Using individual participant data to improve network meta-analysis projects

Weekly Versus Bolus Cisplatin Concurrent With Definitive Radiation Therapy for Squamous Carcinoma of the Head and Neck: A Systematic Review and Network Meta-Analysis

PURPOSE

The schedule of cisplatin concurrent with definitive radiation for squamous carcinoma of the head and neck remains controversial. Most institutions deliver either a high-dose "bolus" schedule once every 3 weeks or a low-dose weekly schedule. We compared these 2 schedules via a simplified network meta-analysis with a common comparator.

METHODS AND MATERIALS

We performed a PRISMA-concordant systematic review to identify randomized controlled trials comparing cisplatin with cetuximab for nonmetastatic, locoregionally advanced squamous carcinoma of the head and neck treated with definitive radiation. Trials incorporating primary surgery or induction therapy were excluded. Patient survival times were extracted on a per-event basis from the published curves using a digitizer and validated against published point estimates and hazard ratios (HRs). Survival was compared using random effects Cox regression under a frequentist framework. Toxicity and secondary endpoints were analyzed qualitatively. The Cochrane method assessed the risk of bias. The analysis plan was preregistered with the Open Science Foundation.

RESULTS

Five randomized trials were identified, including 1678 patients. There was no statistical difference in overall survival between weekly and bolus regimens (HR, 0.90; 95% CI, 0.53-1.52, P = .345). This Cox model suggested that for the average patient in the cohort, the absolute difference in 5-year overall survival between weekly and bolus regimens was +1.2% (95% CI, -6.1%-+5.9%, P = .345). Secondary endpoints and toxicity were not obviously different by regimen, qualitatively.

CONCLUSIONS

The cetuximab trials provide indirect data suggesting that the differences between cisplatin schedules are subtle.

Pulmonary arterial hypertension treatment: an individual participant data network meta-analysis

BACKGROUND AND AIMS

Effective therapies that target three main signalling pathways are approved to treat pulmonary arterial hypertension (PAH). However, there are few large patient-level studies that compare the effectiveness of these pathways. The aim of this analysis was to compare the effectiveness of the treatment pathways in PAH and to assess treatment heterogeneity.

METHODS

A network meta-analysis was performed using individual participant data of 6811 PAH patients from 20 Phase III randomized clinical trials of therapy for PAH that were submitted to the US Food and Drug Administration. Individual drugs were grouped by the following treatment pathways: endothelin, nitric oxide, and prostacyclin pathways.

RESULTS

The mean (±standard deviation) age of the sample was 49.2 (±15.4) years; 78.4% were female, 59.7% had idiopathic PAH, and 36.5% were on background PAH therapy. After covariate adjustment, targeting the endothelin + nitric oxide pathway {β: 43.7 m [95% confidence interval (CI): 32.9, 54.4]}, nitric oxide pathway [β: 29.4 m (95% CI: 22.6, 36.3)], endothelin pathway [β: 25.3 m (95% CI: 19.8, 30.8)], and prostacyclin pathway [oral/inhaled β: 19.1 m (95% CI: 14.2, 24.0), intravenous/subcutaneous β: 24.4 m (95% CI: 15.1, 33.7)] significantly increased 6 min walk distance at 12 or 16 weeks compared with placebo. Treatments also significantly reduced the likelihood of having clinical worsening events. There was significant heterogeneity of treatment effects by age, body mass index, hypertension, diabetes, and coronary artery disease.

CONCLUSIONS

Drugs targeting the three traditional treatment pathways significantly improve outcomes in PAH, with significant treatment heterogeneity in patients with some comorbidities. Randomized clinical trials are warranted to identify the most effective treatment strategies in a personalized approach.

