Matching-Adjusted Indirect Comparisons of Lorlatinib Versus Chemotherapy for Patients With Second-Line or Later Anaplastic Lymphoma Kinase–Positive Non–Small Cell Lung Cancer

ObjectivesThis study aimed to compare the relative efficacy of lorlatinib, an anaplastic lymphoma kinase–tyrosine kinase inhibitor, with chemotherapy, for patients with second-line or later advanced anaplastic lymphoma kinase–positive non–small cell lung cancer. The endpoints of interest were overall survival (OS) and progression-free survival (PFS).MethodsEvidence for lorlatinib was informed by the single-arm phase I/II trial B7461001. A systematic literature review (SLR) was performed to identify OS and PFS data for chemotherapy. Unanchored matching-adjusted indirect comparisons (MAICs) between lorlatinib and chemotherapy (pemetrexed/docetaxel, platinum-based, or systemic therapy) were performed.ResultsThe SLR identified 3 relevant studies reporting PFS. Lorlatinib was associated with a significant decrease in the hazard of progression versus the 2 types of chemotherapy assessed. For PFS, the MAIC of lorlatinib versus the combined treatment arm of docetaxel or pemetrexed resulted in an adjusted hazard ratio (HR) of 0.22 (95% confidence interval [CI] 0.15-0.31). When lorlatinib was compared with platinum-based chemotherapy through an MAIC, the adjusted HR for PFS was 0.40 (95% CI 0.29-0.55). An exploratory comparison was performed for OS with evidence for systemic therapy (assumed equivalent to chemotherapy) not identified in the SLR. Lorlatinib provided a significant decrease in hazard of death (OS) versus systemic therapy, with HRs ranging from 0.12 (95% CI 0.05-0.27) to 0.43 (95% CI 0.27-0.60).ConclusionsLorlatinib demonstrated a significant improvement in PFS compared with chemotherapy, although limitations in the analyses were identified. The evidence informing OS comparisons was highly limited but suggested benefit of lorlatinib compared with systemic therapy.     

Evaluating Matching‐Adjusted Indirect Comparisons in Practice: A Case Study of Patients with Attention‐Deficit/Hyperactivity Disorder

Differences in patient characteristics across trials may bias efficacy estimates from indirect treatment comparisons. To address this issue, matching‐adjusted indirect comparison (MAIC) measures treatment efficacy after weighting individual patient data to match patient characteristics across trials. To date, however, there is no consensus on how best to implement MAIC. To address this issue, we applied MAIC to measure how two attention‐deficit/hyperactivity disorder (ADHD) treatments (guanfacine extended release and atomoxetine hydrochloride) affect patients' ADHD symptoms, as measured by the ADHD Rating Scale IV score. We tested MAIC sensitivity to: matched patient characteristics, matched statistical moments, weighting matrix, and placebo‐arm matching (i.e., matching on outcomes in the placebo arm). After applying MAIC, guanfacine and atomoxetine had similar reductions in ADHD symptoms (Δ: 0.4, p < 0.737). The results were similar for three of four sensitivity analyses. When we applied MAIC with placebo‐arm matching, however, guanfacine reduced symptoms more than atomoxetine (Δ: ?3.9, p < 0.004). We discuss the implication of this finding and advise MAIC practitioners to carefully consider the use of placebo‐arm matching, depending on the presence of residual confounding across trials. Copyright © 2016 John Wiley & Sons, Ltd.     

Assessing the performance of population adjustment methods for anchored indirect comparisons: A simulation study

Standard network meta-analysis and indirect comparisons combine aggregate data from multiple studies on treatments of interest, assuming that any factors that interact with treatment effects (effect modifiers) are balanced across populations. Population adjustment methods such as multilevel network meta-regression (ML-NMR), matching-adjusted indirect comparison (MAIC), and simulated treatment comparison (STC) relax this assumption using individual patient data from one or more studies, and are becoming increasingly prevalent in health technology appraisals and the applied literature. Motivated by an applied example and two recent reviews of applications, we undertook an extensive simulation study to assess the performance of these methods in a range of scenarios under various failures of assumptions. We investigated the impact of varying sample size, missing effect modifiers, strength of effect modification and validity of the shared effect modifier assumption, validity of extrapolation and varying between-study overlap, and different covariate distributions and correlations. ML-NMR and STC performed similarly, eliminating bias when the requisite assumptions were met. Serious concerns are raised for MAIC, which performed poorly in nearly all simulation scenarios and may even increase bias compared with standard indirect comparisons. All methods incur bias when an effect modifier is missing, highlighting the necessity of careful selection of potential effect modifiers prior to analysis. When all effect modifiers are included, ML-NMR and STC are robust techniques for population adjustment. ML-NMR offers additional advantages over MAIC and STC, including extending to larger treatment networks and producing estimates in any target population, making this an attractive choice in a variety of scenarios.     

