Pre-specified Significance Level Research Articles

Association of RNA-sequencing signatures with clinical outcomes of pembrolizumab + chemotherapy in patients with metastatic nonsquamous non–small cell lung cancer (NSCLC) enrolled in the phase 2 KEYNOTE-782 trial.

8578 Background: KEYNOTE-782 (NCT03664024) was a single-arm phase 2 study designed to assess possible biomarkers of response (objective response rate [ORR]) to first-line pembrolizumab + chemotherapy (pemetrexed + carboplatin or cisplatin) in patients with previously untreated metastatic nonsquamous NSCLC. Herein, we examined relationships between the T-cell–inflamed gene expression profile (TcellinfGEP) and other tumor microenvironment consensus signatures and efficacy of pembrolizumab + chemotherapy in an exploratory analysis of KEYNOTE-782. Methods: All patients were to receive pembrolizumab 200 mg Q3W, pemetrexed 500 mg/m2 Q3W, and 4 Q3W cycles of carboplatin AUC 5 mg/mL/min or cisplatin 75 mg/m2. Using tumor samples, RNA sequencing was used to measure expression of the TcellinfGEP and 10 non-TcellinfGEP signatures (angiogenesis, gMDSC, glycolysis, hypoxia, mMDSC, MYC, proliferation, RAS, stromal/EMT/TGFβ, and WNT). The association between each signature and clinical outcomes was analyzed using logistic regression (ORR) and Cox proportional hazards regression (progression-free survival [PFS] and overall survival [OS]); the prespecified significance level was α = 0.05 for TcellinfGEP (hypothesized positive association) and α = 0.10 for non-TcellinfGEP signatures (hypothesized negative associations). The clinical utility of TcellinfGEP was descriptively assessed using a prespecified cutoff of the first tertile. The clinical database cutoff was November 5, 2021. Results: Of 117 patients enrolled, 69 (59.0%) had evaluable RNA-sequencing data. TcellinfGEP was not associated with ORR ( P = 0.183), PFS ( P = 0.071), or OS ( P = 0.142); the area under the receiver operating characteristic curve for discriminating response was 0.56 (95% CI, 0.42-0.71). None of the 10 non-TcellinfGEP consensus signatures showed a statistically significant association with clinical outcomes after adjusting for TcellinfGEP (multiplicity adjusted P > 0.10). ORR and median PFS were comparable in the TcellinfGEP low and nonlow subgroups; there was a trend towards longer OS in the nonlow subgroup (Table). Conclusions: In this exploratory analysis of patients with metastatic nonsquamous NSCLC, none of the RNA-sequencing signatures evaluated were associated with clinical outcomes of pembrolizumab + chemotherapy. There was little evidence of clinical utility of TcellinfGEP when evaluated as a dichotomous variable. Data support the use of pembrolizumab + chemotherapy as first-line therapy for patients with nonsquamous NSCLC regardless of consensus signature status. Clinical trial information: NCT03664024 . [Table: see text]

LFK index does not reliably detect small-study effects in meta-analysis: A simulation study.

The LFK index has been promoted as an improved method to detect bias in meta-analysis. Putatively, its performance does not depend on the number of studies in the meta-analysis. We conducted a simulation study, comparing the LFK index test to three standard tests for funnel plot asymmetry in settings with smaller or larger group sample sizes. In general, false positive rates of the LFK index test markedly depended on the number and size of studies as well as the between-study heterogeneity with values between 0% and almost 30%. Egger's test adhered well to the pre-specified significance level of 5% under homogeneity, but was too liberal (smaller groups) or conservative (larger groups) under heterogeneity. The rank test was too conservative for most simulation scenarios. The Thompson-Sharp test was too conservative under homogeneity, but adhered well to the significance level in case of heterogeneity. The true positive rate of the LFK index test was only larger compared with classic tests if the false positive rate was inflated. The power of classic tests was similar or larger than the LFK index test if the false positive rate of the LFK index test was used as significance level for the classic tests. Under ideal conditions, the false positive rate of the LFK index test markedly and unpredictably depends on the number and sample size of studies as well as the extent of between-study heterogeneity. The LFK index test in its current implementation should not be used to assess funnel plot asymmetry in meta-analysis.

Pooled Exposure-Response (ER) and Quantitative Benefit/Risk (B/R) Analyses Support the Approved Copanlisib Intermittent Dosing Regimen

