point oritavancin became the least costly treatment option. At dalbavancin costs ≥ $2,600, all other treatment strategies cost less except for the
RI PT
strategy of intravenous vancomycin followed by oral linezolid for completion of the antibiotic course. When all parameters were held at baseline, oritavancin was the least costly treatment option up to a cost of $1,905, at which point dalbavancin became the least costly option.
SC
Oritavancin remained less costly than vancomycin based treatments at twice its current cost
M AN U
($2,390 per 400 mg).
When hospitalization costs were decreased to $4,000 per patient, treatment with dalbavancin was less costly than treatment with vancomycin followed by linezolid, but was more expensive than the other vancomycin based regimens. In probabilistic sensitivity analysis, 100% of iterations favored treatment with oritavancin (i.e., it was the least costly option) for all
TE D
scenarios explored except when hospitalization cost $4,000 and oritavancin cost $2,390 per 400 mg (twice the current cost), at which point 100% of iterations favored treatment with
DISCUSSION
EP
vancomycin followed by doxycycline.
AC C
Our systematic review and meta-analysis suggests that newer glycopeptide antibiotics are similar to standard therapy for treatment of cSSTIs caused by a suspected or confirmed gram positive organism. This finding was also true for cSSTIs caused specifically by MRSA with the exception of telavancin, which demonstrated increased odds of clinical response compared to standard care. Head-to-head comparisons suggested no differences in clinical response between the new agents. There was increased odds of adverse events with telavancin. Compared with standard treatment, 14
ACCEPTED MANUSCRIPT
dalbavancin and oritavancin were less costly compared to standard of care under baseline assumptions and many of the scenarios evaluated. Even reducing the total cost of hospitalization for cSSTI patients to $4,000 was not enough to overcome the potential savings of outpatient
RI PT
infusion therapy at oritavancin’s current price point. Importantly, our cost analysis is relevant for cases in which the empirical treatment of choice is a drug that covers MRSA.
In the studies that evaluated telavancin, it was administered intravenously for 4-14 days
SC
in two studies (29, 30) and 7-14 days in one study (31). We found that the clinical response to
M AN U
telavancin was superior compared to standard care for patients infected with MRSA. However, the duration of treatment with telavancin is similar to conventional agents. Further, we found increased odds of adverse events with telavancin, which was supported by a past meta-analysis that compared telavancin with vancomycin among patients with cSSTI and hospital acquired pneumonia, and demonstrated an increased risk of serious adverse events and acute renal failure
TE D
among telavancin recipients (35). In light of this, telavancin may have little net advantage over vancomycin for the treatment of cSSTIs, especially in the presence of other vancomycin alternatives, such as linezolid. A recent large meta-analysis demonstrated that linezolid was
EP
associated with a significantly higher clinical and microbiological cure rate as well as a reduction
AC C
in the length of hospital stay and cost of hospital charges compared with vancomycin among patients with SSTI (36).
The RCTs that evaluated oritavancin and dalbavancin demonstrated that patients with
complicated SSTIs can be treated successfully with one or two doses of these agents without an increase in adverse events. Further, a recent RCT demonstrated that a single 1500 mg IV dose of dalbavanvin was non-inferior to the two-dose regimen (37) and this dosing has now been 15
ACCEPTED MANUSCRIPT
approved by the FDA. These agents represent an advance in the treatment of these infections by eliminating the need for inpatient or prolonged outpatient parenteral therapy, thus potentially decreasing resource utilization, improving compliance and preventing potential complications
RI PT
from peripherally inserted central catheters. Further, no monitoring for blood concentrations is required in contrast to vancomycin (13, 15, 38). However, use of these agents may be limited by their high acquisition costs (39), as well as unexpected adverse events. For example, the SOLO II
SC
trial reported that osteomyelitis occurred in 5 patients on oritavancin and none in the vancomycin
M AN U
arm. This finding, along with data from earlier studies, led to a warning issued by the FDA (40). Our review is strengthened by the fact that all included studies were described as doubleblind and had low risk of bias in our quality evaluation using established criteria. A key limitation of the review is that most of the studies were published prior to the FDA guidance on designing RCTs to evaluate drugs for cSSTIs (6), with the result that the definition of cSSTI
TE D
varied between studies. The comparator regimens were also different as no formal guidance currently exists on empiric therapy and duration of treatment for cSSTIs.
