Eccmid poster

DALBAVANCIN VS LINEZOLID FOR TREATMENT OF ACUTE BACTERIAL INFECTIONS OF THE SKIN:
A COMPARISON OF EARLY AND STANDARD OUTCOME MEASURES IN STUDY VER001-9
Michael W. Dunne,1 George H. Talbot,2 Anita F. Das3
1Durata Therapeutics, Morristown, NJ, USA; 2Talbot Advisors, St. Davids, PA, USA; 3Axistat Inc, San Francisco, CA, USA ABSTRACT
Table 3. ABSSSI Measurements at Baseline
Table 4. Systemic Signs of ABSSSI at Baseline
Table 9. Mean Change from Baseline in Lesion Area by Infection Type
Dalbavancin (N = 135)
Linezolid (N = 85)
Dalbavancin (N = 135)
Linezolid (N = 85)
All Patients (N = 220)
Objectives: FDA Draft Guidance for treatment of skin infections has
Table 1. Demographics and Baseline Characteristics*
Length (cm)
Temperature (
Cellulitis
recommended the implementation of an outcome assessment at 48–72 hours post baseline of the cessation of spread plus resolution of elevated temperatures Dalbavancin (N = 135)
Linezolid (N = 85)
as the primary point for comparisons in noninferiority studies, rather than the test of cure historically measured post therapy. We performed a retrospective Age (years)
analysis of this new endpoint in a previously completed registrational trial and compared the outcome to the protocol prespecified primary endpoint of clinical Temperature (≥38
Methods: The primary endpoint at Day 28 was originally defined in the clinically
˚C), n (%)
evaluable population which was then further analyzed in subgroups of patients Width (cm)
WBC Count (cells/mm3)
Major Abscess
meeting the newly defined FDA inclusion criteria (surface area of lesion >75cm2; one sign of either fever, elevated white blood cell count or bandemia) as well as the early response endpoint at Day 3/4 (lesion size the same or smaller relative to baseline and temperature <37.6°C).
Results: The difference in response rates between the dalbavancin and linezolid
treated patients was similar in all analyses of early response based on lesion size and temperature compared to the historical assessment of clinical response at WBC Count >12,000 cells/mm3, n (%)
the end of therapy. The early response point estimates that include temperature Area (cm2)
Bands (%)
Wound Infection
assessments were lower than the overall clinical response at end of therapy. The Race/Ethnicity
difference in cure rates between treatment regimens as assessed by the early response endpoint did not further differentiate between those regimens relative to the cure rates determined by clinical response at the end of therapy.
Conclusions: Dalbavancin noninferiority relative to linezolid as assessed by the
prespecified primary analysis is reinforced with an early responder analysis performed at Day 3/4, with very similar differences in the point estimate ≥10%, n (%)
Other Deep Soft Tissue Infection
Table 5. FDA Primary Outcome Measure: Cessation of Skin Lesion Spread and Afebrile
BMI (kg/m2)
OBJECTIVES
Timepoint
Analysis Population
Endpoint
Dalbavancin, n/N (%)
Linezolid, n/N (%)
Difference, 95% CI
Perform a retrospective analysis of the performance of the new FDA Draft Guidance early responder endpoint in a previously completed registrational trial of complicated skin and skin structure infections (SSSI) (new FDA terminology: ■ Compare the outcome of that early responder analysis to the protocol’s pre- specified primary endpoint of clinical response at the Test-of-Cure (TOC) visit at Table 10. Early Response at Day 3/4 vs. Clinical Response at End of Therapy
* FDA-defined population, i.e. with lesion >75 cm2 and one systemic sign ■ Compare the early responder analysis results to FDA’s proposed assessment of All Treated (N=220)
Clinical Response at End of Treatment (EOT) Clinical Success at EOT (N1= 187)
Clinical Failure at EOT (N1= 33)
Table 2. Description and Location of the ABSSSI
Early Clinical Response
Dalbavancin (N = 135)
Linezolid (N = 85)
Table 6. Reasons for Non-Response FDA Primary Efficacy Outcome Measure
Table 7. Percent Change from Baseline in Lesion Area for Responders
