Multiple Treatment Comparisons Research Articles

A network meta-analysis in comparing prophylactic treatments of radiotherapy-induced oral mucositis for patients with head and neck cancers receiving radiotherapy

ObjectivesThe objective of this network meta-analysis is to optimize the prophylactic treatment for radiotherapy-induced oral mucositis in patients with head and neck cancer (HNSCC) receiving postoperative or definitive radiotherapy with or without chemotherapy. Materials and methodsWe searched electronic databases to identify all eligible randomized clinical trials on oral mucositis. The endpoint was grade 0–2 oral mucositis. Odds ratios (OR) and the corresponding 95% confidence intervals (CI) were extracted. Network meta-analysis was performed using the frequentist approach to conduct multiple treatment comparisons. ResultsIn total, 57 trials with 5261 patients were eligible for this study. Both direct and network meta-analysis revealed that low-level laser additional to standard oral care (SOC) was better than most of the other treatments and achieved the highest effect on grade 0–2 oral mucositis, with a surface under the cumulative ranking curve (SUCRA) of 95.8%; however, SOC with or without placebo had worse effect than most of the other treatments and was ranked worst (SUCRA = 0.4%). Moreover, sensitivity analysis performed after we had combined the SOC and placebo groups (non-medication treatment, NMT) yielded similar results, with SUCRA of 91.3% and 3.5% for low-level laser and NMT, respectively. ConclusionsLow-level laser additional to SOC may be a more effective prophylactic treatment for reducing severe radiotherapy-induced oral mucositis; SOC alone is insufficient for patients with HNSCC receiving postoperative or definitive radiotherapy with or without chemotherapy.

Overcoming the Straw Man Effect in Oncology: Visualization and Ranking of Chemotherapy Regimens Using an Information Theoretic Approach.

PurposeDespite the plethora of randomized controlled trial (RCT) data, most cancer treatment recommendations are formulated by experts. Alternatively, network meta-analysis (NMA) is one method of analyzing multiple indirect treatment comparisons. However, NMA does not account for mixed end points or temporality. Previously, we described a prototype information theoretical approach for the construction of ranked chemotherapy treatment regimen networks. Here, we propose modifications to overcome an apparent straw man effect, where the most studied regimens were the most negatively valued.MethodsRCTs from two scenarios—upfront treatment of chronic myelogenous leukemia and relapsed/refractory multiple myeloma—were assembled into ranked networks using an automated algorithm based on effect sizes, statistical significance, surrogacy of end points, and time since RCT publication. Vertex and edge color, transparency, and size were used to visually analyze the network. This analysis led to the additional incorporation of value propagation.ResultsA total of 18 regimens with 42 connections (chronic myelogenous leukemia) and 28 regimens with 25 connections (relapsed/refractory multiple myeloma) were analyzed. An initial negative correlation between vertex value and size was ameliorated after value propagation, although not eliminated. Updated rankings were in close agreement with published guidelines and NMAs.ConclusionStraw man effects can distort the comparative efficacy of newer regimens at the expense of older regimens, which are often cheaper or less toxic. Using an automated method, we ameliorated this effect and produced rankings consistent with common practice and published guidelines in two distinct cancer settings. These findings are likely to be generalizable and suggest a new means of ranking efficacy in cancer trials.

Comparison of 7 Surgical Interventions for Lumbar Disc Herniation: A Network Meta-analysis

Background: The number of interventions on intervertebral discs rapidly increased and the treatment options for lumbar disc surgery quickly evolved. It is important that the safety and efficacy of all new innovative procedures be compared with currently accepted forms of treatment; however, the previous pairwise meta-analyses could not develop the hierarchy of these treatments. Objectives: The purpose of the study is to perform a network meta-analysis to evaluate the clinical results of 7 surgical interventions for the treatment of lumbar disc herniation. Study Design: Network meta-analysis of randomized controlled trials (RCTs) for multiple treatment comparisons of lumbar disc herniation. Methods: We performed a Bayesian-framework network meta-analysis of RCTs to compare 7 surgical interventions for people with lumbar disc herniation. The eligible RCTs were identified by searching Embase, Pubmed, the Cochrane Central Register of Controlled Trials (CENTRAL), and Google scholar. Data from 3 outcomes (success, complications, and reoperation rate) were independently extracted by 2 authors. Results: A total of 29 RCTs including 3,146 participants were finally included into this article. Our meta-analysis provides hierarchies of these 7 interventions. For the success rate the rank probability (from best to worst): percutaneous endoscopic lumber discectomy (PELD) > standard open discectomy (SOD) > standard open microsurgical discectomy (SOMD) > chemonucleolysis (CN) > microendoscopic discectomy (MED) > percutaneous laser disc decompression (PLDD) > automated percutaneous lumber discectomy (APLD). For the complication rate the rank probability (from best to worst): PELD > SOMD > SOD > MED > PLDD > CN > APLD. For the reoperation rate the rank probability (from best to worst): SOMD > SOD > MED > PLDD > PELD > CN > APLD. Limitations: The limitations of this network meta-analysis include the range of study populations and inconformity of the follow-up times and outcome measurements. Conclusions: This meta-analysis provides evidence that PELD might be the best choice to increase the success rate and decrease the complication rate, moreover SOMD might be the best option to drop the reoperation rate. APLD might lead to the lowest success rate and the highest complication and reoperation rate. Higher quality RCTs and direct head to head trials are needed to confirm these results.