Implementing Multilevel Network Meta-Regression for Time-To-Event Outcomes: A Case Study in Relapsed Refractory Multiple Myeloma

OBJECTIVES

Multilevel network meta-regression (ML-NMR) leverages individual patient data (IPD) and aggregate data from a network of randomized controlled trials (RCTs) to assess the comparative efficacy of multiple treatments, while adjusting for between-study differences. We provide an overview of ML-NMR for time-to-event outcomes and apply it to an illustrative case study, including example R code.

METHODS

The case study evaluated the comparative efficacy of idecabtagene vicleucel (ide-cel), selinexor+dexamethasone (Sd), belantamab mafodotin (BM), and conventional care (CC) for patients with triple-class exposed relapsed/refractory multiple myeloma in terms of overall survival. Single-arm clinical trials and real-world data were naively combined to create an aggregate data artificial RCT (aRCT) (MAMMOTH-CC versus DREAMM-2-BM versus STORM-2-Sd) and an IPD aRCT (KarMMa-ide-cel versus KarMMa-RW-CC). With some assumptions, we incorporated continuous covariates with skewed distributions, reported as median and range. The ML-NMR models adjusted for number of prior lines, triple-class refractory status, and age and were compared using the leave-one-out information criterion. We summarized predicted hazard ratios and survival (95% credible intervals) in the IPD aRCT population.

RESULTS

The Weibull ML-NMR model had the lowest leave-one-out information criterion. Ide-cel was more efficacious than Sd, BM, and CC in terms of overall survival. Effect modifiers had minimal impact on the model, and only triple-class refractory was a prognostic factor.

CONCLUSIONS

We demonstrate an application of ML-NMR for time-to-event outcomes and introduce code that can be used to aid implementation. Given its benefits, we encourage practitioners to utilize ML-NMR when population adjustment is necessary for comparisons of multiple treatments.

A collaborative network trial to evaluate the effectiveness of implementation strategies to maximize adoption of a school-based healthy lunchbox program: a study protocol

Introduction

An important impediment to the large-scale adoption of evidence-based school nutrition interventions is the lack of evidence on effective strategies to implement them. This paper describes the protocol for a "Collaborative Network Trial" to support the simultaneous testing of different strategies undertaken by New South Wales Local Health Districts to facilitate the adoption of an effective school-based healthy lunchbox program ('SWAP IT'). The primary objective of this study is to assess the effectiveness of different implementation strategies to increase school adoption of the SWAP across New South Wales Local Health Districts.

Methods

Within a Master Protocol framework, a collaborative network trial will be undertaken. Independent randomized controlled trials to test implementation strategies to increase school adoption of SWAP IT within primary schools in 10 different New South Wales Local Health Districts will occur. Schools will be randomly allocated to either the intervention or control condition. Schools allocated to the intervention group will receive a combination of implementation strategies. Across the 10 participating Local Health Districts, six broad strategies were developed and combinations of these strategies will be executed over a 6 month period. In six districts an active comparison group (containing one or more implementation strategies) was selected. The primary outcome of the trial will be adoption of SWAP IT, assessed via electronic registration records captured automatically following online school registration to the program. The primary outcome will be assessed using logistic regression analyses for each trial. Individual participant data component network meta-analysis, under a Bayesian framework, will be used to explore strategy-covariate interactions; to model additive main effects (separate effects for each component of an implementation strategy); two way interactions (synergistic/antagonistic effects of components), and full interactions.

Discussion

The study will provide rigorous evidence of the effects of a variety of implementation strategies, employed in different contexts, on the adoption of a school-based healthy lunchbox program at scale. Importantly, it will also provide evidence as to whether health service-centered, collaborative research models can rapidly generate new knowledge and yield health service improvements.

Clinical trial registration

This trial is registered prospectively with the Australian New Zealand Clinical Trials Registry (ACTRN12623000558628).

