Performance of unanchored matching-adjusted indirect comparison (MAIC) for the evidence synthesis of single-arm trials with time-to-event outcomes

Jiang, Yawen; Ni, Weiyi

doi:10.1186/s12874-020-01124-6

Cited by 16 publications

(23 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…13 When RCTs are not available, methods for 'unanchored' meta-analysis can still be used, such as matching-adjusted indirect comparisons (MAIC) and simulated treatment comparisons (STC). [14][15][16] When patient level data is available, more direct comparison methods can be applied such as propensity score matching or Bayesian methods to help reduce selection bias and differences between comparator cohorts. [17][18][19] Some formal HTA guidance is available on these approaches, and their limitations from National Institute for Health and Care Excellence's (NICE) decision support unit.…”

Section: Introductionmentioning

confidence: 99%

Use of External Comparators for Health Technology Assessment Submissions Based on Single-Arm Trials

et al. 2021

View full text Add to dashboard Cite

Background: Single-arm trial (SAT) data is increasingly reviewed for drug approvals by regulators and Health Technology Assessment (HTA) bodies. Supplementary data in the form of external comparators (ECs) can be used to provide clinical context to support these drug evaluations. In this study we characterized HTAs for SAT-based submissions, the use of supplementary EC data and outcomes from HTA review.Methods: HTA Accelerator database was used to describe SAT-based HTA submissions with decisions (2011-2019).Results: A total of 433 SAT-based HTA submissions were identified between 2011 and 2019 with a 13-fold increase during this period. Around 65%(283/433) were in oncology or hem-oncology. Around 52%(226/433) of submissions contained some type of EC data, including prior clinical trials (24%, 104) and real-world data (RWD) (20%, 87), but 40%(175) contained no EC data. The overall acceptance rate for SAT-based submissions was 48% and with RWD EC data acceptance was 59%. In the latest 5year period (2015-2019), use of RWD ECs increased 22% as a proportion of submissions per year, whereas, prior trial ECs decreased (-14%) and use of no EC remained stable (-2%). Between 2015 to 2017 and 2018 to 2019, acceptance rate for RWD ECs increased by 20% (41% in 2015-2017 to 61% in 2018-2019) whereas prior trial EC use decreased by 10% and no EC submissions decreased 16%. Of 226 submissions using ECs, only 29%(66) used an adjusted indirect treatment comparison method.Conclusions: SAT-based submissions to HTA bodies are rapidly evolving in terms of composition and acceptance. Types of EC and methodological approach used are important determinants of positive outcomes.

show abstract

Section: Introductionmentioning

confidence: 99%

Use of External Comparators for Health Technology Assessment Submissions Based on Single-Arm Trials

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Under no failures of assumptions, MAIC has produced unbiased treatment effect estimation in simulation studies [7,[14][15][16][17][18][19][20]. Nevertheless, there are some concerns about its inefficiency and instability, particularly where covariate overlap is poor and effective sample sizes (ESSs) after weighting are small [21].…”

mentioning

confidence: 99%

Two-stage matching-adjusted indirect comparison

Remiro‐Azócar

2022

BMC Med Res Methodol

View full text Add to dashboard Cite

Background Anchored covariate-adjusted indirect comparisons inform reimbursement decisions where there are no head-to-head trials between the treatments of interest, there is a common comparator arm shared by the studies, and there are patient-level data limitations. Matching-adjusted indirect comparison (MAIC), based on propensity score weighting, is the most widely used covariate-adjusted indirect comparison method in health technology assessment. MAIC has poor precision and is inefficient when the effective sample size after weighting is small. Methods A modular extension to MAIC, termed two-stage matching-adjusted indirect comparison (2SMAIC), is proposed. This uses two parametric models. One estimates the treatment assignment mechanism in the study with individual patient data (IPD), the other estimates the trial assignment mechanism. The first model produces inverse probability weights that are combined with the odds weights produced by the second model. The resulting weights seek to balance covariates between treatment arms and across studies. A simulation study provides proof-of-principle in an indirect comparison performed across two randomized trials. Nevertheless, 2SMAIC can be applied in situations where the IPD trial is observational, by including potential confounders in the treatment assignment model. The simulation study also explores the use of weight truncation in combination with MAIC for the first time. Results Despite enforcing randomization and knowing the true treatment assignment mechanism in the IPD trial, 2SMAIC yields improved precision and efficiency with respect to MAIC in all scenarios, while maintaining similarly low levels of bias. The two-stage approach is effective when sample sizes in the IPD trial are low, as it controls for chance imbalances in prognostic baseline covariates between study arms. It is not as effective when overlap between the trials’ target populations is poor and the extremity of the weights is high. In these scenarios, truncation leads to substantial precision and efficiency gains but induces considerable bias. The combination of a two-stage approach with truncation produces the highest precision and efficiency improvements. Conclusions Two-stage approaches to MAIC can increase precision and efficiency with respect to the standard approach by adjusting for empirical imbalances in prognostic covariates in the IPD trial. Further modules could be incorporated for additional variance reduction or to account for missingness and non-compliance in the IPD trial.