EP
Although our cost analysis focuses on the cost of cSSTI treatments in the United States, the direction of our results is likely generalizable to other countries, particularly those with
AC C
similar standards of care. In addition, we examined non-monetary outcomes to compare the “value” of each approach, such as number of treatment days avoided, which may be less sensitive to differences in direct drug costs between countries. Another potential limitation of the evidence-base is that all RCTs were sponsored by the
pharmaceutical companies that produced the antibiotics. In addition, our findings are limited to cases of cSSTI where a gram-positive organism is suspected or confirmed. Future research 16
ACCEPTED MANUSCRIPT
should test the comparative effectiveness of these newer drugs in patients with cSSTI that meet the criteria defined by the FDA.
RI PT
CONCLUSION Dalbavancin and oritavancin demonstrate efficacy and safety comparable to vancomycin alone or followed by linezolid when used to treat cSSTI where a gram-positive organism is suspected
SC
or confirmed. Oritavancin yielded greater cost savings compared to dalbavancin. Treatment with these new antibiotics consists of one or two doses and offers cost savings over standard therapies
M AN U
for cSSTI when standard care is treatment that covers MRSA. These drugs are potential options for the outpatient management of complicated SSTIs, defined as SSTIs with systemic signs of infection or significant medical comorbidities.
Funding: none
TE D
Acknowledgements: none
Transparency declaration: The authors have no conflicts of interest and no sources of industry
AC C
EP
support to declare
17
ACCEPTED MANUSCRIPT
REFERENCES 1.
Jenkins TC, Sabel AL, Sarcone EE, Price CS, Mehler PS, Burman WJ. Skin and soft-
antimicrobial stewardship. Clin Infect Dis. 2010;51(8):895-903. 2.
RI PT
tissue infections requiring hospitalization at an academic medical center: opportunities for
Lee BY, Singh A, David MZ, Bartsch SM, Slayton RB, Huang SS, et al. The economic
SC
burden of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA). Clin Microbiol Infect. 2013;19(6):528-36.
Miller LG, Eisenberg DF, Liu H, Chang CL, Wang Y, Luthra R, et al. Incidence of skin
M AN U
3.
and soft tissue infections in ambulatory and inpatient settings, 2005-2010. BMC Infect Dis. 2015;15:362. 4.
Russo A, Concia E, Cristini F, De Rosa FG, Esposito S, Menichetti F, et al. Current and
future trends in antibiotic therapy of acute bacterial skin and skin-structure infections. Clin
5.
TE D
Microbiol Infect. 2016;22 Suppl 2:S27-36.
Stevens DL, Bisno AL, Chambers HF, Dellinger EP, Goldstein EJ, Gorbach SL, et al.
Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014
US Food and Drug Administration CfDEaR. Guidance for industry. Acute bacterial skin
AC C
6.
EP
update by the Infectious Diseases Society of America. Clin Infect Dis. 2014;59(2):e10-52.
and skin-structure infections: developing drugs for treatment 2010 [Available from: http://www.fda.gov/downloads/Drugs/.../Guidances/ucm071185.pdf. 7.
Miller LG, Daum RS, Creech CB, Young D, Downing MD, Eells SJ, et al. Clindamycin
versus trimethoprim-sulfamethoxazole for uncomplicated skin infections. N Engl J Med. 2015;372(12):1093-103. 18
ACCEPTED MANUSCRIPT
8.
Chambers HF. The changing epidemiology of Staphylococcus aureus? Emerg Infect Dis.