Unanchored Population-Adjusted Indirect Comparison Methods for Time-to-Event Outcomes Using Inverse Odds Weighting, Regression Adjustment, and Doubly Robust Methods With Either Individual Patient or Aggregate Data

OBJECTIVES

Several methods for unanchored population-adjusted indirect comparisons (PAICs) are available. Exploring alternative adjustment methods, depending on the available individual patient data (IPD) and the aggregate data (AD) in the external study, may help minimize bias in unanchored indirect comparisons. However, methods for time-to-event outcomes are not well understood. This study provides an overview and comparison of methods using a case study to increase familiarity. A recent method is applied to marginalize conditional hazard ratios, which allows for the comparisons of methods, and a doubly robust method is proposed.

METHODS

The following PAIC methods were compared through a case study in third-line small cell lung cancer, comparing nivolumab with standard of care based on a single-arm phase II trial (CheckMate 032) and real-world study (Flatiron) in terms of overall survival: IPD-IPD analyses using inverse odds weighting, regression adjustment, and a doubly robust method; IPD-AD analyses using matching-adjusted indirect comparison, simulated treatment comparison, and a doubly robust method.

RESULTS

Nivolumab extended survival versus standard of care with hazard ratios ranging from 0.63 (95% CI 0.44-0.90) in naive comparisons (identical estimates for IPD-IPD and IPD-AD analyses) to 0.69 (95% CI 0.44-0.98) in the IPD-IPD analyses using regression adjustment. Regression-based and doubly robust estimates yielded slightly wider confidence intervals versus the propensity score-based analyses.

CONCLUSIONS

The proposed doubly robust approach for time-to-event outcomes may help to minimize bias due to model misspecification. However, all methods for unanchored PAIC rely on the strong assumption that all prognostic covariates have been included.

Mechanical circulatory support for cardiogenic shock: a network meta-analysis of randomized controlled trials and propensity score-matched studies

PURPOSE

Cardiogenic shock is associated with high mortality. In refractory shock, it is unclear if mechanical circulatory support (MCS) devices improve survival. We conducted a network meta-analysis to determine which MCS devices confers greatest benefit.

METHODS

We searched MEDLINE, Embase, and Scopus databases through 27 August 2023 for relevant randomized controlled trials (RCTs) and propensity score-matched studies (PSMs). We conducted frequentist network meta-analysis, investigating mortality (either 30 days or in-hospital) as the primary outcome. We assessed risk of bias (Cochrane risk of bias 2.0 tool/Newcastle-Ottawa Scale) and as sensitivity analysis reconstructed survival data from published survival curves for a one-stage unadjusted individual patient data (IPD) meta-analysis using a stratified Cox model.

RESULTS

We included 38 studies (48,749 patients), mostly reporting on patients with Society for Cardiovascular Angiography and Intervention shock stages C-E cardiogenic shock. Compared with no MCS, extracorporeal membrane oxygenation with intra-aortic balloon pump (ECMO-IABP; network odds ratio [OR]: 0.54, 95% confidence interval (CI): 0.33-0.86, moderate certainty) was associated with lower mortality. There were no differences in mortality between ECMO, IABP, microaxial ventricular assist device (mVAD), ECMO-mVAD, centrifugal VAD, or mVAD-IABP and no MCS (all very low certainty). Our one-stage IPD survival meta-analysis based on the stratified Cox model found only ECMO-IABP was associated with lower mortality (hazard ratio, HR, 0.55, 95% CI 0.46-0.66).

CONCLUSION

In patients with cardiogenic shock, ECMO-IABP may reduce mortality, while other MCS devices did not reduce mortality. However, this must be interpreted within the context of inter-study heterogeneity and limited certainty of evidence.