The Effects of Model Misspecification in Unanchored Matching-Adjusted Indirect Comparison: Results of a Simulation Study

ObjectivesTo assess the performance of unanchored matching-adjusted indirect comparison (MAIC) by matching on first moments or higher moments in a cross-study comparisons under a variety of conditions. A secondary objective was to gauge the performance of the method relative to propensity score weighting (PSW).MethodsA simulation study was designed based on an oncology example, where MAIC was used to account for differences between a contemporary trial in which patients had more favorable characteristics and a historical control. A variety of scenarios were then tested varying the setup of the simulation study, including violating the implicit or explicit assumptions of MAIC.ResultsUnder ideal conditions and under a variety of scenarios, MAIC performed well (shown by a low mean absolute error [MAE]) and was unbiased (shown by a mean error [ME] of about zero). The performance of the method deteriorated where the matched characteristics had low explanatory power or there was poor overlap between studies. Only when important characteristics are not included in the matching did the method become biased (nonzero ME). Where the method showed poor performance, this was exaggerated if matching was also performed on the variance (ie, higher moments). Relative to PSW, MAIC provided similar results in most circumstances, although it exhibited slightly higher MAE and a higher chance of exaggerating bias.ConclusionsMAIC appears well suited to adjust for cross-trial comparisons provided the assumptions underpinning the model are met, with relatively little efficiency loss compared with PSW.     

Indirect Treatment Comparisons of Lenacapavir Plus Optimized Background Regimen Versus Other Treatments for Multidrug-Resistant Human Immunodeficiency Virus

Background/AimsHeavily treatment-experienced (HTE) people with human immunodeficiency virus (HIV) (PWH) may not achieve virologic suppression (VS) with combination antiretroviral therapy due to multidrug resistance (MDR), intolerance, and safety concerns. These PWH often receive highly individualized treatment regimens, but these regimens may not enable PWH to achieve VS, thereby halting disease progression. Novel medications are required for treating individuals with MDR HIV. Lenacapavir (LEN), a first-in-class HIV capsid inhibitor, is under investigation for the treatment of HTE individuals with MDR HIV in the phase 2/3 CAPELLA study. This study aimed to compare LEN plus optimized background regimen (OBR) with fostemsavir (FTR) + OBR, ibalizumab (IBA) + OBR, and OBR alone in terms of VS, CD4 cell count change from baseline, immunologic recovery, and discontinuation due to adverse events, using indirect treatment comparisons.MethodsA systematic review identified clinical evidence on HIV-1 treatments in HTE PWH. A feasibility assessment evaluated the identified studies for indirect treatment comparison analyses based on population characteristics, interventions, comparators, and outcomes of interest. Unanchored simulated treatment comparisons of LEN + OBR versus comparators were conducted.ResultsLEN + OBR had 6.57 times higher odds of VS at weeks 24 to 28 than FTR + OBR (95% confidence interval [CI] 1.34-32.28), 8.93 times higher odds of VS than IBA + OBR (95% CI 2.07-38.46), and 12.74 times higher odds of VS than OBR alone (95% CI 1.70-95.37). Change from baseline in CD4 cell count was similar across LEN + OBR, FTR + OBR, and IBA + OBR.ConclusionLEN + OBR has statistically significantly greater odds of VS at weeks 24 to 28 than its comparators and represents a novel treatment for people with MDR HIV.     