2001;7(2):178-82. 9.
Daum RS. Clinical practice. Skin and soft-tissue infections caused by methicillin-
10.
Cosgrove SE, Carroll KC, Perl TM. Staphylococcus aureus with reduced susceptibility to
vancomycin. Clin Infect Dis. 2004;39(4):539-45.
Tsoulas C, Nathwani D. Review of meta-analyses of vancomycin compared with new
SC
11.
RI PT
resistant Staphylococcus aureus. N Engl J Med. 2007;357(4):380-90.
Antimicrob Agents. 2015;46(1):1-7. 12.
M AN U
treatments for Gram-positive skin and soft-tissue infections: Are we any clearer? Int J
Cardona AF, Wilson SE. Skin and soft-tissue infections: a critical review and the role of
telavancin in their treatment. Clin Infect Dis. 2015;61 Suppl 2:S69-78. 13.
Klinker KP, Borgert SJ. Beyond Vancomycin: The Tail of the Lipoglycopeptides. Clin
14.
TE D
Ther. 2015;37(12):2619-36.
Lin G, Credito K, Ednie LM, Appelbaum PC. Antistaphylococcal activity of dalbavancin,
an experimental glycopeptide. Antimicrob Agents Chemother. 2005;49(2):770-2. Saravolatz LD, Stein GE. Oritavancin: A Long-Half-Life Lipoglycopeptide. Clin Infect
EP
15.
Dis. 2015;61(4):627-32.
Zhanel GG, Schweizer F, Karlowsky JA. Oritavancin: mechanism of action. Clin Infect
AC C
16.
Dis. 2012;54 Suppl 3:S214-9. 17.
Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The
PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339:b2700. 18.
Cipriani A, Higgins JP, Geddes JR, Salanti G. Conceptual and technical challenges in 19
ACCEPTED MANUSCRIPT
network meta-analysis. Ann Intern Med. 2013;159(2):130-7. 19.
Salanti G. Indirect and mixed-treatment comparison, network, or multiple-treatments
synthesis tool. Res Synth Methods. 2012;3(2):80-97. 20.
Collaboration TC. Cochrane handbook for systematic reviews of interventions (Version
5.1.0) March 2011 [Available from: http://cochrane.handbook.org.
Collins R, Gray R, Godwin J, Peto R. Avoidance of large biases and large random errors
SC
21.
RI PT
meta-analysis: many names, many benefits, many concerns for the next generation evidence
1987;6(3):245-54. 22.
Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-
analyses. BMJ. 2003;327(7414):557-60. 23.
M AN U
in the assessment of moderate treatment effects: the need for systematic overviews. Stat Med.
DerSimonian R, Kacker R. Random-effects model for meta-analysis of clinical trials: an
24.
TE D
update. Contemp Clin Trials. 2007;28(2):105-14.
Boucher HW, Wilcox M, Talbot GH, Puttagunta S, Das AF, Dunne MW. Once-weekly
dalbavancin versus daily conventional therapy for skin infection. N Engl J Med.
25.
EP
2014;370(23):2169-79.
Corey GR, Good S, Jiang H, Moeck G, Wikler M, Green S, et al. Single-dose oritavancin
AC C
versus 7-10 days of vancomycin in the treatment of gram-positive acute bacterial skin and skin structure infections: the SOLO II noninferiority study. Clin Infect Dis. 2015;60(2):254-62. 26.
United States Department of Health & Human Services. HCUP facts and figures:
statistics on hospital-based care in the United States Rockville, MD: AHRQ: Agency for Healthcare Research and Quality; 2013 [Available from: http://hcupnet.ahrq.gov/HCUPnet.jsp. 27.
Red Book Online [Internet]. Truven Health Analytics. 2016 [cited October 24, 2016]. 20
ACCEPTED MANUSCRIPT
28.
M.R. Gold JES, L.B. Russell, and M.C. Weinstein. New York: Oxford University Press;
1996. 29.