The lymphatic filariasis treatment study landscape: A systematic review of study characteristics and the case for an individual participant data platform

BACKGROUND

Lymphatic filariasis (LF) is a neglected tropical disease (NTD) targeted by the World Health Organization for elimination as a public health problem (EPHP). Since 2000, more than 9 billion treatments of antifilarial medicines have been distributed through mass drug administration (MDA) programmes in 72 endemic countries and 17 countries have reached EPHP. Yet in 2021, nearly 900 million people still required MDA with combinations of albendazole, diethylcarbamazine and/or ivermectin. Despite the reliance on these drugs, there remain gaps in understanding of variation in responses to treatment. As demonstrated for other infectious diseases, some urgent questions could be addressed by conducting individual participant data (IPD) meta-analyses. Here, we present the results of a systematic literature review to estimate the abundance of IPD on pre- and post-intervention indicators of infection and/or morbidity and assess the feasibility of building a global data repository.

METHODOLOGY

We searched literature published between 1st January 2000 and 5th May 2023 in 15 databases to identify prospective studies assessing LF treatment and/or morbidity management and disease prevention (MMDP) approaches. We considered only studies where individual participants were diagnosed with LF infection or disease and were followed up on at least one occasion after receiving an intervention/treatment.

PRINCIPAL FINDINGS

We identified 138 eligible studies from 23 countries, having followed up an estimated 29,842 participants after intervention. We estimate 14,800 (49.6%) IPD on pre- and post-intervention infection indicators including microfilaraemia, circulating filarial antigen and/or ultrasound indicators measured before and after intervention using 8 drugs administered in various combinations. We identified 33 studies on MMDP, estimating 6,102 (20.4%) IPD on pre- and post-intervention clinical morbidity indicators only. A further 8,940 IPD cover a mixture of infection and morbidity outcomes measured with other diagnostics, from participants followed for adverse event outcomes only or recruited after initial intervention.

CONCLUSIONS

The LF treatment study landscape is heterogeneous, but the abundance of studies and related IPD suggest that establishing a global data repository to facilitate IPD meta-analyses would be feasible and useful to address unresolved questions on variation in treatment outcomes across geographies, demographics and in underrepresented groups. New studies using more standardized approaches should be initiated to address the scarcity and inconsistency of data on morbidity management.

Short, medium, and long deferral of umbilical cord clamping compared with umbilical cord milking and immediate clamping at preterm birth: a systematic review and network meta-analysis with individual participant data

BACKGROUND

Deferred (also known as delayed) cord clamping can improve survival of infants born preterm (before 37 weeks of gestation), but the optimal duration of deferral remains unclear. We conducted a systematic review and individual participant data network meta-analysis with the aim of comparing the effectiveness of umbilical cord clamping strategies with different timings of clamping or with cord milking for preterm infants.

METHODS

We searched medical databases and trial registries from inception until Feb 24, 2022 (updated June 6, 2023) for randomised controlled trials comparing cord clamping strategies for preterm infants. Individual participant data were harmonised and assessed for risk of bias and quality. Interventions were grouped into immediate clamping, short deferral (≥15 s to <45 s), medium deferral (≥45 s to <120 s), long deferral (≥120 s), and intact cord milking. The primary outcome was death before hospital discharge. We calculated one-stage, intention-to-treat Bayesian random-effects individual participant data network meta-analysis. This study was registered with PROSPERO, CRD42019136640.

FINDINGS

We included individual participant data from 47 trials with 6094 participants. Of all interventions, long deferral reduced death before discharge the most (compared with immediate clamping; odds ratio 0·31 [95% credibility interval] 0·11-0·80; moderate certainty). The risk of bias was low for 10 (33%) of 30 trials, 14 (47%) had some concerns, and 6 (20%) were rated as having a high risk of bias. Heterogeneity was low, with no indication of inconsistency.

INTERPRETATION

This study found that long deferral of clamping leads to reduced odds of death before discharge in preterm infants. In infants assessed as requiring immediate resuscitation, this finding might only be generalisable if there are provisions for such care with the cord intact. These results are based on thoroughly cleaned and checked individual participant data and can inform future guidelines and practice.

FUNDING

Australian National Health and Medical Research Council.