No Study Left Behind: A Network Meta-Analysis in Non–Small-Cell Lung Cancer Demonstrating the Importance of Considering All Relevant Data

Objective:  To demonstrate the importance of considering all relevant indirect data in a network meta-analysis of treatments for non–small-cell lung cancer (NSCLC).
Methods:  A recent National Institute for Health and Clinical Excellence appraisal focussed on the indirect comparison of docetaxel with erlotinib in second-line treatment of NSCLC based on trials including a common comparator. We compared the results of this analysis to a network meta-analysis including other trials that formed a network of evidence. We also examined the importance of allowing for the correlations between the estimated treatment effects that can arise when analysing such networks.
Results:  The analysis of the restricted network including only trials of docetaxel and erlotinib linked via the common placebo comparator produced an estimated mean hazard ratio (HR) for erlotinib compared with docetaxel of 1.55 (95% confidence interval [CI] 0.72–2.97). In contrast, the network meta-analysis produced an estimated HR for erlotinib compared with docetaxel of 0.83 (95% CI 0.65–1.06). Analyzing the wider network improved the precision of estimated treatment effects, altered their rankings and also allowed further treatments to be compared. Some of the estimated treatment effects from the wider network were highly correlated.
Conclusions:  This empirical example shows the importance of considering all potentially relevant data when comparing treatments. Care should therefore be taken to consider all relevant information, including correlations induced by the network of trial data, when comparing treatments.     

Radical Cystectomy versus Bladder-Preserving Therapy for Muscle-Invasive Urothelial Carcinoma: Examining Confounding and Misclassification Biasin Cancer Observational Comparative Effectiveness Research

ObjectivesRadical cystectomy (RC) is the standard treatment for muscle-invasive urothelial carcinoma of the bladder. Trimodality bladder-preserving therapy (BPT) is an alternative to RC, but randomized comparisons of RC versus BPT have proven infeasible. To compare RC versus BPT, we undertook an observational cohort study using registry and administrative claims data from the Surveillance, Epidemiology and End Results-Medicare database.MethodsWe identified patients age 65 years or older diagnosed between 1995 and 2005 who received RC (n = 1426) or BPT (n = 417). We examined confounding and stage misclassification in the comparison of RC and BPT by using multivariable adjustment, propensity score–based adjustment, instrumental variable (IV) analysis, and simulations.ResultsPatients who received BPT were older and more likely to have comorbid disease. After propensity score adjustment, BPT was associated with an increased hazard of death from any cause (hazard ratio [HR] 1.26; 95% confidence interval [CI] 1.05–1.53) and from bladder cancer (HR 1.31; 95% CI 0.97–1.77). Using the local area cystectomy rate as an instrument, IV analysis demonstrated no differences in survival between BPT and RC (death from any cause HR 1.06; 95% CI 0.78–1.31; death from bladder cancer HR 0.94; 95% CI 0.55–1.18). Simulation studies for stage misclassification yielded results consistent with the IV analysis.ConclusionsSurvival estimates in an observational cohort of patients who underwent RC versus BPT differ by analytic method. Multivariable and propensity score adjustment revealed greater mortality associated with BPT relative to RC, while IV analysis and simulation studies suggest that the two treatments are associated with similar survival outcomes.     

Matching-Adjusted Indirect Comparisons: A New Tool for Timely Comparative Effectiveness Research

ObjectiveIn 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.MethodsMatching-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.ResultsExample 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.ConclusionsIndirect 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.     

Alternative Weighting Approaches for Anchored Matching-Adjusted Indirect Comparisons via a Common Comparator

Background

Adjusted indirect comparisons (anchored via a common comparator) are an integral part of health technology assessment. These methods are challenged when differences between studies exist, including inclusion/exclusion criteria, outcome definitions, patient characteristics, as well as ensuring the choice of a common comparator.

Extending Treatment Networks in Health Technology Assessment: How Far Should We Go?