A Bayesian meta-analysis of multiple treatment comparisons of pancreatic duct stent drainage for pancreaticoduodenectomy

Perivascular epithelioid cell tumor (PEComa) is a rare mesenchymal tumor, with distinctive morphology and expression of myo-melanocytic markers.Current scientific literature reported just 24 cases of pancreatic PEComas. With our systematic review, we aimed at improving our understanding of the disease, focusing on the knowledge gained on epidemiology, etiology, clinical presentation, diagnosis, treatment and prognosis.Based on the PRISMA guidelines, a systematic research was carried out on PubMed/MEDLINE, Web of Science, Scopus, EMBASE, and EBSCO using the search terms: (“perivascular epithelioid cell tumor” OR ″PEComa”) and (“pancreas “OR″ pancreatic”).The 4th-6th decades of life and female sex (86.9%) turned out as the most affected. Pancreatic head was the most involved site (50%), with a single lesion in almost all cases. The diagnosis was only obtained after histopathological examination (70.8%). The clinical presentation was non-specific, abdominal pain being the main symptom (60.9%). At immunohistochemistry, PEComa showed benign epithelioid predominance and a strong positivity for HMB-45, Melan-A, and α-SMA. Surgical resection was performed in almost all cases, while for one patient the multidisciplinary group chose just endoscopic and imaging follow-up, based on the benign nature of the lesion.The biological characteristics of pancreatic PEComa remain an enigma. Its prognosis seems to depend on whether atypical (“worrisome”) histological features are available or not. Surgery turned out as the most appropriate treatment, without reaching any agreement on surgery timing. Further studies on larger population are needed to better understand the biological features of pancreatic PEComa, in order to set up guidelines in the diagnosis, treatment and follow-up.

Multiple treatment comparison of seven new drugs for patients with advanced malignant melanoma: a systematic review and health economic decision model in a Norwegian setting

ObjectiveTo assess the relative effectiveness and cost-effectiveness of seven new drugs (cobimetinib, dabrafenib, ipilimumab, nivolumab, pembrolizumab, trametinib and vemurafenib) used for treatment of patients with advanced malignant melanoma in the...

Effectiveness of non-steroidal anti-inflammatory drugs for the treatment of pain in knee and hip osteoarthritis: a network meta-analysis