Introduction: Copanlisib is a potent and highly selective pan-class phosphatidylinositol 3-kinase (PI3K) inhibitor with preferential activity against the p110α and p110δ isoforms. Unlike other PI3K inhibitors, copanlisib is uniquely administered intravenously using an intermittent schedule, supported by preclinical data and literature model-based analyses showing high tumor residence and time intervals enabling for recovery of normal tissue. The pivotal studies, CHRONOS-1 and CHRONOS-3, established the benefit/risk of copanlisib 60 mg on Day 1, 8, 15 of 28-day cycle as monotherapy in 3L+ relapsed follicular lymphoma and in combination with rituximab in 2L+ non-Hodgkin's lymphoma, respectively. While prior ER analyses in CHRONOS-1 suggested a positive ER trend for copanlisib safety, similar to oral continuously dosed PI3K inhibitors (2022 FDA ODAC), ER analyses from the larger Phase 3 study, CHRONOS-3, revealed no significant ER for safety (Morcos et al. CPT Pharmacometrics Syst Pharmacol 2023). This analysis investigates pooled ER relationships for copanlisib safety across the pivotal studies and considers them in the context of the ER for efficacy in CHRONOS-3, as part of a quantitative B/R analysis for the approved copanlisib intermittent dosing regimen. Methods: Individual static and time-varying exposure metrics in CHRONOS-1 and CHRONOS-3 were derived from a previously established comprehensive population pharmacokinetics (popPK) analysis (Morcos et al. CPT Pharmacometrics Syst Pharmacol 2023). ER analyses investigated the relationship to the efficacy endpoint: progression-free survival (PFS) in CHRONOS-3 using 2-year follow-up; and to the safety endpoints: time-to-first Grade ≥ 3 adverse events (Gr≥3 AEs) and serious adverse events (SAEs), and frequency of individual AEs associated with copanlisib treatment (hyperglycemia, diarrhea, hypertension, nausea, fatigue, neutropenia, and lung infections of any grade), pooled across CHRONOS-1 using 6-year follow-up and CHRONOS-3 using 2-year follow-up. Multivariate parametric time-to-event (TTE) and logistic regression models investigated ER relationships at a prespecified significance level of p<0.01 after accounting for other potentially prognostic demographic, laboratory, and disease related baseline covariates. To quantitatively assess the B/R of the copanlisib 60 mg intermittent dosing regimen, a clinical utility index (CUI) function was defined with a 1:1 weighting for efficacy and safety to consider the tradeoffs between any identified ER relationships for efficacy and safety. Results: In CHRONOS-3, the parametric TTE ER analysis for PFS was characterized by a 2-knot spline baseline hazard function and demonstrated a statistically significant positive ER relationship (p<0.00001). Analysis results are consistent with a prior ER analysis using a semi-parametric Cox proportional hazards model of earlier 1-year follow-up (Morcos et al. CPT Pharmacometrics Syst Pharmacol 2023). Therefore, greater copanlisib exposure is associated with more prolonged PFS. Across pooled CHRONOS-1 and CHRONOS-3, parametric TTE for first Gr≥3 AEs and SAEs were characterized by 2-knot spline baseline hazard functions and demonstrated a borderline significant ER for Gr≥3 AEs (p=0.014) and no significant ER for SAEs (p=0.26). Multivariate logistic regression analyses showed no significant ER for individual AEs after adjusting for relevant baseline covariates. Across the range of exposures achieved with the copanlisib 60 mg intermittent dosing regimen, a steeper relationship for the hazard ratio relative to placebo in is seen for the significant ER of PFS compared with the borderline ER of Gr≥3 AEs (Figure 1). The CUI accounting for the quantitative ER relationships confirmed the favorable B/R across the range of copanlisib exposures. Thus, lower copanlisib doses may result in lower efficacy while not necessarily resulting in substantially improved safety. Conclusions: The completed quantitative analyses substantiate the favorable B/R for the copanlisib 60 mg on days 1, 8, and 15 of a 28-day cycle dosing regimen.

A Case Study of 2-Stage Seamless Adaptive Sample Size Re-Estimation Design with Efficacy Interim Analysis When Slope Is the Primary Endpoint

Due to highly unmet medical needs in rare disease areas, there is great desire to speed up the drug development process. A two-stage adaptive design option for a placebo-controlled registration study is being evaluated. Stage 1 consists of participants in an ongoing phase 2 study with 1-year double blinded (DB) treatment and Stage 2 includes newly enrolled participants with 2-year DB treatment period. The primary endpoint is the annualized rate of change (slope) for a continuous longitudinal measurement, which will be evaluated through a random coefficient linear mixed model. An unblinded interim analysis will be performed using Stage 1 data to re-estimate the sample size for Stage 2, followed by another interim analysis for potential early efficacy stopping when all participants completed the 1-year DB treatment. To control the overall Type 1 error rate, rather than using a conservative approach, the actual correlation between the interim and final test statistics will be taken into account to determine the final significance level given the pre-specified significance level for the interim efficacy analysis. Multiplicity adjustments for secondary endpoints including considerations for order switching between interim and final analysis in this specific case study will also be discussed.

The Smooth Away From Expected (SAFE) non-inferiority frontier: theory and implementation with an application to the D3 trial

BackgroundIn a non-inferiority trial, the choice of margin depends on the expected control event risk. If the true risk differs from expected, power and interpretability of results can be affected. A non-inferiority frontier pre-specifies an appropriate non-inferiority margin for each value of control event risk. D3 is a non-inferiority trial comparing two treatment regimens in children living with HIV, designed assuming a control event risk of 12%, a non-inferiority margin of 10%, 80% power and a significance level (α) of 0.025. We consider approaches to choosing and implementing a frontier for this already funded trial, where changing the sample size substantially would be difficult.MethodsIn D3, we fix the non-inferiority margin at 10%, 8% and 5% for control event risks of ≥9%, 5% and 1%, respectively. We propose four frontiers which fit these fixed points, including a Smooth Away From Expected (SAFE) frontier. Analysis approaches considered are as follows: using the pre-specified significance level (α=0.025); always using a reduced significance level (to achieve α≤0.025 across control event risks); reducing significance levels only when the control event risk differs significantly from expected (control event risk <9%); and using a likelihood ratio test. We compare power and type 1 error for SAFE with other frontiers.ResultsChanging the significance level only when the control event risk is <9% achieves approximately nominal (<3%) type I error rate and maintains reasonable power for control event risks between 1 and 15%. The likelihood ratio test method performs similarly, but the results are more complex to present. Other analysis methods lead to either inflated type 1 error or badly reduced power. The SAFE frontier gives more interpretable results with low control event risks than other frontiers (i.e. it uses more reasonable non-inferiority margins). Other frontiers do not achieve power close (i.e. within 1%) to SAFE across the range of likely control event risks while controlling type I error. ConclusionsThe SAFE non-inferiority frontier will be used in D3, and the non-inferiority margin and significance level will be modified if the control event risk is lower than expected. This ensures results will remain interpretable if design assumptions are incorrect, while achieving similar power. A similar approach could be considered for other non-inferiority trials where the control event risk is uncertain.