BackgroundNetwork meta-analysis may require substantially more resources than does a standard systematic review. One frequently asked question is “how far should I extend the network and which treatments should I include?”ObjectiveTo explore the increase in precision from including additional evidence.MethodsWe assessed the benefit of extending treatment networks in terms of precision of effect estimates and examined how this depends on network structure and relative strength of additional evidence. We introduced a “star”-shaped network. Network complexity is increased by adding more evidence connecting treatments under five evidence scenarios. We also examined the impact of heterogeneity and absence of evidence facilitating a “first-order” indirect comparison.ResultsIn all scenarios, extending the network increased the precision of the A versus B treatment effect. Under a fixed-effect model, the increase in precision was modest when the existing direct A versus B evidence was already strong and was substantial when the direct evidence was weak. Under a random-effects model, the gain in precision was lower when heterogeneity was high. When evidence is available for all “first-order” indirect comparisons, including second-order evidence has limited benefit for the precision of the A versus B estimate. This is interpreted as a “ceiling effect.”ConclusionsIncluding additional evidence increases the precision of a “focal” treatment comparison of interest. Once the comparison of interest is connected to all others via “first-order” indirect evidence, there is no additional benefit in including higher order comparisons. This conclusion is generalizable to any number of treatment comparisons, which would then all be considered “focal.” The increase in precision is modest when direct evidence is already strong, or there is a high degree of heterogeneity.     

Critical Appraisal of Published Indirect Comparisons and Network Meta-Analyses of Competing Interventions for Multiple Myeloma

ObjectivesIn the field of relapsed or refractory multiple myeloma (RRMM), between-trial or indirect comparisons are required to estimate relative treatment effects between competing interventions based on the available evidence. Two approaches are frequently used in RRMM: network meta-analysis (NMA) and unanchored matching-adjusted indirect comparison (MAIC). The objective of the current study was to evaluate the relevance and credibility of published NMA and unanchored MAIC studies aiming to estimate the comparative efficacy of treatment options for RRMM.MethodsTwelve relevant studies were identified in the published literature (n = 7) and from health technology assessment agencies (n = 5). Data from trials were extracted to identify between-trial differences that may have biased results. Credibility of the performed analyses and relevance of the research questions were critically appraised using the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) checklist and feedback based on consultations with clinical experts.ResultsThe identified studies concerned NMAs of randomized controlled trials (RCTs; n = 7), unanchored MAICs (n = 4), or both types of analyses (n = 1). According to clinical expert consultation, the majority of the identified NMAs did not consider differences in prior therapies or treatment duration across the RCTs included in the analyses, thereby compromising the relevance.ConclusionBased on the results and feedback from clinicians, the majority of NMAs did not consider prior treatment history or treatment duration, which resulted in nonrelevant comparisons. Furthermore, it may have compromised the credibility of the estimates owing to differences in effect-modifiers between the different trials. Pairwise comparisons by means of unanchored MAICs require clear justification given the reliance on non-randomized comparisons.     

Skeletal Muscle Mass Predicts Poor Prognosis in Patients with Advanced Pancreatic Cancer Undergoing Second-Line FOLFIRINOX Chemotherapy

Objectives: Although the significance of skeletal muscle mass has been investigated in pancreatic cancer, there are no reports regarding the impact of skeletal muscle mass on prognosis in patients who have undergone second-line chemotherapy. We aimed to identify prognostic factors in patients with advanced pancreatic cancer treated with second-line FOLFIRINOX (folinic acid, fluorouracil, irinotecan, and oxaliplatin).

Methods: We retrospectively reviewed the data of 57 pancreatic cancer patients treated with second-line FOLFIRINOX. Age, sex, body mass index, Eastern Cooperative Oncology Group (ECOG) performance status, carbohydrate antigen 19-9 levels, skeletal muscle area, skeletal muscle index (SMI), progression free survival (PFS), and overall survival (OS) were analyzed.

Results: The median age of the 57 patients (male, 56.1%) was 60.4?years (38–78). Median PFS and OS were 2.6 and 6.6?months. On Kaplan-Meier curves, high SMI was associated with prolonged OS and PFS (P value?=?0.003 and 0.015). In multivariate analysis, baseline SMI was significant independent prognostic factor in patients treated with second-line FOLFIRINOX.

Conclusion: Baseline SMI has an impact on prognosis in patients who undergoing second-line chemotherapy for pancreatic cancer. Skeletal muscle mass may warrant consideration as a predictive factor with which to identify candidates for second-line chemotherapy for advanced pancreatic cancer.     

