Keith A. Betts

Matching-adjusted indirect comparisons: a new tool for timely comparative effectiveness research.

OBJECTIVE In the absence of head-to-head randomized trials, indirect comparisons of treatments across separate trials can be performed. However, these analyses may be biased by cross-trial differences in patient populations, sensitivity to modeling assumptions, and differences in the definitions of outcome measures. The objective of this study was to demonstrate how incorporating individual patient data (IPD) from trials of one treatment into indirect comparisons can address several limitations that arise in analyses based only on aggregate data. METHODS Matching-adjusted indirect comparisons (MAICs) use IPD from trials of one treatment to match baseline summary statistics reported from trials of another treatment. After matching, by using an approach similar to propensity score weighting, treatment outcomes are compared across balanced trial populations. This method is illustrated by reviewing published MAICs in different therapeutic areas. A novel analysis in attention deficit/hyperactivity disorder further demonstrates the applicability of the method. The strengths and limitations of MAICs are discussed in comparison to those of indirect comparisons that use only published aggregate data. RESULTS Example applications were selected to illustrate how indirect comparisons based only on aggregate data can be limited by cross-trial differences in patient populations, differences in the definitions of outcome measures, and sensitivity to modeling assumptions. The use of IPD and MAIC is shown to address these limitations in the selected examples by reducing or removing the observed cross-trial differences. An important assumption of MAIC, as in any comparison of nonrandomized treatment groups, is that there are no unobserved cross-trial differences that could confound the comparison of outcomes. CONCLUSIONS Indirect treatment comparisons can be limited by cross-trial differences. By combining IPD with published aggregate data, MAIC can reduce observed cross-trial differences and provide decision makers with timely comparative evidence.

Comparative efficacy of nilotinib and dasatinib in newly diagnosed chronic myeloid leukemia: a matching-adjusted indirect comparison of randomized trials.

Abstract Objective: Nilotinib and dasatinib have not been directly compared in a randomized trial for the treatment of newly diagnosed chronic myeloid leukemia in the chronic phase (CML-CP). The purpose of this study was to indirectly compare rates of major molecular response (MMR), progression-free survival (PFS) and overall survival by month 12 with nilotinib and dasatinib treatment of newly diagnosed CML-CP. Methods: Individual patient data from a randomized trial of nilotinib vs. imatinib (ENESTnd) and published summary data from a separate randomized trial of dasatinib vs. imatinib (DASISION) were utilized. A matching-adjusted indirect comparison was conducted by weighting individual patients treated with nilotinib to match baseline characteristics reported for dasatinib-treated patients, including age, gender, ECOG performance status and hematology lab values. After matching, efficacy outcomes were compared for patients treated with nilotinib 300 mg twice daily vs. dasatinib 100 mg once daily. Patients randomized to imatinib 400 mg once daily in each trial were used to assess the adequacy of the matching. Results: Before matching, patients randomized to nilotinib in ENESTnd (n = 273) were older, with a lower median platelet count and more favorable performance status compared to patients randomized to dasatinib in DASISION (n = 259). After matching, all baseline characteristics were balanced across treatment groups. Matched patients treated with nilotinib vs. dasatinib experienced significantly higher rates of MMR (56.8 vs. 45.9%, p = 0.014) and overall survival (99.5 vs. 97.3%, p = 0.046) and numerically higher rates of PFS (98.8 vs. 96.5%). Matched imatinib arms showed no statistically significant or clinically meaningful differences in these outcomes. Limitations: Baseline measures unavailable in one or both trials could not be matched. Adverse event rates were not formally compared across trials due to differences in reporting. Conclusion: Nilotinib was associated with significantly higher rates of MMR and overall survival compared with dasatinib by month 12 in the treatment of newly diagnosed CML-CP.

Comparative efficacy of biological treatments for moderate-to-severe psoriasis: a network meta-analysis adjusting for cross-trial differences in reference arm response

Multiple biological therapies are approved for the treatment of moderate‐to‐severe psoriasis.