Stryjewski ME, O'Riordan WD, Lau WK, Pien FD, Dunbar LM, Vallee M, et al.
due to gram-positive bacteria. Clin Infect Dis. 2005;40(11):1601-7. 30.
RI PT
Telavancin versus standard therapy for treatment of complicated skin and soft-tissue infections
Stryjewski ME, Chu VH, O'Riordan WD, Warren BL, Dunbar LM, Young DM, et al.
SC
Telavancin versus standard therapy for treatment of complicated skin and skin structure
infections caused by gram-positive bacteria: FAST 2 study. Antimicrob Agents Chemother.
31.
M AN U
2006;50(3):862-7.
Stryjewski ME, Graham DR, Wilson SE, O'Riordan W, Young D, Lentnek A, et al.
Telavancin versus vancomycin for the treatment of complicated skin and skin-structure infections caused by gram-positive organisms. Clin Infect Dis. 2008;46(11):1683-93. Jauregui LE, Babazadeh S, Seltzer E, Goldberg L, Krievins D, Frederick M, et al.
TE D
32.
Randomized, double-blind comparison of once-weekly dalbavancin versus twice-daily linezolid therapy for the treatment of complicated skin and skin structure infections. Clin Infect Dis.
33.
EP
2005;41(10):1407-15.
Corey GR, Kabler H, Mehra P, Gupta S, Overcash JS, Porwal A, et al. Single-dose
90. 34.
AC C
oritavancin in the treatment of acute bacterial skin infections. N Engl J Med. 2014;370(23):2180-
Seltzer E, Dorr MB, Goldstein BP, Perry M, Dowell JA, Henkel T, et al. Once-weekly
dalbavancin versus standard-of-care antimicrobial regimens for treatment of skin and soft-tissue infections. Clin Infect Dis. 2003;37(10):1298-303. 35.
Polyzos KA, Mavros MN, Vardakas KZ, Makris MC, Rafailidis PI, Falagas ME. Efficacy 21
ACCEPTED MANUSCRIPT
and safety of telavancin in clinical trials: a systematic review and meta-analysis. PLoS One. 2012;7(8):e41870. 36.
Yue J, Dong BR, Yang M, Chen X, Wu T, Liu GJ. Linezolid versus vancomycin for skin
37.
RI PT
and soft tissue infections. Cochrane Database Syst Rev. 2016(1):CD008056.
Dunne MW, Puttagunta S, Giordano P, Krievins D, Zelasky M, Baldassarre J. A
Randomized Clinical Trial of Single-Dose Versus Weekly Dalbavancin for Treatment of Acute
38.
SC
Bacterial Skin and Skin Structure Infection. Clin Infect Dis. 2016;62(5):545-51.
Nathwani D, Eckmann C, Lawson W, Stephens JM, Macahilig C, Solem CT, et al. Pan-
M AN U
European early switch/early discharge opportunities exist for hospitalized patients with methicillin-resistant Staphylococcus aureus complicated skin and soft tissue infections. Clin Microbiol Infect. 2014;20(10):993-1000.
Drugs for MRSA skin and soft-tissue infections. JAMA. 2014;312(15):1583-4.
40.
Oritavancin: Highlights of prescribing information 2014 [Available from:
TE D
39.
https://www.accessdata.fda.gov/drugsatfda_docs/label/2014/206334s000lbl.pdf. 41.
Centers for Medicare & Medicaid Services. Physician Fee Schedule Baltimore, MD:
EP
Centers for Medicare & Medicaid; 2016 [updated October 18. Available from: https://www.cms.gov/apps/physician-fee-schedule/overview.aspx. Centers for medicare & Medicaid Services. Durable Medical Equipment, Prosthetics,
AC C
42.
Orthotics, and Supplies (DMEPOS) Fee Schedule Baltimore, MD: Centers for medicare & Medicaid 2016 [updated January. Available from: https://www.cms.gov/medicare/medicare-feefor-service-payment/DMEPOSFeeSched/DMEPOS-Fee-Schedule.html. 43.