Relationship Between Oral Health and Fractures in Community-Dwelling Older Japanese Adults

ObjectiveTo investigate the relationship between poor oral health and the incidence of fall-related fractures in older Japanese individuals.DesignA 9-year prospective cohort study.Setting and ParticipantsParticipants comprised 937 community-dwelling older Japanese adults aged 70 years or older. They all lived in the Tsurugaya district, a suburban area of Sendai city, and underwent comprehensive geriatric assessment, including an oral examination, in a public facility.MeasurementsThe exposure variables were related to oral health status (posterior occlusal support, number of remaining teeth, and occlusal force). The outcome measure was the incidence of fall-related fractures, which was determined by National Health Insurance data. Analyzed covariates included age, sex, medical history, smoking, alcohol drinking, educational level, depressive symptoms, cognitive impairment, physical function, body mass index, and history of falls. Statistical relationships were examined by calculating hazard ratios (HRs) at 95% confidence intervals (CIs) using the Cox proportional hazard model.ResultsIn the multivariate analysis, the HRs of fall-related fractures were significantly higher in those with unilateral posterior occlusal support (HR, 2.72; 95% CI, 1.13-6.55) and no posterior occlusal support (HR, 2.58; 95% CI, 1.29-5.15) than in those with bilateral posterior occlusal support. The HRs (95% CIs) of fall-related fractures in individuals with 10-19 and 1-9 teeth and edentulous individuals were 1.77 (0.81-3.89), 2.67 (1.24-5.75), and 2.31 (1.01-5.28), respectively, compared to those with ≥20 teeth.Conclusions and ImplicationsPoor oral health status is a risk factor for the incidence of fall-related fractures in community-dwelling older Japanese individuals. The findings suggest that attention should be focused on oral health status to further understand the risk of fall-related fractures among community-dwelling older adults.     

Prognostic Value of Prealbumin in Liver Cancer: A Systematic Review and Meta-Analysis

Abstract

Accumulated studies have reported the prognostic significance of prealbumin in liver cancer, but the results were not conclusive. The aim of this study was to evaluate the association between pretreatment serum prealbumin and clinical outcome of liver cancer patients through a meta-analysis. We comprehensively searched EMBASE, PubMed, Web of Science and the Cochrane library to identify eligible studies. The pooled hazard ratios (HRs) and their 95% confidence intervals (CIs) were utilized to evaluate the prognostic value of pretreatment serum prealbumin in overall survival (OS) and recurrence-free survival (RFS) of liver cancer patients. A total of 3470 patients from 10 eligible studies were finally included for analysis. The combined effects of prealbumin on liver cancer patients’ OS and RFS were HR?=?1.83, 95% CI: 1.46–2.30, P?<?0.001 and HR?=?1.47, 95% CI: 1.01–2.14, P?=?0.045, respectively. Sensitivity and subgroup analysis showed that the pooled HR of prealbumin on liver cancer patients’ OS was stable. Since potential publication bias was identified in the OS studies, the trim-and-fill method therefore was performed to explore publication bias, and the results showed reliability. This meta-analysis shows that low pretreatment serum prealbumin is significantly associated with poor prognosis of liver cancer patients.     

Comparison of treatment effect estimates of non-vitamin K antagonist oral anticoagulants versus warfarin between observational studies using propensity score methods and randomized controlled trials

Emerging observational studies using propensity score (PS) methods assessed real-world comparative effectiveness of non-vitamin K antagonist oral anticoagulants (NOACs) versus warfarin in patients with non-valvular atrial fibrillation (AF). We aimed to compare treatment effect estimates of NOACs between PS studies and randomized controlled trials (RCTs). Electronic databases and conference proceedings were searched systematically. Primary outcomes included stroke or systemic embolism (SE) and major bleeding. A random-effects meta-analysis was performed to synthesize the data by pooling the PS- and RCT-derived hazard ratios (HRs) separately. The ratio of HRs (RHR) from the ratio of PS-derived HRs relative to RCT-derived HRs was used to determine whether there was a difference between estimates from PS studies and RCTs. There were 10 PS studies and 5 RCTs included for analysis. No significant difference of treatment effect estimates between the PS studies and RCTs was observed: RHR 1.11, 95 % CI 0.98–1.23 for stroke or SE; RHR 1.07, 95 % CI 0.87–1.34 for major bleeding. A significant association between NOACs and risk of stroke or SE was observed: HR 0.88, 95 % CI 0.83–0.94 for the PS studies; HR 0.79, 95 % CI 0.72–0.87 for the RCTs. However, no relationship between NOACs and risk of major bleeding was found: HR 0.91, 95 % CI 0.79–1.05 for the PS studies; HR 0.85, 95 % CI 0.73–1.00 for the RCTs. In this study, treatment effect estimates of NOACs versus warfarin in patients with non-valvular AF from PS studies are found to be in agreement with those from RCTs.     