Non-steroidal anti-inflammatory drugs (NSAIDs) are the backbone of osteoarthritis pain management. We aimed to assess the effectiveness of different preparations and doses of NSAIDs on osteoarthritis pain in a network meta-analysis. For this network meta-analysis, we considered randomised trials comparing any of the following interventions: NSAIDs, paracetamol, or placebo, for the treatment of osteoarthritis pain. We searched the Cochrane Central Register of Controlled Trials (CENTRAL) and the reference lists of relevant articles for trials published between Jan 1, 1980, and Feb 24, 2015, with at least 100 patients per group. The prespecified primary and secondary outcomes were pain and physical function, and were extracted in duplicate for up to seven timepoints after the start of treatment. We used an extension of multivariable Bayesian random effects models for mixed multiple treatment comparisons with a random effect at the level of trials. For the primary analysis, a random walk of first order was used to account for multiple follow-up outcome data within a trial. Preparations that used different total daily dose were considered separately in the analysis. To assess a potential dose-response relation, we used preparation-specific covariates assuming linearity on log relative dose. We identified 8973 manuscripts from our search, of which 76 randomised trials with a total of 58 451 patients were included in this analysis. 23 nodes concerning seven different NSAIDs or paracetamol with specific daily dose of administration or placebo were considered. All preparations, irrespective of dose, improved point estimates of pain symptoms when compared with placebo. For six interventions (diclofenac 150 mg/day, etoricoxib 30 mg/day, 60 mg/day, and 90 mg/day, and rofecoxib 25 mg/day and 50 mg/day), the probability that the difference to placebo is at or below a prespecified minimum clinically important effect for pain reduction (effect size [ES] -0·37) was at least 95%. Among maximally approved daily doses, diclofenac 150 mg/day (ES -0·57, 95% credibility interval [CrI] -0·69 to -0·45) and etoricoxib 60 mg/day (ES -0·58, -0·74 to -0·43) had the highest probability to be the best intervention, both with 100% probability to reach the minimum clinically important difference. Treatment effects increased as drug dose increased, but corresponding tests for a linear dose effect were significant only for naproxen (p=0·034). We found no evidence that treatment effects varied over the duration of treatment. Model fit was good, and between-trial heterogeneity and inconsistency were low in all analyses. All trials were deemed to have a low risk of bias for blinding of patients. Effect estimates did not change in sensitivity analyses with two additional statistical models and accounting for methodological quality criteria in meta-regression analysis. On the basis of the available data, we see no role for single-agent paracetamol for the treatment of patients with osteoarthritis irrespective of dose. We provide sound evidence that diclofenac 150 mg/day is the most effective NSAID available at present, in terms of improving both pain and function. Nevertheless, in view of the safety profile of these drugs, physicians need to consider our results together with all known safety information when selecting the preparation and dose for individual patients. Swiss National Science Foundation (grant number 405340-104762) and Arco Foundation, Switzerland.

Multiple treatment comparisons in analysis of covariance with interaction

When multiple treatments are analyzed together with a covariate, a treatment-covariate interaction complicates the interpretation of the treatment effects. The construction of simultaneous confidence bands for differences of the treatment specific regression lines is one option to proceed. The application of these methods is difficult because they are described as a collection of special cases and the implementation requires additional programming or relies on non-standard or proprietary software. If inferential interest can be restricted to a pre-specified set of covariate values, a flexible alternative is to compute simultaneous confidence intervals for multiple contrasts of the treatment effects over this grid. This approach is available in the R software: next to treatment differences in the linear model, approximate simultaneous confidence intervals for ratios of expected values and asymptotic extensions to generalized linear models are straightforward. The paper summarizes the available methodology and presents three case studies to illustrate the application to different models, differences and ratios, as well as different types of between treatment comparisons. Simulation studies in the general linear model, for different parameters and different types of comparisons are provided. The R code to reproduce the case studies and a hint to a related R package is provided.

Evaluation of the correlations between patient-derived xenograft (PDX) model-based mouse trials and cancer patient-based clinical trials.

e23140 Background: Clinical trial to evaluate the safety and efficacy of an oncology drug candidate in general is a long and expensive process required for regulatory approval in drug development. High failure rates and difficulties to predict the outcome of clinical trials have wasted the resources and hindered the progress to bring more cancer drugs to patients. PDX models have been widely used recently in preclinical efficacy studies of new cancer agents. Comparing with tumor cell line derived xenografts, PDX models showed more similarities to cancer patients in tumor pathological features, molecular changes and drug responses. More importantly, the large number of available PDX models make it feasible to do preclinical tests in large scales which can represent cancer complexity similar to clinical trials with cancer patients. Methods: At GenenDesign, we have established over 1000 PDX tumor models in various cancer types and derived about 200 acquired drug resistance models. In our in house studies, we have tested near 40 treatments with standard of cares (SoCs) and late stage clinical drug candidates either as mono-therapies or combination therapies, and have generated over 3300 drug response datasets. Many tests with targeted drugs are based on cancer gene aberrations as predictive biomarkers. The drug responses were analyzed by Modifying Response Evaluation Criteria in Solid Tumors (mRECIST) method. Results: By comparing the mRECIST scores in mouse trials with results from about two dozens of historical clinical trials at phase 3 or phase 2, we found that there are high correlations in the rates of objective response (OR), stable disease (SD) and progressive disease (PD) between the mouse trials and their counterpart clinical trials. Progression-free survival also showed similar patterns in these analyses. Conclusions: PDX based mouse trials have advantages in flexibility of study design, quickness in testing new drug candidates and direct comparison of multiple drug treatments in the same models. It can provide valuable information on clinical trial outcome prediction and support clinical study design and patient selection.

Comparative outcome of PFNA, Gamma nails, PCCP, Medoff plate, LISS and dynamic hip screws for fixation in elderly trochanteric fractures: a systematic review and network meta-analysis of randomized controlled trials.

The ideal implant for the treatment of an unstable intertrochanteric femoral fracture is still a matter of discussion. The aim of this systematic review is to conduct a network meta-analysis of randomized controlled trials (RCTs) comparing clinical outcomes between dynamic hip screws (DHS), Medoff sliding plating, percutaneous compression plating (PCCP), proximal femoral nails (PFN), Gamma nails and less invasive stabilization system fixation in femoral trochanteric fractures in the elderly. These clinical outcomes consist of total intra-operative time, intra-operative fluoroscopy time, intra-operative blood loss, blood component transfusion, length of hospital stay, postoperative general complications, wound complications, late complications and reoperation rates. This systematic review was conducted using PubMed and Scopus search engines for RCTs comparing clinical outcomes between treatments from inception to February 22, 2015. Thirty-six of 785 studies identified were eligible. Compared to the other implants, PCCP showed the lowest total operative time and units of blood transfusion with an unstandardized mean difference (UMD) of 29.27min (95% CI 5.24, 53.50) and 0.89 units (95% CI 0.52, 1.25). The lowest incidence of general complications, wound complications and late complications of PCCP was 0.09 (95% CI 0.04, 0.18), 0.01 (95% CI 0.01, 0.04) and 0.05 (95% CI 0.02, 0.11), respectively, when compared to others. The lowest fluoroscopic time was with DHS with an UMD of 0.24min (95% CI 0.16, 0.32), whereas the lowest blood loss and shortest hospital stay were with PFN with an UMD of 233.61ml of blood loss (95% CI 153.17, 314.04) and 7.23days of hospital stay (95% CI 7.15, 7.31) when compared to all other fixation methods. Reoperation rates of all implants had no statistically significant difference. The network meta-analysis suggested that fixation with PCCP significantly shortens operative time and decreases the units of blood transfusion required, while also lowering risks of general complications, wound complications and late complications when compared to fixation. Use of PFN showed the least intra-operative blood loss and shortest hospital stay. Multiple active treatment comparisons indicate that PCCP fixation in trochanteric fractures in the elderly is the treatment of choice in terms of intra-operative outcomes and postoperative complications.

Network meta-analysis: a technique to gather evidence from direct and indirect comparisons.

Systematic reviews and pairwise meta-analyses of randomized controlled trials, at the intersection of clinical medicine, epidemiology and statistics, are positioned at the top of evidence-based practice hierarchy. These are important tools to base drugs approval, clinical protocols and guidelines formulation and for decision-making. However, this traditional technique only partially yield information that clinicians, patients and policy-makers need to make informed decisions, since it usually compares only two interventions at the time. In the market, regardless the clinical condition under evaluation, usually many interventions are available and few of them have been studied in head-to-head studies. This scenario precludes conclusions to be drawn from comparisons of all interventions profile (e.g. efficacy and safety). The recent development and introduction of a new technique – usually referred as network meta-analysis, indirect meta-analysis, multiple or mixed treatment comparisons – has allowed the estimation of metrics for all possible comparisons in the same model, simultaneously gathering direct and indirect evidence. Over the last years this statistical tool has matured as technique with models available for all types of raw data, producing different pooled effect measures, using both Frequentist and Bayesian frameworks, with different software packages. However, the conduction, report and interpretation of network meta-analysis still poses multiple challenges that should be carefully considered, especially because this technique inherits all assumptions from pairwise meta-analysis but with increased complexity. Thus, we aim to provide a basic explanation of network meta-analysis conduction, highlighting its risks and benefits for evidence-based practice, including information on statistical methods evolution, assumptions and steps for performing the analysis.

Indirect comparisons of treatment effects: Network meta-analyses

Many therapeutic options are now available for patients with multiple sclerosis. While the efficacy of each drug has been assessed against placebo or, more recently, against interferon, no direct comparisons of these new therapies have been conducted in randomized clinical trials. Therefore, indirect treatment comparisons are needed to inform clinical decisions. In this brief report, some basic concepts about network meta-analyses that are the formal methods used to run multiple indirect treatment comparisons are reviewed when applied in the context of multiple sclerosis studies.

Discovering the Relative Efficacy of Inhaled Medications for Chronic Obstructive Pulmonary Disease: Multiple Treatment Comparisons

Background: Chronic obstructive pulmonary disease (COPD) is managed by three major classes of inhaled medications: inhaled corticosteroids (ICS), long-acting Beta 2-agonist (LABA), long-acting muscarinic antagonist (LAMA). Single inhaled medication is usually replaced by combined inhaled medications for efficacy enhancement. However, this practice should be supported by clinical evidence for large-scale implementations. Methods: The relative efficacy of inhaled medications is determined by three endpoints: changes in the trough forced expiratory volume in 1 second (tFEV₁), changes in the St George’s Respiratory Questionnaire (SGRQ) score and the proportion of SGRQ responders which represents a reduction in SGRQ total score at week 24 of ≥4.0. A total of 76 eligible studies were identified in PubMed and Embase. Relevant data were extracted for the purpose of evidence synthesis. Then, raw mean differences (MD) and odds ratios (ORs) were produced by using the network meta-analysis. Results: Patients with ICS + LABA, ICS + LABA + LAMA, LABA, LABA + LAMA, LAMA exhibited significant increases in the tFEV1 compared to those with placebo (P < 0.05). Moreover, patients with ICS + LABA + LAMA exhibited the largest increase in the average tFEV1 and the largest decrease in the average SGRQ scores compared to those with placebo. COPD patients with ICS + LABA + LAMA were far more likely to achieve a significant reduction in the SGRQ scores compared to those with placebo or other inhaled medications (OR > 1). Conclusions: The combined inhaled medication of ICS + LABA + LAMA may be more efficacious than other inhaled medications for COPD patients.

Performing Arm-Based Network Meta-Analysis in R with the pcnetmeta Package.

Network meta-analysis is a powerful approach for synthesizing direct and indirect evidence about multiple treatment comparisons from a collection of independent studies. At present, the most widely used method in network meta-analysis is contrast-based, in which a baseline treatment needs to be specified in each study, and the analysis focuses on modeling relative treatment effects (typically log odds ratios). However, population-averaged treatment-specific parameters, such as absolute risks, cannot be estimated by this method without an external data source or a separate model for a reference treatment. Recently, an arm-based network meta-analysis method has been proposed, and the R package pcnetmeta provides user-friendly functions for its implementation. This package estimates both absolute and relative effects, and can handle binary, continuous, and count outcomes.

Therapeutic Efficacy Comparison of 5 Major EGFR-TKIs in Advanced EGFR-positive Non–Small-cell Lung Cancer: A Network Meta-analysis Based on Head-to-Head Trials

BackgroundFive major first- and second-generation epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs), including erlotinib, gefitinib, icotinib, afatinib, and dacomitinib, are currently optional for patients with advanced non–small-cell lung cancer (NSCLC) who harbor EGFR mutations. However, there was no head-to-head-based network meta-analysis among all the TKIs in EGFR-mutated populations. MethodsEligible literature was searched from an electronic database. Data of objective response rate, disease control rate, progression-free survival, and overall survival were extracted from enrolled studies. Multiple treatment comparisons based on Bayesian network integrated the efficacy of all included treatments. ResultsSix phase III randomized trials involving 1055 EGFR-mutated patients with advanced NSCLC were enrolled. Multiple treatment comparisons showed that 5 different EGFR-TKIs shared equivalent therapeutic efficacy in terms of all outcome measures. Rank probabilities indicated that dacomitinib and afatinib had potentially better efficacy compared with erlotinib, gefitinib, and icotinib in the EGFR-mutated patients. When compared with other agents, potential survival benefits (progression-free and overall survival) were observed in dacomitinib, whereas afatinib showed a better rank probability in overall response rate and disease control rate. ConclusionOur study indicated a preferable therapeutic efficacy in the second-generation TKIs (dacomitinib and afatinib) when compared with the first-generation TKIs (erlotinib, gefitinib, and icotinib).

Systematic Review and Network Meta-Analysis: Comparative Effectiveness of Transoral Incisionless Fundoplication(TIF)vs Laparascopic Nissen Fundoplication(Nissen)vs PPIs for the Treatment of Gastroesophageal Reflux Disease

Introduction: TIF has recently emerged as an effective therapy for GERD, however it is not recommended as first line therapy. Gold standard treatment for GERD is either long term PPIs or Nissen. While both TIF & Nissen have been compared with either PPIs or sham, head to head comparison of TIF vs NISSEN has not. We performed a systematic review and network meta-analysis of randomized controlled trials(RCTs) to assess the efficacy of TIF vs NISSEN for the treatment of GERD. Methods: Cochrane Central Register of Controlled Trials & MEDLINE from inception to May 1, 2016. Data were extracted independently by two authors. Methodological quality was assessed using the Cochrane risk of bias tool. A network meta-analysis was performed using the Bayesian methods under random-effects multiple treatment comparisons. Results were summarized as odds ratio along with credibility intervals. The ranking probability was assessed by surface under the cumulative ranking curve(SUCRA). To assess the impact of follow up time a meta-regression was performed on the Health Related Quality of Life(HRQOL). The Maude database was also reviewed for adverse events. Results: Overall 7 studies were included: 2 RCTs (n=173)TIF vs sham, 2 RCTs(n=120)TIF to PPIs, and 3 RCTs(n=835)Nissen vs PPI(Figure 1 summarizes these comparison). SUCRA ranking indicated TIF had the highest probability of symptomatic remission followed by Nissen, sham, then PPIs(0.96, 0.66, 0.35 and 0.042, respectively). Persistent esophagitis was highest in sham followed by TIF, Nissen, then PPIs(0.74, 0.69, 0.38 and 0.19). Nissen had the highest probability of achieving improved % time pH < 4 followed by PPIs, TIF, then sham(0.99, 0.64, 0.32, and 0.05), as well as increase in LES pressure followed by TIF then PPI(0.78, 0.72 and 0.01) Figure 2. Meta-regression showed a significant association between follow-up time and HRQOL for NISSEN but not TIF indicating follow up time was a significant confounder. Maude database revealed 54 adverse events with at least 28 hospitalizations and 14 esophageal perforations.Figure 1Figure 2Conclusion: Nissen has a higher ability to improve physiological parameters associated with GERD including improving LES pressure and % time pH < 4. PPIs were superior to all for reducing esophagitis. Although TIF appeared superior for achieving symptomatic remission this is likely related to shorter follow up time compared to Nissen and PPIs.

Efficacy and safety of TNF-α inhibitors for active ankylosing spondylitis patients: Multiple treatment comparisons in a network meta-analysis.

Ankylosing spondylitis (AS) is an inflammatory rheumatic disease with impact on axial skeleton, peripheral joints and enthuses, and it may result in severe disabilities of those parts. Tumor necrosis factor-α (TNF-α) inhibitors are considered as an effective treatment for patients with active AS. In this study, we conducted a network meta-analysis to compare the clinical outcomes of active AS patients treated with TNF-α inhibitors. Randomized controlled trials (RCTs) evaluating the efficacy and safety of TNF-α inhibitors were retrieved in literature search and selected for meta-analysis. Changes in ASAS20 response, ASAS40 response and BASDAI 50% response were regarded as efficacy outcomes; serious adverse events (SAE) and all cause withdrawals were regarded as safety outcomes. Both traditional pairwise meta-analysis and network meta-analysis were performed. The results showed that adalimumab and infliximab had better clinical outcomes. Infliximab consistently appeared to be the most effective TNF-α inhibitors with a high risk of adverse events for patients with active AS; meanwhile, adalimumab ranked highest with respect to adverse effects with efficacy secondary to infliximab. As a result, we were unable to conclude the optimal TNF-α inhibitor and this issue should be solved by future researchers.

Node-Splitting Generalized Linear Mixed Models for Evaluation of Inconsistency in Network Meta-Analysis.

Network meta-analysis for multiple treatment comparisons has been a major development in evidence synthesis methodology. The validity of a network meta-analysis, however, can be threatened by inconsistency in evidence within the network. One particular issue of inconsistency is how to directly evaluate the inconsistency between direct and indirect evidence with regard to the effects difference between two treatments. A Bayesian node-splitting model was first proposed and a similar frequentist side-splitting model has been put forward recently. Yet, assigning the inconsistency parameter to one or the other of the two treatments or splitting the parameter symmetrically between the two treatments can yield different results when multi-arm trials are involved in the evaluation. We aimed to show that a side-splitting model can be viewed as a special case of design-by-treatment interaction model, and different parameterizations correspond to different design-by-treatment interactions. We demonstrated how to evaluate the side-splitting model using the arm-based generalized linear mixed model, and an example data set was used to compare results from the arm-based models with those from the contrast-based models. The three parameterizations of side-splitting make slightly different assumptions: the symmetrical method assumes that both treatments in a treatment contrast contribute to inconsistency between direct and indirect evidence, whereas the other two parameterizations assume that only one of the two treatments contributes to this inconsistency. With this understanding in mind, meta-analysts can then make a choice about how to implement the side-splitting method for their analysis.