Study Design and Treatments:
Type of Infection
■ Randomized, double-blind study conducted in 7 countries from January 2003 Dalbavancin (N = 135)
Linezolid (N = 85)
Dalbavancin (N = 135)
Linezolid (N = 85)
■ Patients randomized 2:1 to receive either of the following: – Dalbavancin, 1000-mg on day 1 followed by 500-mg on day 8, with a possible Nonresponders, N1
Responders, N1
CONCLUSIONS
– Linezolid 600 mg intravenously every 12 h, with a possible switch to orally Primary Endpoint:
■ Dalbavancin noninferiority relative to linezolid as assessed by the prespecified primary analysis is reinforced ■ Protocol-specified primary endpoint: Clinical success in the clinically evaluable with an early responder analysis performed at Day 3/4, with very similar differences in the point estimate (CE) population at Day 28, the TOC visit (i.e., signs and symptoms of SSSI improved so no further antibacterial therapy warranted) Anatomical Site of Infection
■ The addition of resolution of fever to the early response definition did not further differentiate between – Clinically evaluable patients received at least 72 h of treatment with the blinded treatment regimens beyond lesion measurement alone.
study medication, did not have an indeterminate clinical response, and fulfilled ■ Most patients who were an early responder were also a clinical success at EOT (PPV of response at all other protocol requirements relating to concurrent use of antibacterials, Day 3/4 = 90.6%). However, most patients who were an early nonresponder became a clinical success at surgical intervention of the SSSI, and efficacy-related inclusion/exclusion criteria EOT (NPV of non-response at Day 3/4 = 34%).
– For this reanalysis, the primary endpoint in the CE population was further ■ In our analysis, the early response endpoint had limitations, specifically a low NPV.
analyzed in subgroups of patients meeting the newly defined FDA inclusion ■ A validated measure of response at EOT and TOC that is acceptable to FDA and other regulatory agencies criteria, specifically: surface area of lesion >75 cm2 or > 50 cm2 if on face AND Table 8. Percent Change from Baseline in Lesion Area for Responders by Infection Type
one sign of either fever, elevated white blood cell count or bandemia would be more intuitive to clinicians and more relevant to patients.
■ FDA early response endpoint: Success at Day 3 or 4, defined as an ABSSSI Cellulitis (N = 81)
Major Abscess (N = 78)
Wound Infection (N = 34)
Other Infection Type (N = 27)
lesion size the same or smaller than at baseline (i.e., “cessation of lesion spread”) Responders, N1
– The analysis population was the intent-to-treat (ITT) population excluding REFERENCES
patients with missing data for the skin lesion size – Failures included death or receipt of another potentially effective antibiotic 1. Jauregui LE, Babazadeh S, Seltzer E, et al. Randomized, Double-Blind Comparison of Once-Weekly Dalbavancin versus Twice-Daily – For this re-analysis, patients with missing data for skin lesion size or Linezolid Therapy for the Treatment of Complicated Skin and Skin Structure Infections. Clin Infect Dis. 2005;41:1407–15.
temperature were removed from the analysis 2. FDA Draft Guidance for Industry. Acute Bacterial Skin and Skin Structure Infections: Developing Drugs for Treatment. August 2010.
■ Data are shown for the population defined by the new FDA inclusion criteria ACKNOWLEDGEMENTS
■ The Positive Predictive Value (PPV) and Negative Predictive Value (NPV) for an ● This analysis was sponsored by Durata Therapeutics, Inc.; editorial support was funded by Durata Therapeutics, Inc.
early response or non-response was calculated using the EOT response in the ● Dr. Dunne is an employee of Durata Therapeutics, Inc.; Drs. Talbot and Das are consultants to the company.
● Drs. Dunne, Talbot and Das are members of the Foundation for NIH Project Team on New Endpoints for ABSSSI.

Source: http://a.eqcdn.com/duratatherapeutics/files/Posters/Final-ECCMID-2011-42611.pdf?1353358529