U.S. Department of Health and Human Services, Centers for Disease Control and
Prevention, National Center for Health Statistics. Anthropometric Reference Data for Children 22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
and Adults: United States, 2011-2014. 2016.
23
ACCEPTED MANUSCRIPT Tables Table 1. Model input parameters, values, and sources
Table 2: Results from pairwise and network meta-analysis
Figure Legends: Figure 1: Selection process for included studies Figure 2: Cochrane risk of bias table
AC C
EP
TE D
M AN U
SC
Figure 3: Network of treatment comparisons for primary outcome
RI PT
Table 3: Average total cost in dollars (95% uncertainty interval) of treatment regimens
24
ACCEPTED MANUSCRIPT Table 1. Model input parameters, values, and sources Parameter
Mean or Median
Standard Deviation or Range
Source
1,194.8 1,841.64 0.015 190.05 3.16
0.008 23.69 0.85
(27) (27) (27) (27) (27)
Costs (2017 $US)
RI PT
Oritavancin (per 400mg) Dalbavancin (per 500mg) Vancomycin (per mg) Linezolid (per 600mg) Trimethoprim-sulfamethoxazole (per 800mg-160mg) Doxycycline (per 100mg)
3.31
1.85
(27)
Clindamycin (per 300mg)
4.45
0.54
(27)
5,318.03 8,113.79 7,973.10 8,657.41 71.91 19.55 111.37 45.07
565.01 818.24 956.68 970.19 0.93
(26) (26) (26) (26) (41) (41) (41) (42)
3 0.5 -
7 – 10 10 – 14
(25) (24) (25) (24)
-
10-14
(5)
21.3 34.7 36.6 7.5
-
(26) (26) (26) (26)
Hospitalization
¶
M AN U
≥ 85 years First hour of infusion Additional hours of infusion Office visit for infusion (evaluation and management)
SC
18 to 44 years 45 to 64 years 65 to 84 years
TE D
Infusion supplies Duration Oritavancin (hours) Dalbavancin (hours) Vancomycin only (days) Vancomycin followed by linezolid (days)* Vancomycin followed by trimethoprim-sulfamethoxazole, doxycycline or clindamycin (days)* Patient Characteristics Age distribution of patients (%) 18 to 44 years 45 to 64 years 65 to 84 years ≥ 85 years
AC C
EP
Average Weight (kg) 18 to 44 years 82.8 73.5 – 91.5 (43) 45 to 64 years 84.5 76.6 – 91.5 (43) 65 to 84 years 78.6 64.4 – 90.6 (43) ≥ 85 years 71.8 (43) ¶ Mean and standard deviation of hospitalization costs for patients with a discharge diagnosis of skin and soft-tissue infection (ICD-9 code 682.9) *Total treatment duration represented three days of intravenous vancomycin followed by the oral antibiotic to complete therapy
ACCEPTED MANUSCRIPT
Table 2: Results from pairwise and network meta-analysis Treatment comparison
Number of studies
Pairwise meta-analysis Odds ratio (95% CI)
AC C
EP
TE D
M AN U
SC
RI PT