Genome-wide association study of treatment-related toxicity two years following radiotherapy for breast cancer

Background and purposeUp to a quarter of breast cancer patients treated by surgery and radiotherapy experience clinically significant toxicity. If patients at high risk of adverse effects could be identified at diagnosis, their treatment could be tailored accordingly. This study was designed to identify common single nucleotide polymorphisms (SNPs) associated with toxicity two years following whole breast radiotherapy. Materials and MethodsA genome-wide association study (GWAS) was performed in 1,640 breast cancer patients with complete SNP, clinical, treatment and toxicity data, recruited across 18 European and US centres into the prospective REQUITE cohort study. Toxicity data (CTCAE v4.0) were collected at baseline, end of radiotherapy, and annual follow-up. A total of 7,097,340 SNPs were tested for association with the residuals of toxicity endpoints, adjusted for clinical, treatment co-variates and population substructure. ResultsQuantile-quantile plots showed more associations with toxicity above the p < 5 × 10-5 level than expected by chance. Eight SNPs reached genome-wide significance. Nipple retraction grade ≥ 2 was associated with the rs188287402 variant (p = 2.80 × 10-8), breast oedema grade ≥ 2 with rs12657177 (p = 1.12 × 10-10), rs75912034 (p = 1.12 × 10-10), rs145328458 (p = 1.06 × 10-9) and rs61966612 (p = 1.23 × 10-9), induration grade ≥ 2 with rs77311050 (p = 2.54 × 10-8) and rs34063419 (p = 1.21 × 10-8), and arm lymphoedema grade ≥ 1 with rs643644 (p = 3.54 × 10-8). Heritability estimates across significant endpoints ranged from 25% to 39%. Our study did not replicate previously reported SNPs associated with breast radiation toxicity at the pre-specified significance level. ConclusionsThis GWAS for long-term breast radiation toxicity provides further evidence for significant association of common SNPs with distinct toxicity endpoints.

Astaxanthin, a natural antioxidant, lowers cholesterol and markers of cardiovascular risk in individuals with prediabetes and dyslipidaemia.

To determine the effects of astaxanthin treatment on lipids, cardiovascular disease (CVD) markers, glucose tolerance, insulin action and inflammation in individuals with prediabetes and dyslipidaemia. Adult participants with dyslipidaemia and prediabetes (n = 34) underwent baseline blood draw, an oral glucose tolerance test and a one-step hyperinsulinaemic-euglycaemic clamp. They were then randomized (n = 22 treated, 12 placebo) to receive astaxanthin 12 mg daily or placebo for 24 weeks. Baseline studies were repeated after 12 and 24 weeks of therapy. After 24 weeks, astaxanthin treatment significantly decreased low-density lipoprotein (-0.33 ± 0.11 mM) and total cholesterol (-0.30 ± 0.14 mM) (both P < .05). Astaxanthin also reduced levels of the CVD risk markers fibrinogen (-473 ± 210 ng/mL), L-selectin (-0.08 ± 0.03 ng/mL) and fetuin-A (-10.3 ± 3.6 ng/mL) (all P < .05). While the effects of astaxanthin treatment did not reach statistical significance, there were trends toward improvements in the primary outcome measure, insulin-stimulated, whole-body glucose disposal (+0.52 ± 0.37 mg/m2 /min, P = .078), as well as fasting [insulin] (-5.6 ± 8.4 pM, P = .097) and HOMA2-IR (-0.31 ± 0.16, P = .060), suggesting improved insulin action. No consistent significant differences from baseline were observed for any of these outcomes in the placebo group. Astaxanthin was safe and well tolerated with no clinically significant adverse events. Although the primary endpoint did not meet the prespecified significance level, these data suggest that astaxanthin is a safe over-the-counter supplement that improves lipid profiles and markers of CVD risk in individuals with prediabetes and dyslipidaemia.

Two-Directional Simultaneous Inference for High-Dimensional Models

This article proposes a general two-directional simultaneous inference (TOSI) framework for high-dimensional models with a manifest variable or latent variable structure, for example, high-dimensional mean models, high-dimensional sparse regression models, and high-dimensional latent factors models. TOSI performs simultaneous inference on a set of parameters from two directions, one to test whether the assumed zero parameters indeed are zeros and one to test whether exist zeros in the parameter set of nonzeros. As a result, we can better identify whether the parameters are zeros, thereby keeping the data structure fully and parsimoniously expressed. We theoretically prove that the single-split TOSI is asymptotically unbiased and the multi-split version of TOSI can control the Type I error below the prespecified significance level. Simulations are conducted to examine the performance of the proposed method in finite sample situations and two real datasets are analyzed. The results show that the TOSI method can provide more predictive and more interpretable estimators than existing methods.