Extracorporeal shock wave therapy, ultrasound-guided percutaneous lavage, corticosteroid injection and combined treatment for the treatment of rotator cuff calcific tendinopathy: a network meta-analysis of RCTs.

Treatment of calcific tendinitis using extracorporeal shock wave therapy (ESWT), ultrasound-guided percutaneous lavage (UGPL or barbotage), subacromial corticosteroid injection (SAI) and combined treatment is still controversial. This systematic review and meta-regression aimed to compare clinical outcomes between treatments. Relevant RCTs were identified using PubMed and Scopus search engines to date of September 23, 2015. Seven of 920 studies identified were eligible. Compared to the other treatments, the results of this study indicate that ESWT significantly improved CMS and VAS when compared to placebo. Barbotage plus ESWT significantly improved CMS, VAS and decreased size of calcium deposit when compared to ESWT, while barbotage plus SAI significantly improved CMS and decreased size of calcium deposit when compared to SAI. There have no different adverse effects of all treatment groups. Multiple active treatment comparisons indicated that barbotage plus SAI significantly improved VAS and size of calcium deposit when compared to other groups, while barbotage plus SAI improved CMS when compared to other groups. But there was no significant difference. The network meta-analysis suggested that combined US-guided needling and subacromial corticosteroid injection significantly decreased shoulder pain VAS, improved CMS score and decreased the size of calcium deposits, while also lowering risks of adverse event when compared to barbotage plus ESWT, ESWT and subacromial corticosteroid injection; therefore, the evidence points to UGPL as being the treatment of choice for nonsurgical options of treatment in calcific tendinitis of the shoulder. Level of evidence I.

Comparative safety profile of hyaluronic acid products for knee osteoarthritis: a systematic review and network meta-analysis

Intra-articular (IA) hyaluronic acid (HA) is considered a safer alternative to oral Non-Steroidal Antiinflammatory Drugs (NSAIDs) and opioids for knee osteoarthritis (OA). A recent review raised potential safety concerns about HA, warranting further review of safety outcomes. We examined the risks of HA compared with IA placebo and investigated whether the risks vary among individual HA preparations. We searched all relevant databases from inception to October 2015 and sought unpublished data. We included all knee OA trials which compared any of the 18 HA products and reported on adverse events (AEs) and withdrawals. We calculated odds ratios for safety data reported at the longest follow-up. Network meta-analysis was performed using a Bayesian hierarchical random effects model for mixed multiple treatment comparisons. We identified 74 studies involving 13,032 participants aged between 45 and 75 years. The proportion of women ranged from 28% to 100%. The overall incidence of local reactions reported across all products was 8.5%. Commonly reported AEs were transient local reactions, such as pain, swelling and arthralgia, which subsided rapidly. None of the HA products were statistically significantly different from IA placebo or from each other with regard to incidence of AEs. Three treatment-related serious adverse events (SAEs) were reported among 9214 participants. Given the very low incidence of any particular AEs, we conclude that HA products are relatively well tolerated. These products have a similar safety profile compared to each other. This information along with the comparative effectiveness profile and relative cost would be helpful for clinicians in delivering individualized patient care.

Adaptive contrast weighted learning for multi-stage multi-treatment decision-making.

Dynamic treatment regimes (DTRs) are sequential decision rules that focus simultaneously on treatment individualization and adaptation over time. To directly identify the optimal DTR in a multi-stage multi-treatment setting, we propose a dynamic statistical learning method, adaptive contrast weighted learning. We develop semiparametric regression-based contrasts with the adaptation of treatment effect ordering for each patient at each stage, and the adaptive contrasts simplify the problem of optimization with multiple treatment comparisons to a weighted classification problem that can be solved by existing machine learning techniques. The algorithm is implemented recursively using backward induction. By combining doubly robust semiparametric regression estimators with machine learning algorithms, the proposed method is robust and efficient for the identification of the optimal DTR, as shown in the simulation studies. We illustrate our method using observational data on esophageal cancer.