A Preliminary Study on a Form of the 24-h Recall That Balances Survey Cost and Accuracy,Based on the NCI Method

The 24-h recall (24HR) is a short-term dietary assessment instrument that is widely used in large-scale nutrition surveys. The number of survey days is critical in the accuracy of estimates. The multiple, repeated collection of 24HRs can yield reliable dietary intakes, whereas that is not always feasible due to staffing, equipment, financial, and temporal constraints. The NCI (National Cancer Institute) method was developed to address this limitation by using only within-person variance to calculate usual dietary intake. However, the performance of different forms of 24HRs based on the NCI method remains unclear. The aim of this study was to explore a form of 24HR based on the NCI method that can balance accuracy and survey cost. A total of 595 subjects completed 7 consecutive 24HRs in each season, for a total of 28 24HRs. The averages of the 28 collection days were defined as the reference value to compare the performance of 24HRs for two consecutive days (C2), three consecutive days (C3), two non-consecutive days (NC2), and three non-consecutive days (NC3) for estimating the dietary intakes of Chinese adults. The equivalence test was used to evaluate whether the estimates of scenarios NC2 and NC3 were equivalent. Additionally, the accuracy of a scenario of NC2 which included a weekend was compared to that of a scenario of NC2 which included two weekdays. All results of the 24HRs in each scenario were corrected by the NCI method. Bias/relative bias and mean bias/mean relative bias were used as measures of precision and accuracy, respectively. The results showed that the precision was similar among the four scenarios, while the accuracy relationship varied among the different dietary components. In general, scenario NC3 was the most accurate, followed by scenario NC2, which was close to the former. The form using non-consecutive days was more accurate than that using consecutive days, and the main factor affecting the accuracy of the 24HRs was the continuity between multiple survey days rather than the number of days. The means and major percentiles of energy, nutrients, and frequently consumed food in scenarios NC2 and NC3 were functionally identical. The accuracy of the scenario of NC2 which included a weekend was higher than that of scenario NC2, which consisted of only weekdays. The above results indicated that the adoption of two, non-consecutive 24HRs consisting of a weekend and a weekday to collect dietary data prior to correction by the NCI method, is a feasible approach to balancing survey costs and accuracy in large-scale nutrition surveys.     

Network meta-regression for ordinal outcomes: Applications in comparing Crohn's disease treatments

Crohn's disease (CD) is a life-long condition associated with recurrent relapses characterized by abdominal pain, weight loss, anemia, and persistent diarrhea. In the US, there are approximately 780 000 CD patients and 33 000 new cases added each year. In this article, we propose a new network meta-regression approach for modeling ordinal outcomes in order to assess the efficacy of treatments for CD. Specifically, we develop regression models based on aggregate covariates for the underlying cut points of the ordinal outcomes as well as for the variances of the random effects to capture heterogeneity across trials. Our proposed models are particularly useful for indirect comparisons of multiple treatments that have not been compared head-to-head within the network meta-analysis framework. Moreover, we introduce Pearson residuals and construct an invariant test statistic to evaluate goodness-of-fit in the setting of ordinal outcome data. A detailed case study demonstrating the usefulness of the proposed methodology is carried out using aggregate ordinal outcome data from 16 clinical trials for treating CD.     