Evinacumab in severe hypertriglyceridemia with or without lipoprotein lipase pathway mutations: a phase 2 randomized trial

Severe hypertriglyceridemia (sHTG) is an established risk factor for acute pancreatitis. Current therapeutic approaches for sHTG are often insufficient to reduce triglycerides and prevent acute pancreatitis. This phase 2 trial (NCT03452228) evaluated evinacumab (angiopoietin-like 3 inhibitor) in three cohorts of patients with sHTG: cohort 1, familial chylomicronemia syndrome with bi-allelic loss-of-function lipoprotein lipase (LPL) pathway mutations (n = 17); cohort 2, multifactorial chylomicronemia syndrome with heterozygous loss-of-function LPL pathway mutations (n = 15); and cohort 3, multifactorial chylomicronemia syndrome without LPL pathway mutations (n = 19). Fifty-one patients (males, n = 27; females, n = 24) with a history of hospitalization for acute pancreatitis were randomized 2:1 to intravenous evinacumab 15 mg kg−1 or placebo every 4 weeks over a 12-week double-blind treatment period, followed by a 12-week single-blind treatment period. The primary end point was the mean percent reduction in triglycerides from baseline after 12 weeks of evinacumab exposure in cohort 3. Evinacumab reduced triglycerides in cohort 3 by a mean (s.e.m.) of −27.1% (37.4) (95% confidence interval −71.2 to 84.6), but the prespecified primary end point was not met. No notable differences in adverse events between evinacumab and placebo treatment groups were seen during the double-blind treatment period. Although the primary end point of a reduction in triglycerides did not meet the prespecified significance level, the observed safety and changes in lipid and lipoprotein levels support the further evaluation of evinacumab in larger trials of patients with sHTG. Trial registration number: ClinicalTrials.gov NCT03452228.

Association of rare, recurrent nonsynonymous variants in the germline of prostate cancer patients of African ancestry.

Although men of African ancestry (AA) have the highest mortality rate from prostate cancer (PCa), relatively little is known about the germline variants that are associated with PCa risk in AA men. The goal of this study is to systematically evaluate rare, recurrent nonsynonymous variants across the exome for their association with PCa in AA men. Whole exome sequencing (WES) of germline DNA in two AA PCa patient cohorts of Johns Hopkins Hospital (N = 960) and Wayne State University (N = 747) was performed. All nonsynonymous variants present in both case cohorts, with a carrier rate between 0.5% and 1%, were identified. Their carrier rates were compared with rates from 8128 African/African American (AFR) control subjects from The Genome Aggregation Database (gnomAD) using Fisher's exact test. Significant variants, defined as false discovery rate (FDR) adjusted p-value ≤ 0.05, were further evaluated in AA PCa cases (N = 132) and controls (N = 1184) from the UK Biobank (UKB). Two variants reached a pre-specified statistical significance level. The first was p.R14Q in GPRC5C (found in 0.47% of PCa cases and 0.01% of population controls); odds ratio (OR) for PCa was 37.46 (95% confidence interval CI 4.68-299.72), pexact = 7.01E-06, FDR-adjusted p-value = 0.05. The second was p.R511Q in IGF1R (found in 0.53% of PCa cases and 0.01% of population controls); OR for PCa was 21.54 (95%CI 4.65-99.76), pexact = 5.51E-06, FDR-adjusted p-value = 0.05. The mean percentage of African ancestry was similar between variant carriers and noncarriers of each variant, p > 0.05. In the UKB AA men, GPRC5C R14Q was 0.76% and 0.08% in cases and controls, respectively, OR for PCa was 9.00 (95%CI 0.56-145.23), pexact = 0.19. However, IGF1R R511Q was not found in cases or controls. This WES study identified two rare, recurrent nonsynonymous PCa risk-associated variants in AA. Confirmation in additional large populations of AA PCa cases and controls is required.

Obesity Status and Physical Rehabilitation in Older Patients Hospitalized With Acute HF: Insights From REHAB-HF

BackgroundIn the REHAB-HF (Rehabilitation Therapy in Older Acute Heart Failure Patients) trial, a novel, early, transitional, multidomain rehabilitation intervention improved physical function, frailty, quality of life (QOL), and depression in older patients hospitalized for acute decompensated heart failure (ADHF), but the potential impact of baseline obesity on this intervention has not been studied. ObjectivesThis study assessed for treatment interactions by body mass index (BMI) subgroups for a novel rehabilitation intervention in ADHF. MethodsThree-month outcomes including Short Physical Performance Battery (SPPB) (primary outcome), 6-minute walk distance (6MWD), and Kansas City Cardiomyopathy Questionnaire (KCCQ) were assessed by baseline BMI (≥30 kg/m2 vs <30 kg/m2). Six-month end points included all-cause rehospitalization and death. All analyses were adjusted for age, sex, clinical site, and ejection fraction category, and 3-month outcomes were also adjusted for baseline measure. The prespecified significance level for treatment interaction by BMI category was P ≤ 0.10. ResultsOf 349 trial participants, 204 (58%) had BMI ≥30 kg/m2 and 145 (42%) <30 kg/m2. Compared with patients with BMI <30 kg/m2, participants with BMI ≥30 kg/m2 were younger (age 71 ± 7 years vs 75 ± 9 years), more frequently women (57% vs 46%), and had significantly worse baseline physical function and QOL. Although interaction P values for 3-month outcomes by BMI were not significant (interaction P > 0.15 for overall measures), adjusted SPPB effect sizes were nominally larger for participants with BMI ≥30 kg/m2 compared with those with BMI <30 kg/m2: +1.7 (95% CI: 0.8-2.7) vs +1.1 (95% CI: −0.1 to 2.2). This difference in SPPB effect size was due largely to improvements in the balance component of the SPPB for participants with BMI ≥30 kg/m2: +0.6 (95% CI: 0.2-1.0) vs 0.0 (−0.6 to 0.5) for those with BMI <30 kg/m2 (interaction P = 0.02). In contrast, adjusted 6MWD and KCCQ effect sizes were smaller for participants with BMI ≥30 kg/m2 compared with those with BMI <30 kg/m2: +21 meters (−17 to 59) vs +53 meters (6-100), and +5.0 (−4 to 14) vs +11 (−0.5 to 22), respectively. There was no significant interaction by BMI for 6-month clinical outcomes (all interaction P > 0.30). ConclusionsOlder patients with ADHF benefit from the rehabilitation therapy regardless of BMI. Benefits for patients with obesity may be more evident in the multidomain measure of physical function (SPPB), compared with the 6MWD or KCCQ, which may be driven, in part, by the unique aspects of the novel rehabilitation intervention. (A Trial of Rehabilitation Therapy in Older Acute Heart Failure Patients [REHAB-HF]; NCT02196038)