A Comparison of Relative-Efficacy Estimate(S) Derived From Both Matching-Adjusted Indirect Comparisons and Standard Anchored Indirect Treatment Comparisons: A Review of Matching-Adjusted Indirect Comparisons

OBJECTIVES

We present an empirical comparison of relative-efficacy estimate(s) from matching-adjusted indirect comparisons (MAICs) with estimates from corresponding standard anchored indirect treatment comparisons.

METHODS

A total of 80 comparisons were identified from 17 publications through a systematic rapid review. A standardized metric that used reported relative treatment efficacy estimates and their associated uncertainty was used to compare the methods across different treatment indications and outcome measures.

RESULTS

On aggregate, MAICs presented for connected networks tended to report a more favorable relative-efficacy estimate for the treatment for which individual-level patient data were available relative to the reported indirect treatment comparison estimate.

CONCLUSIONS

Although we recognize the importance of MAIC and other population adjustment methods in certain situations, we recommend that results from these analyses are interpreted with caution. Researchers and analysts should carefully consider if MAICs are appropriate where presented and whether MAICs would have added value where omitted.

Individual Participant Data Network Meta-Analysis of Neoadjuvant Chemotherapy or Chemoradiotherapy in Esophageal or Gastroesophageal Junction Carcinoma

PURPOSE

The optimal neoadjuvant treatment for resectable carcinoma of the thoracic esophagus (TE) or gastroesophageal junction (GEJ) remains a matter of debate. We performed an individual participant data (IPD) network meta-analysis (NMA) of randomized controlled trials (RCTs) to study the effect of chemotherapy or chemoradiotherapy, with a focus on tumor location and histology subgroups.

PATIENTS AND METHODS

All, published or unpublished, RCTs closed to accrual before December 31, 2015 and having compared at least two of the following strategies were eligible: upfront surgery (S), chemotherapy followed by surgery (CS), and chemoradiotherapy followed by surgery (CRS). All analyses were conducted on IPD obtained from investigators. The primary end point was overall survival (OS). The IPD-NMA was analyzed by a one-step mixed-effect Cox model adjusted for age, sex, tumor location, and histology. The NMA was registered in PROSPERO (CRD42018107158).

RESULTS

IPD were obtained for 26 of 35 RCTs (4,985 of 5,807 patients) corresponding to 12 comparisons for CS-S, 12 for CRS-S, and four for CRS-CS. CS and CRS led to increased OS when compared with S with hazard ratio (HR) = 0.86 (0.75 to 0.99), P = .03 and HR = 0.77 (0.68 to 0.87), P < .001 respectively. The NMA comparison of CRS versus CS for OS gave a HR of 0.90 (0.74 to 1.09), P = .27 (consistency P = .26, heterogeneity P = .0038). For CS versus S, a larger effect on OS was observed for GEJ versus TE tumors (P = .036). For the CRS versus S and CRS versus CS, a larger effect on OS was observed for women (P = .003, .012, respectively).

CONCLUSION

Neoadjuvant chemotherapy and chemoradiotherapy were consistently better than S alone across histology, but with some variation in the magnitude of treatment effect by sex for CRS and tumor location for CS. A strong OS difference between CS and CRS was not identified.

Correcting calculation and data errors reveals that the original conclusions were incorrect in "The best drug supplement for obesity treatment: a systematic review and network meta-analysis"

The goal of this study was to reproduce and evaluate the reliability of the network meta-analysis performed in the article "The best drug supplement for obesity treatment: A systematic review and network meta-analysis" by Salari et al. In recent years, it has become more common to employ network meta-analysis to assess the relative efficacy of treatments often used in clinical practice. To duplicate Salari et al.'s research, we pulled data directly from the original trials and used Cohen's D to determine the effect size for each treatment. We reanalyzed the data since we discovered significant differences between the data we retrieved and the data given by Salari et al. We present new effect size estimates for each therapy and conclude that the prior findings were somewhat erroneous. Our findings highlight the importance of ensuring the accuracy of network meta-analyses to determine the quality and strength of existing evidence.