ASAS40 and ASDAS clinical responses in the ABILITY-1 clinical trial translate to meaningful improvements in physical function, health-related quality of life and work productivity in patients with non-radiographic axial spondyloarthritis

Objective. To assess the impact of achieving Assessment in SpondyloArthritis international Society 40% (ASAS40) response or an Ankylosing Spondylitis Disease Activity Score inactive disease (ASDAS-ID) state on patient-reported outcomes (PROs) among patients with non-radiographic axial SpA (nr-axSpA). Methods. Data are from ABILITY-1, a phase 3 trial of adalimumab vs placebo in nr-axSpA patients. PROs included the HAQ for Spondyloarthropathies (HAQ-S), 36-item Short Form Health Survey (SF-36) physical component summary (PCS) score and Work Productivity and Activity Impairment Questionnaire. Patients were grouped by clinical response using ASAS40 response and ASDAS disease states at week 12. Changes in PROs from baseline to week 12 were compared between groups using analysis of covariance with adjustment for baseline scores. Results. At week 12, 47 of 179 patients were ASAS40 responders and 26 of 176 patients achieved ASDAS-ID (ASDAS <1.3). Compared with non-responders (n = 132), ASAS40 responders (n = 47) had a significantly greater improvement in mean HAQ-S (–0.65 vs -0.05, P < 0.0001), SF-36 PCS (12.4 vs 0.7, P < 0.0001), presenteeism (–24.7 vs -2.2, P < 0.0001), overall work impairment (–23.9 vs -2.5, P < 0.0001) and activity impairment (–33.5 vs -0.9, P < 0.0001) at week 12. Similarly, ASDAS-ID, ASDAS clinically important improvement (ASDAS-CII; improvement >1.1) and major improvement (ASDAS-MI; improvement >2.0) were associated with significantly greater improvements from baseline in the majority of the PROs. Conclusion. Among nr-axSpA patients, ASAS40, ASDAS-CII and ASDAS-MI response and achievement of ASDAS-ID were associated with statistically significant and clinically meaningful improvements in physical function, health-related quality of life and work productivity in a higher percentage of patients.

An indirect comparison and cost per responder analysis of adalimumab, methotrexate and apremilast in the treatment of methotrexate-naïve patients with psoriatic arthritis

Abstract Objective Apremilast was recently approved for the treatment of active psoriatic arthritis (PsA). However, no studies compare apremilast with methotrexate or biologic therapies, so its relative comparative efficacy remains unknown. This study compared the response rates and incremental costs per responder associated with methotrexate, apremilast, and biologics for the treatment of active PsA. Methods A systematic literature review was performed to identify phase 3 randomized controlled clinical trials of approved biologics, methotrexate, and apremilast in the methotrexate-naïve PsA population. Using Bayesian methods, a network meta-analysis was conducted to indirectly compare rates of achieving a ≥20% improvement in American College of Rheumatology component scores (ACR20). The number needed to treat (NNT) and the incremental costs per ACR20 responder (2014 US

One-year and long-term molecular response to nilotinib and dasatinib for newly diagnosed chronic myeloid leukemia: a matching-adjusted indirect comparison

) relative to placebo were estimated for each of the therapies. Results Three trials (MIPA for methotrexate, PALACE-4 for apremilast, and ADEPT for adalimumab) met all inclusion criteria. The NNTs relative to placebo were 2.63 for adalimumab, 6.69 for apremilast, and 8.31 for methotrexate. Among methotrexate-naïve PsA patients, the 16 week incremental costs per ACR20 responder were

Comparative efficacy of long-acting β2-agonists as monotherapy for chronic obstructive pulmonary disease: a network meta-analysis

3622 for methotrexate,

Effect of Adalimumab on Visual Functioning in Patients With Noninfectious Intermediate Uveitis, Posterior Uveitis, and Panuveitis in the VISUAL-1 and VISUAL-2 Trials

26,316 for adalimumab, and

Network Meta-Analysis and Cost Per Responder of Tumor Necrosis Factor-α and Interleukin Inhibitors in the Treatment of Active Ankylosing Spondylitis

45,808 for apremilast. The incremental costs per ACR20 responder were

Comparative efficacy and safety of daclatasvir/asunaprevir versus IFN-based regimens in genotype 1b hepatitis C virus infection.

222,488 for apremilast vs. methotrexate. Conclusion Among methotrexate-naive PsA patients, adalimumab was found to have the lowest NNT for one additional ACR20 response and methotrexate was found to have the lowest incremental costs per ACR20 responder. There was no statistical evidence of greater efficacy for apremilast vs. methotrexate. A head-to-head trial between apremilast and methotrexate is recommended to confirm this finding.