Assessing and visualizing fragility of clinical results with binary outcomes in R using the fragility package.

With the growing concerns about research reproducibility and replicability, the assessment of scientific results' fragility (or robustness) has been of increasing interest. The fragility index was proposed to quantify the robustness of statistical significance of clinical studies with binary outcomes. It is defined as the minimal event status modifications that can alter statistical significance. It helps clinicians evaluate the reliability of the conclusions. Many factors may affect the fragility index, including the treatment groups in which event status is modified, the statistical methods used for testing for the association between treatments and outcomes, and the pre-specified significance level. In addition to assessing the fragility of individual studies, the fragility index was recently extended to both conventional pairwise meta-analyses and network meta-analyses of multiple treatment comparisons. It is not straightforward for clinicians to calculate these measures and visualize the results. We have developed an R package called "fragility" to offer user-friendly functions for such purposes. This article provides an overview of methods for assessing and visualizing the fragility of individual studies as well as pairwise and network meta-analyses, introduces the usage of the "fragility" package, and illustrates the implementations with several worked examples.

Measurement of endocrine activity (SET2,3) related to prognosis and prediction of benefit from dose-dense (DD) chemotherapy in estrogen receptor-positive (ER+) cancer: CALGB 9741 (Alliance).

505 Background: We investigated the clinical utility of SET2,3, a novel biomarker designed to measure to endocrine sensitivity. SET2,3 measures nonproliferative hormone receptor-related transcription (SETER/PR) adjusted for a baseline prognosis index derived from tumor size, nodes involved and a 4-gene molecular subtype (RNA4). CALGB 9471 is a seminal phase III study that showed improved DFS and OS from 2-weekly dose-dense (DD) vs 3-weekly chemotherapy in ER-negative cancers. Risk of recurrence (ROR-PT) score (intrinsic subtype, proliferation score and tumor size) measured by Nanostring assay was reported to be prognostic in CALGB 9741, but did not predict benefit from DD chemotherapy. Methods: SET2,3 was performed using an aliquot of 200-300 ng RNA (residual from prior ROR-PT testing) from 682 ER+ tumor samples and tested using the QuantiGene Plex bead-based hybridization assay (ThermoFisher, Luminex). We report results for the primary and two secondary objectives of the NCI/CTEP-approved correlative science proposal CSC0154) to evaluate SET2,3 in CALGB 9741 for prognosis (primary endpoint: 95%CI for 5-year (yr) DFS &gt; 75% for High SET2,3 using the predefined prognostic cutpoint 2.10), SET2,3 prognostic independence from ROR-PT, and prediction of outcome according to chemotherapy regimen. We used Cox models to estimate hazard ratios (HR) for prognosis and comparison with ROR-PT results (using c-indices) and for prediction according to chemotherapy schedule using an interaction term (prespecified significance level for interaction: p &lt; 0.10). Results: The study met its primary endpoint with a 5-yr DFS of 85.6% (95%CI 81.3-90.2) in the High-SET subset (244/613, 40%). High-SET vs Low-SET was significantly associated with favorable outcomes at 5 yr (DFS 85.6% vs 69%, p &lt;.0001; OS 95.3% vs 84.6%, p &lt;.0001) and 10 yr (DFS 77.7% vs 58.2%, p &lt;.0001; OS 86.9% vs 65.9%, p &lt;.0001). PAM50 ROR-PT and SET classification were available for 596 tumors. In multivariate models for DFS and OS, SET2,3 remained an independent prognostic variable for DFS (SET high vs low HR = 0.46, 95% CI, 0.34 – 0.63, p &lt; 0.0001; PAM50 ROR-PT high vs low HR = 1.22, 95% CI, 0.91 – 1.64, p = 0.18) and for OS (SET high vs low HR = 0.36, 95% CI, 0.25 – 0.53, p &lt; 0.0001; PAM50 ROR-PT high vs low HR = 1.26, 95% CI, 0.91 – 1.75, p = 0.16). Similar observations were seen in models including SET and PAM50 ROR-PT as continuous variables. Lower SET2,3 values predicted improved outcomes from DD vs 3-weekly chemotherapy (interaction p=0.0998 for DFS, 0.042 for RFS and 0.027 for OS). This was unrelated to menopausal status and lower SET2,3 values favored DD concurrent treatments. Conclusions: SET2,3 index was strongly prognostic, independent of ROR-PT, and predicted survival benefit from DD chemotherapy in pre- and postmenopausal women with ER+ cancer. Clinical trial information: NCT00003088.