Clinical response – all patients Oritavancin vs. Standard care 2 1.06 (0.85-1.33) Dalbavancin vs. Standard care 2 0.79 (0.53-1.20) Telavancin vs. Standard care 3 1.09 (0.90-1.33) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Clinical response – patients infected with methicillin-resistant Staphylococcus aureus Oritavancin vs. Standard care 2 0.95 (0.56-1.61) Dalbavancin vs. Standard care 1 0.73 (0.06-8.33) Telavancin vs. Standard care 3 1.62 (1.01-2.61) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Early clinical response at 48-72 hours among patients with acute bacterial skin and skin-structure infection (ABSSSI) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Standard care 2 1.02 (0.80-1.27) Dalbavancin vs. Standard care 1 0.99 (0.76-1.30) Overall adverse events Oritavancin vs. Standard care 2 0.94 (0.79-1.13) Dalbavancin vs. Standard care 2 0.77 (0.64-0.93) Telavancin vs. Standard care 3 1.33 (1.10-1.60) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Serious adverse events Oritavancin vs. Standard care 2 0.99 (0.68-1.45) Dalbavancin vs. Standard care 2 0.77 (0.52-1.14) Telavancin vs. Standard care 3 1.52 (1.07-2.17) Oritavancin vs. Dalbavancin NA NA Oritavancin vs. Telavancin NA NA Dalbavancin vs. Telavancin NA NA Abbreviations: CI – confidence interval; NA – not available
Network meta-analysis Odds ratio (95% CI) 1.06 (0.85-1.33) 0.78 (0.52-1.18) 1.09 (0.90-1.33) 1.36 (0.85-2.18) 0.98 (0.72-1.31) 0.72 (0.45-1.13) 0.95 (0.56-1.61) 0.73 (0.06-8.33) 1.62 (1.00-2.61) 1.29 (0.11-15.48) 0.59 (0.29-1.19) 0.45 (0.04-5.39) 1.02 (0.72-1.46) 1.02 (0.80-1.27) 0.99 (0.76-1.30) 0.94 (0.79-1.13) 0.77 (0.64-0.93) 1.33 (1.10-1.61) 1.23 (0.94-1.58) 0.71 (0.55-0.92) 0.58 (0.45-0.76) 0.99 (0.68-1.45) 0.77 (0.52-1.15) 1.54 (1.07-2.21) 1.28 (0.74-2.22) 0.64 (0.38-1.09) 0.50 (0.29-0.86)
ACCEPTED MANUSCRIPT
Table 3: Average total cost in dollars (95% uncertainty interval) of treatment regimens Total Cost for Treatment in Outpatient Settings
Total Cost for Treatment Requiring Hospitalization Median ($5,466) SSTI Hospitalization Cost
$8,000 SSTI Hospitalization Cost
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
7,569 (4,425 – 8,958)
5,782 (5,744 – 5,806)
4,318 (4,283 – 4,339)
10,784 (7,626 – 12,200)
8,998 (8,955 – 9,041)
7,534 (7,491 – 7,577)
8,316 (8,279 – 8,339) 11,532 (11,490 – 11,580)
7,478 (4,329 – 8,890)
5,692 (5,678 – 5,700)
4,226 (4,213 – 4,234)
8,225 (8,212 – 8,234)
7,524 (4,376 – 8,935)
5,738 (5,724 – 5,747)
4,272 (4,260 – 4,281)
8,272 (8,259 – 8,280)
5,637 (5,624 – 5,646)
4,172 (4,159 – 4,180)
8,171 (8,158 – 8,179)
RI PT
Age-Specific SSTI Hospitalization Cost -
SC
-
M AN U
-
7,423 (4,274 – 8,834)
AC C
EP
Dalbavancin ($1,842 per 500 5,982 (5,982 – 5,982) mg)* Dalbavancin ($921 per 500 mg)^ 3,219 (3,219 – 3,219) Dalbavancin ($3,684 per 500 11,507 (11,507 – 11,507) mg)† Oritavancin ($1,195 per 400 mg)* 3,852 (3,852 – 3,852) Oritavancin ($598 per 400 mg)^ 2,060 (2,060 – 2,060) Oritavancin ($2,390 per 400 mg) 7,436 (7,436 – 7,436) † Vancomycin Vancomycin followed by linezolid Vancomycin followed by trimethoprim-sulfamethoxazole Vancomycin followed by clindamycin Vancomycin followed by doxycycline *Current cost ^Half the current cost †Two times the current cost Abbreviations: SSTI – skin and soft tissue infection
$4,000 SSTI Hospitalization Cost
TE D
Treatment Regimen
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT