1095P - The comparative efficacy of first-line (1L) treatments for stage IIIC and stage IV melanoma: Results of a systematic review and network meta-analysis

Date 28 September 2014
Event ESMO 2014
Session Poster Display session
Topics Anti-Cancer Agents & Biologic Therapy
Cancer Immunology and Immunotherapy
Melanoma and other Skin Tumours
Presenter Christin Bexelius
Citation Annals of Oncology (2014) 25 (suppl_4): iv374-iv393. 10.1093/annonc/mdu344
Authors C. Bexelius1, J.M. Quigley2, P. Thuresson3, N. Hawkins2
  • 1Pharmaceuticals Division, F. Hoffmann-La Roche, 4070 - Basel/CH
  • 2Health Economics, ICON Health Economics, OX2 0JJ - Oxford/GB
  • 3Global Pricing And Market Access, F. Hoffmann-La Roche, Basel/CH

Abstract

Aim

Network meta-analysis (NMA) is used to compare treatment effects, taking all available evidence into account. This study describes a NMA of 1L treatments for stages IIIC and IV melanoma using the accelerated failure time (AFT) scale. The comparators were selected based on European clinical practice and included: dacarbazine (DTIC), DTIC + tamoxifen (TMX), temozolomide (TMZ), interferon (IFN), ipilimumab (IPI), vemurafenib (VM), dabrafenib (DB) and trametinib (TR). Reported outcomes were overall survival (OS) and progression-free survival (PFS).

Methods

Studies in the systematic literature were identified through the Embase, Medline, and Cochrane databases. Kaplan-Meier curves were scanned, and individual patient level data were recreated using a published algorithm. Treatment effect estimates for each study were obtained using Cox proportional hazards (PH) and the AFT models. For the AFT, a generalized gamma distribution was assumed. Bayesian fixed-effects NMAs were conducted and assumed that treatment effects were comparable across trials.

Results

Fourteen studies were identified and included in the analysis. Diagnostic tests results indicated that treatment effects were not constant over time in the PH model, and the AFT model better fit the trial data and, therefore, was used in the analysis. Studies using lower dose intensities of DTIC (800 mg/m2/d) were excluded due to heterogeneity. The inverse acceleration factors for OS and PFS for each treatment, compared with DTIC, are shown in the table (values lower than 1 indicate increased survival).

Conclusions

VM is the only treatment that shows a statistically significant increase in both OS and PFS, compared with DTIC, when published RCT data are compared.

Disclosure

C. Bexelius: I am an employee of F.Hoffmann-La Roche; J.M. Quigley: Consultant for Roche with compensation for conducting systematic review; P. Thuresson: I am an employee of F. Hoffmann-La Roche; N. Hawkins: I am employed by ICON PLC, a consultant to F. Hoffmann-La Roche.

Table: 1095P

Inverse Acceleration Factor Compared With DTIC. Median (95% confidence interval)

DB DB + TR 1mg DB + TR 2 mg DTIC + IFN DTIC + IFN + TMX DTIC + TMX IPI IPI + DTIC TMZ VM
OS 0.86 (0.62,1.2) 0.87 (0.53,1.43) 0.7 (0.43,1.16) 1.13 (0.95,1.33) 3.35 (2.23,5.02) 3.53 (2.36,5.3) 1.05 (0.65,1.69) 0.77 (0.63,0.96) 0.97 (0.86,1.1) 0.74 (0.61,0.89)
PFS 0.37 (0.29,0.46) 0.27 (0.19,0.38) 0.21 (0.15,0.29) 1.11 (0.89,1.38) 1.22 (0.85,1.75) 1.15 (0.8,1.65) 0.84 (0.7,1.02) - 0.91 (0.8,1.03) 0.32 (0.28,0.37)