Cardiac MRI Findings in COVID-19 Vaccine-Related Myocarditis: A Pooled Analysis of 468 Patients.

Understanding the pattern and severity of myocarditis caused by the coronavirus disease 2019 (COVID‐19) vaccine is imperative for improving the care of the patients, and cardiac evaluation by MRI plays a key role in this regard. Our systematic review and meta‐analysis aimed to summarize cardiac MRI findings in COVID‐19 vaccine‐related myocarditis. We performed a comprehensive systematic review of literature in PubMed, Scopus, and Google Scholar databases using key terms covering COVID‐19 vaccine, myocarditis, and cardiac MRI. Individual‐level patient data (IPD) and aggregated‐level data (AD) studies were pooled through a two‐stage analysis method. For this purpose, all IPD were first gathered into a single data set and reduced to AD, and then this AD (from IPD studies) was pooled with existing AD (from the AD studies) using fixed/random effect models. I 2 was used to assess the degree of heterogeneity, and the prespecified level of statistical significance (P value for heterogeneity) was <0.1. Based on meta‐analysis of 102 studies (n = 468 patients), 79% (95% confidence interval [CI]: 54%–97%) of patients fulfilled Lake Louise criteria (LLC) for diagnosis of myocarditis. Cardiac MRI abnormalities included elevated T2 in 72% (95% CI: 50%–90%), myocardial late gadolinium enhancement (LGE) in 93% (95% CI: 83%–99%; nearly all with a subepicardial and/or midwall pattern), impaired left ventricular ejection fraction (LVEF) (<50%) in 4% (95% CI: 1.0%–9.0%). Moreover, elevated T1 and extracellular volume fraction (ECV) (>30), reported only by some IPD studies, were detected in 74.5% (76/102) and 32% (16/50) of patients, respectively. In conclusion, our findings may suggest that over two‐thirds of patients with clinically suspected myocarditis following COVID‐19 vaccination meet the LLC. COVID‐19 vaccine‐associated myocarditis may show a similar pattern compared to other acute myocarditis entities. Notably, preserved LVEF is probably a common finding in these patients.Evidence Level4Technical EfficacyStage 3

P260 Secukinumab in Patients With Psoriatic Arthritis and Axial Manifestations: Predictors of Response From the Double-blind, Randomised, Phase 3b MAXIMISE Trial

Abstract Background/Aims Psoriatic arthritis (PsA) is a heterogeneous disease with variability of response to different therapeutic modalities. Identifying potential demographic and disease characteristics as predictors of treatment response may define personalised treatment optimisation strategies. This post-hoc exploratory analysis of the MAXIMISE trial investigated the differential treatment effect of demographics and baseline characteristics as predictive factors in biologic-naïve patients with active PsA and symptoms of active spinal disease. Methods Full analysis set (FAS) comprised all patients from the randomised set assigned to study treatment, fulfilling the predefined clinical criteria for active axial disease, and for whom Assessment in Ankylosing Spondylitis 20% improvement criteria (ASAS20) data were available at week 12. The research hypothesis was that the odds ratio associated with the effect of treatment on ASAS20 responder status at week 12 would be different depending on 12 pre-specified predictor variables. A logistic regression model was initially fitted to the FAS that included 12 pre-specified covariates. A second logistic regression model was then fitted to the FAS that allowed for all 12 pre-specified variables to interact with treatment. The log-likelihood of the two fitted models were compared using a likelihood ratio test at a pre-specified significance level of 20% (i.e. P-value ≤0.20) to test whether any of the predefined variables interacted with treatment. If the above test was statistically significant at the 20% level, the variables of the second model were formally examined to determine whether overall effect of treatment was not applicable. Hypothesis tests were employed to determine the strength of evidence for each individual variable. Results The likelihood ratio test provided evidence against the assumption that overall effect of treatment is applicable to all patients (P-value, 0.08). Notably, the odds of being an ASAS20 responder if nail dystrophy is present at baseline were 3 times greater in the secukinumab 150-mg group and 5 times greater in the 300-mg group versus placebo (interaction P-value, 0.029). Although males fared worse than females in the placebo group, in the secukinumab 150-mg and 300-mg groups the odds of being a responder were similar to females (interaction P-value, 0.039). Current smokers were less likely to be ASAS20 responders versus never-smokers regardless of treatment group (interaction P-value, 0.589). Age, C-reactive protein level, Berlin MRI spine/SIJ score, time since first axial signs, number of swollen joints, new bone formation, and body mass index did not show a differential treatment effect on ASAS20 responses. Conclusion Of the 12 baseline variables of a unique population of 473 PsA patients with active axial disease diagnosed by clinical criteria, our analyses showed evidence of differential treatment effect, most notably for nail dystrophy, suggesting that the presence of nail dystrophy may predict a better response to secukinumab in PsA patients with axial manifestations. Disclosure L.C. Coates: Consultancies; AbbVie, Amgen, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Eli Lilly, Gilead, Janssen, Novartis, Pfizer, UCB. Member of speakers’ bureau; AbbVie, Amgen, Biogen, Medac, Celgene, Eli Lilly, Gilead, Janssen, Novartis, Pfizer, UCB. Grants/research support; AbbVie, Amgen, Celgene, Eli Lilly, Pfizer, Novartis. X. Baraliakos: Consultancies; AbbVie, BMS, Celgene, Chugai, Galapagos, Gilead, MSD, Novartis, Pfizer, UCB. Member of speakers’ bureau; AbbVie, BMS, Celgene, Chugai, MSD, Novartis, Pfizer, UCB. Grants/research support; AbbVie, Novartis. E. Pournara: Shareholder/stock ownership; Novartis. Other; Employee of Novartis. L. Gossec: Consultancies; AbbVie, Amgen, BMS, Galapagos, Gilead, Janssen, Lilly, Novartis, Pfizer, Samsung Bioepis, Sanofi-Aventis, UCB. Grants/research support; Amgen, Galapagos, Lilly, Pfizer, Sandoz, Sanofi. H. Marzo-Ortega: Consultancies; Janssen, Novartis, AbbVie, Celgene, Lilly, Pfizer, Takeda, UCB. Member of speakers’ bureau; AbbVie, Celgene, Janssen, Eli Lilly, Novartis, Pfizer, Takeda, UCB. Grants/research support; Janssen, Novartis, AbbVie, Celgene, Lilly, Pfizer, UCB. P.J. Mease: Consultancies; AbbVie, Amgen, BMS, Boehringer Ingelheim, Galapagos, Celgene, Gilead, Janssen, Lilly, Novartis, Pfizer, SUN Pharma, UCB. Member of speakers’ bureau; AbbVie, Amgen, Genentech, Janssen, Lilly, Merck, Novartis, Pfizer, UCB. Grants/research support; AbbVie, Amgen, BMS, Celgene, Galapagos, Genentech, Gilead, Janssen, Lilly, Novartis, Pfizer, SUN, UCB. R. White: Shareholder/stock ownership; Novartis. Other; Employee of Novartis. E. O'brien: Shareholder/stock ownership; Novartis. Other; Employee of Novartis. B. Schulz: Other; Employee of Novartis.

Efficacy and safety of itacitinib versus placebo in combination with corticosteroids for initial treatment of acute graft-versus-host disease (GRAVITAS-301): a randomised, multicentre, double-blind, phase 3 trial

Acute graft-versus-host disease (GVHD) is a common and life-threatening complication of allogeneic haematopoietic stem cell transplantation (HSCT); there is an urgent unmet need for effective therapies. We aimed to evaluate the Janus kinase 1 inhibitor itacitinib versus placebo, both in combination with corticosteroids, for initial treatment of acute GVHD. GRAVITAS-301 was an international, double-blind, adaptive (group sequential design) phase 3 study conducted at 129 hospitals and community practices in 19 countries. Eligible patients were aged 18 years or older, had previously received allogeneic HSCT for a haematological malignancy, developed grades II-IV acute GVHD, and received up to 2 days of systemic corticosteroids. Patients were stratified by clinical standard-risk or high-risk acute GVHD and randomly assigned (1:1), via a centralised interactive voice response system, to receive either oral itacitinib (200 mg) or placebo once daily, both in addition to corticosteroids. The primary endpoint was overall response rate (ORR) at day 28 (defined as the proportion of patients with complete response, very good partial response, or partial response 28 days after the start of treatment). For sample size determination, an absolute improvement in ORR at day 28 over standard therapy of 16% was considered clinically meaningful. Efficacy analyses were performed in the intention-to-treat population; safety analyses included patients who received at least one dose of study drug. GRAVITAS-301 is registered with ClinicalTrials.gov (NCT03139604) and is complete. Between July 19, 2017, and Oct 3, 2019, 439 patients were randomly assigned to receive either itacitinib plus corticosteroids (n=219; itacitinib group) or placebo plus corticosteroids (n=220; placebo group). 173 (39%) patients were female and 390 (89%) were White. At baseline, 107 (24%) of 439 patients (itacitinib 51 [23%] of 219; placebo 56 [25%] of 220) had clinical high-risk acute GVHD. The ORR at day 28 was 74% (95% CI 67·6-79·7; 162 of 219; complete response 53% [116 of 219]) for itacitinib and 66% (59·7-72·6; 146 of 220; complete response, 40% [89 of 220]) for placebo (odds ratio for ORR 1·45, 95% CI 0·96-2·20; two-sided p=0·078). Grade 3 or worse adverse events occurred in 185 (86%) of 215 itacitinib recipients and 178 (82%) of 216 placebo recipients, and most commonly included thrombocytopenia or platelet count decreased (78 [36%] vs 68 [31%]), neutropenia or neutrophil count decreased (49 [23%] vs 45 [21%]), anaemia (42 [20%] vs 26 [12%]), and hyperglycaemia (26 [12%] vs 28 [13%]). Treatment-related deaths occurred in three of 215 patients (1%) in the itacitinib group and four of 216 (2%) in the placebo group. The observed improvement in ORR at day 28 with the addition of itacitinib versus placebo to corticosteroids did not reach the prespecified significance level. Further studies might provide additional insight into the utility of selective JAK1 inhibition for the treatment of acute GVHD. Incyte.

Rehabilitation Intervention in Older Patients With Acute Heart Failure With Preserved Versus Reduced Ejection Fraction

ObjectivesThis study assessed for treatment interactions by ejection fraction (EF) subgroup (≥45% [heart failure with preserved ejection fraction (HFpEF); vs <45% [heart failure with reduced ejection fraction (HFrEF)]). BackgroundThe REHAB-HF trial showed that an early multidomain rehabilitation intervention improved physical function, frailty, quality-of-life, and depression in older patients hospitalized with acute decompensated heart failure (ADHF). MethodsThree-month outcomes were: Short Physical Performance Battery (SPPB), 6-min walk distance (6MWD), and Kansas City Cardiomyopathy Questionnaire (KCCQ). Six-month end points included all-cause rehospitalization and death and a global rank of death, all-cause rehospitalization, and SPPB. Prespecified significance level for interaction was P ≤ 0.1. ResultsAmong 349 total participants, 185 (53%) had HFpEF and 164 (47%) had HFrEF. Compared with HFrEF, HFpEF participants were more often women (61% vs 43%) and had significantly worse baseline physical function, frailty, quality of life, and depression. Although interaction P values for 3-month outcomes were not significant, effect sizes were larger for HFpEF vs HFrEF: SPPB +1.9 (95% CI: 1.1-2.6) vs +1.1 (95% CI: 0.3-1.9); 6MWD +40 meters (95% CI: 9 meters-72 meters) vs +27 (95% CI: −6 meters to 59 meters); KCCQ +9 (2-16) vs +6 (−2 to 14). All-cause rehospitalization rate was nominally lower with intervention in HFpEF but not HFrEF [effect size 0.83 (95% CI: 0.64-1.09) vs 0.99 (95% CI: 0.74-1.33); interaction P = 0.40]. There were significantly greater treatment benefits in HFpEF vs HFrEF for all-cause death [interaction P = 0.08; intervention rate ratio 0.63 (95% CI: 0.25-1.61) vs 2.21 (95% CI: 0.78-6.25)], and the global rank end point (interaction P = 0.098) with benefit seen in HFpEF [probability index 0.59 (95% CI: 0.50-0.68)] but not HFrEF. ConclusionsAmong older patients hospitalized with ADHF, compared with HFrEF those with HFpEF had significantly worse impairments at baseline and may derive greater benefit from the intervention. (A Trial of Rehabilitation Therapy in Older Acute Heart Failure Patients [REHAB-HF]; NCT02196038)

Timely Administration of Tocilizumab Improves Survival of Hospitalized COVID-19 Patients

Background: Accumulating evidence points to an overactive immune response in Covid-19 disease and potential clinical benefit of the interleukin-6 inhibitor tocilizumab. We assessed the efficacy of early tocilizumab treatment for hospitalized patients in a randomized phase II study.Methods: Patients admitted to the general ward with proven Covid-19 and in need of supplemental oxygen were randomly assigned to receive standard of care with or without intravenous tocilizumab 8 mg/kg (maximal 800 mg). A second dose of tocilizumab was permitted if hypoxia persisted after 8 hrs. The primary endpoint of the study was 30-day mortality with a prespecified 2-sided significance level of α=0.10. A post-hoc analysis was performed for a combined endpoint of mechanical ventilation or death at 30 days.Findings: A total of 354 patients (67% men; median age 66 years) were enrolled of whom 88% received dexamethasone. Thirty-day mortality was 19% (95% CI 14%-26%) in the standard arm versus 12% (95% CI: 8%-18%) in the tocilizumab arm, hazard ratio (HR)=0.62 (90% CI 0.39-0.98; p=0.086). 21% of patients were admitted to the ICU in each arm (p=0.89). The median stay in the ICU was 16 days (IQR 8-30) in the standard arm versus 9 days (IQR 5-16) in the tocilizumab arm (p=0.025). Mechanical ventilation or death at thirty days was 31% (95% CI 24%-38%) in the standard arm versus 21% (95% CI 16%-28%) in the tocilizumab arm, HR = 0.65 (95% CI 0.42-0.98; p=0.042).Interpretation: Various studies have suggested a beneficial effect of tocilizumab in the treatment of COVID-19. This randomized phase II study, which met its primary endpoint, confirms these observations and demonstrates a clinically meaningful efficacy when given early in the disease course in hospitalized patients who need oxygen support, even when concomitantly treated with dexamethasone.Trial Registration: The trial was designed as a prospective randomized (1:1) open label phase II trial and was registered in the Netherlands Trial register (https://www.trialregister.nl/trial/8504).Funding Statement: Academic study, funded by participating hospitals. Roche supplied tocilizumab.Declaration of Interests: None to declare. Ethics Approval Statement: The trial was approved by the relevant medical ethical committee and was performed in accordance with Good Clinical Practice guidelines and the Helsinki Declaration.

Exact unconditional inference for analyzing contingency tables in finite populations

ABSTRACT With recent developments in computer power the application of exact inferential methods has become more feasible which has resulted in increasing popularity of these approaches. However, there is a lack of such methodology for populations with more complex structure, such as finite populations. When a small sample is drawn from a finite population, the number of individuals with a specific characteristic of interest follows hypergeometric distribution. In order to test for the comparison of two proportions in finite populations we develop an exact unconditional test. We utilize the information gained from the sample to restrict our search for the maximum p-value. Our proposed test has power equal to its competitors while maintains the pre-specified nominal significance level.

Author

In clinical trials, power calculation is often performed based on a single primary endpoint to determine sample size required for achieving study objective with a desired power at a pre-specified level of significance. In practice, power calculation based on a single primary endpoint has been criticized. First, how to select the single primary endpoint among a group of primary endpoints? Second, a single primary endpoint may not be sufficient to adequately inform complex cohorts, the disease status and/or treatment effect of the test treatment under investigation. Third, different study endpoints with different data types (e.g., continuous versus binary response) may result in different sample sizes. In addition, with a given sample size, some (single) endpoints may achieve the study objective while others fail to do so. In this opinion article, a conceptual innovation is the development of a therapeutic index that fully utilizes information from all relevant study endpoints proposed.