Citrus Consumption and Risk of Cutaneous Malignant Melanoma in the Women’s Health Initiative

Abstract

Citrus products are rich sources of furocoumarins, a class of photoactive compounds. Certain furocoumarins combined with ultraviolet radiation can induce skin cancer. We examined the relationship between citrus consumption and cutaneous melanoma risk among 56,205 Caucasian postmenopausal women in the Women’s Health Initiative. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of melanoma by citrus intake level. During a mean follow-up of 15.7?years, 956 incident melanoma cases were documented. In multivariable adjusted models, the HR (95% CI) for melanoma was 1.12 (0.91, 1.37) among the highest citrus consumers (1.5+ servings/day of fruit or juice) versus the lowest (<2 servings/week), 0.95 (0.76, 1.20) among the highest citrus fruit consumers (5+ servings/week) versus non-consumers, and was 1.13 (0.96, 1.32) for the highest citrus juice consumers (1+ servings/day) versus the lowest (<1 serving/week). In stratified analyses, an increased melanoma risk associated with citrus juice intake was observed among women who spent the most time outdoors in summer as adults; the HR for the highest versus lowest intake was 1.22 (1.02, 1.46) (p trend = 0.03). Further research is needed to explore the association of melanoma with citrus juices among women with high sun exposure.     

Prognostic Value of the Geriatric Nutritional Risk Index in Patients Exceeding 70 Years Old with Esophageal Squamous Cell Carcinoma

Abstract

To investigate the prognostic value of the Geriatric Nutritional Risk Index (GNRI) in esophageal squamous cell carcinoma (ESCC) patients treated with radiotherapy (RT) or definitive concurrent chemoradiotherapy (dCRT). Fifty-two ESCC patients were included from July 2014 to December 2018. RT was delivered at a dose of 1.8–2.0?Gy per day to a total dose of 50–60?Gy. Tumor response was assessed using the RECIST 1.1 system. Overall survival (OS) and progression-free survival (PFS) were calculated and compared with the Kaplan–Meier method. Multivariate analysis of predictive factors of response and survival was performed using a logistic regression and a Cox model, respectively. In multivariate analysis, GNRI score (HR 0.278, P?=?0.036) was the only independent prognostic factor for tumor response. As for survival outcomes, GNRI score (OS: HR 0.505, P?=?0.028; PFS: HR 0.583, P?=?0.045) and treatment modality (OS: HR 0.356, P?=?0.015; PFS: HR 0.392, P?=?0.0014) were both independent prognostic factors for better OS and PFS. Additionally, there was no correlation between GNRI score and treatment modality (Spearman’s ρ?=?0.200; P?=?0.154). In conclusion, routine use of the GNRI criteria may help in the risk stratification of elderly patients undergoing RT/dCRT. The dCRT treatment could provide survival benefits for elderly ESCC patients.     

Narcolepsy and hypersomnia in Norwegian children and young adults following the influenza A(H1N1) 2009 pandemic

BackgroundAssociations between influenza infection and sleep disorders are poorly studied. We investigated if pandemic influenza infection or vaccination with Pandemrix in 2009/2010 was associated with narcolepsy or hypersomnia in children and young adults.MethodsWe followed the Norwegian population under age 30 from January 2008 through December 2012 by linking national health registry data. Narcolepsy diagnoses were validated using hospital records. Risks of narcolepsy or hypersomnia were estimated as adjusted hazard ratios (HRs) in Cox regression models with influenza infection and vaccination as time-dependent exposures.ResultsAmong the 1,638,526 persons under age 30 in Norway in 2009, 3.6% received a physician diagnosis of influenza during the pandemic, while 41.9% were vaccinated against pandemic influenza. Between October 1st 2009 and December 31st 2012, 72 persons had onset of narcolepsy and 305 were diagnosed with hypersomnia. The risk of a sleep disorder was associated with infection during the first six months, adjusted HR 3.31 with 95% confidence interval [CI], 1.01–10.79 for narcolepsy and adjusted HR 3.13 (95% CI, 1.12–8.76) for hypersomnia. The risk of narcolepsy was strongly associated with vaccination during the first six months adjusted HR 17.21 (95% CI, 6.28–47.14), while the adjusted HR for hypersomnia was 1.54 (95% CI, 0.81–2.93).ConclusionsThe study confirms an increased HR of narcolepsy following pandemic vaccination. Slightly increased HRs of narcolepsy and hypersomnia are also seen after influenza infection. However, the role of infection should be viewed with caution due to underreporting of influenza.