show abstract

“…Second, although the RCT-MAIC cross validation suggested good validity of the MAIC analyses, residual effect modifiers that might confound the results could not be fully ruled out. 17 The E-values suggest that, whereas the effects of AB and LP compared with sorafenib were relatively robust to residual confounders, the effects of AB in relation to LP might entail some sensitivity to bias. Third, the sample size of the KEYNOTE 524 trial and the ESS of the corresponding MAIC were moderate, which might undermine the power of the analyses to detect true effectiveness.…”

Section: Discussionmentioning

confidence: 98%

“…The EB process created weights that could be used to align the characteristics of the IPD with those of the AgD in subsequent analyses. 17 , 18 Using this process, sorafenib IPD were first balanced to the AB AgD, following which the effective sample size (ESS) with the EB weights was also calculated. After the EB process, the OS and PFS K-M curves of the AB arm in the IMbrave150 trial were then digitized using Engauge Digitizer 10.11 to reconstruct the corresponding individual-level survival data with an established method.…”

Section: Methodsmentioning

confidence: 99%

Indirect Comparisons via Sorafenib for the Comparative Effectiveness of Two PD-1/PD-L1 Inhibitors to Treat Advanced Hepatocellular Carcinoma Patients without Prior Systemic Therapies

Jiang¹,

Dan²,

Shi³

2022

CLEP

Self Cite

View full text Add to dashboard Cite

Introduction Advanced hepatocellular carcinoma (HCC) represents a major public health threat. Several emerging combination therapies have shown promising results for the first-line treatment of advanced HCC. The present study compared the efficacy of atezolizumab plus bevacizumab (AB) with lenvatinib plus pembrolizumab (LP), which were two of the leading combination therapies. Methods The outcomes of the present analysis were overall survival (OS) time and progression-free survival (PFS) time. Two matching-adjusted indirect comparisons (MAICs) were first conducted using the individual-level patient data (IPD) of the sorafenib arm from a previous clinical trial and the aggregate data (AgD) of the AB and LP arms from the corresponding published trials. From the MAICs, the hazard ratios (HRs) of AB and LP vs sorafenib were estimated by conducting weighted Cox regressions. The HRs from the two MAICs were then pooled to conduct a second-order indirect comparison of AB vs LP. Results In the MAIC analyses, AB had better efficacy on both OS (HR: 0.58, 95% CI: 0.42–0.79) and PFS (HR: 0.59, 95% CI: 0.47–0.76) than sorafenib, whereas LP had significantly better efficacy on PFS (HR: 0.62, 95% CI: 0.41–0.94) but not OS (HR: 0.83, 95% CI: 0.52–1.32). In the second-order comparison, AB was insignificantly more efficacious on OS (HR: 0.71, 95% CI: 0.42–1.23) than and similarly efficacious on PFS (HR: 0.95, 95% CI: 0.60–1.51) as the LP regimen. Conclusion LP regimen may be a potential first-line immunotherapy option for advanced HCC given its comparative effectiveness in relation to AB.

show abstract

Performance of unanchored matching-adjusted indirect comparison (MAIC) for the evidence synthesis of single-arm trials with time-to-event outcomes

Cited by 16 publications

References 19 publications

Use of External Comparators for Health Technology Assessment Submissions Based on Single-Arm Trials

Use of External Comparators for Health Technology Assessment Submissions Based on Single-Arm Trials

Two-stage matching-adjusted indirect comparison

Indirect Comparisons via Sorafenib for the Comparative Effectiveness of Two PD-1/PD-L1 Inhibitors to Treat Advanced Hepatocellular Carcinoma Patients without Prior Systemic Therapies

Contact Info

Product

Resources

About