A comparison of estimation methods adjusting for selection bias in adaptive enrichment designs with time-to-event endpoints.

Adaptive enrichment designs in clinical trials have been developed to enhance drug developments. They permit, at interim analyses during the trial, to select the sub-populations that benefits the most from the treatment. Because of this selection, the naive maximum likelihood estimation of the treatment effect, commonly used in classical randomized controlled trials, is biased. In the literature, several methods have been proposed to obtain a better estimation of the treatments' effects in such contexts. To date, most of the works have focused on normally distributed endpoints, and some estimators have been proposed for time-to-event endpoints but they have not all been compared side-by-side. In this work, we conduct an extensive simulation study, inspired by a real case-study in heart failure, to compare the maximum-likelihood estimator (MLE) with an unbiased estimator, shrinkage estimators, and bias-adjusted estimators for the estimation of the treatment effect with time-to-event data. The performances of the estimators are evaluated in terms of bias, variance, and mean squared error. Based on the results, along with the MLE, we recommend to provide the unbiased estimator and the single-iteration bias-adjusted estimator: the former completely eradicates the selection bias, but is highly variable with respect to a naive estimator; the latter is less biased than the MLE estimator and only slightly more variable.

Predictive probability of success using surrogate endpoints.

The predictive probability of success of a future clinical trial is a key quantitative tool for decision-making in drug development. It is derived from prior knowledge and available evidence, and the latter typically comes from the accumulated data on the clinical endpoint of interest in previous clinical trials. However, a surrogate endpoint could be used as primary endpoint in early development and, usually, no or limited data are collected on the clinical endpoint of interest. We propose a general, reliable, and broadly applicable methodology to predict the success of a future trial from surrogate endpoints, in a way that makes the best use of all the available evidence. The predictions are based on an informative prior, called surrogate prior, derived from the results of past trials on one or several surrogate endpoints. If available, in a Bayesian framework, this prior could be combined with data from past trials on the clinical endpoint of interest. Two methods are proposed to address a potential discordance between the surrogate prior and the data on the clinical endpoint. We investigate the patterns of behavior of the predictions in a comprehensive simulation study, and we present an application to the development of a drug in Multiple Sclerosis. The proposed methodology is expected to support decision-making in many different situations, since the use of predictive markers is important to accelerate drug developments and to select promising drug candidates, better and earlier.

Randomized Trial of Macitentan/Tadalafil Single-Tablet Combination Therapy for Pulmonary Arterial Hypertension.

BACKGROUND: Endothelin receptor antagonist (ERA) and phosphodiesterase 5 inhibitor (PDE5i) combination therapy is recommended for low-/intermediate-risk pulmonary arterial hypertension (PAH) patients. A fixed-dose combination of the ERA macitentan and PDE5i tadalafil (M/T FDC) in a once-daily, single tablet would simplify treatment. OBJECTIVES: The multicenter, double-blind, adaptive phase 3 A DUE study investigated the efficacy and safety of M/T FDC vs macitentan 10 mg and vs tadalafil 40 mg monotherapies in PAH patients, including treatment-naïve and prior ERA or PDE5i monotherapy-treated patients. METHODS: World Health Organization functional class II-III patients were randomized to M/T FDC, macitentan, or tadalafil depending on their PAH treatment (treatment-naïve, ERA, or PDE5i monotherapy) at baseline. The primary endpoint was change in pulmonary vascular resistance (PVR) at week 16. RESULTS: In total, 187 patients were randomized to single-tablet M/T FDC (n = 108), macitentan (n = 35), or tadalafil (n = 44). PVR reduction with M/T FDC was significantly greater vs macitentan (29%; geometric mean ratio 0.71; 95% CL: 0.61-0.82; P < 0.0001) and vs tadalafil (28%; geometric mean ratio 0.72; 95% CL: 0.64-0.80; P < 0.0001). Three patients died in the M/T FDC arm (judged unrelated to treatment). Adverse events (AEs) leading to discontinuation, serious AEs, and those of special interest (anemia, hypotension, and edema) were more frequent with M/T FDC. CONCLUSIONS: Macitentan and tadalafil FDC significantly improved PVR vs monotherapies in PAH patients, with a safety and tolerability profile consistent with the individual components. The A DUE study supports M/T FDC as a once-daily, single-tablet combination for initial therapy and escalation to double combination therapy in patients with PAH. (Clinical Study to Compare the Efficacy and Safety of Macitentan and Tadalafil Monotherapies With the Corresponding Fixed-dose Combination Therapy in Subjects With Pulmonary Arterial Hypertension [PAH]) [A DUE]; NCT03904693).

UCART19, a first-in-class allogeneic anti-CD19 chimeric antigen receptor T-cell therapy for adults with relapsed or refractory B-cell acute lymphoblastic leukaemia (CALM): a phase 1, dose-escalation trial.

BACKGROUND: The prognosis for adults with relapsed or refractory B-cell acute lymphoblastic leukaemia remains poor. UCART19, an allogeneic genome-edited anti-CD19 chimeric antigen receptor (CAR) T-cell product derived from healthy donors and available for immediate clinical use, offers a potential therapeutic option for such patients. The CALM trial is a first-in-human study evaluating the safety and antileukaemic activity of UCART19 in adult patients with relapsed or refractory B-cell acute lymphoblastic leukaemia. METHODS: This phase 1, open-label study was conducted at eight centres across France, the UK, the USA, and Japan. Adult patients aged 16-70 years with CD19-positive relapsed or refractory B-cell acute lymphoblastic leukaemia who had morphological relapse or a minimal residual disease level of at least 1 × 10-3 and had exhausted standard treatment options were enrolled in the study, which comprised a dose-escalation phase of up to three UCART19 doses followed by a safety expansion phase. Patients underwent lymphodepletion with fludarabine (30 mg/m2 per day intravenously for 3 days) and cyclophosphamide (500 mg/m2 per day intravenously for 3 days) with or without alemtuzumab (1 mg/kg or 40 mg or 60 mg over 5 days) and received UCART19 doses of 6 × 106, 6-8 × 107, or 1·8-2·4 × 108 total CAR T cells intravenously, followed by safety evaluation and disease response assessments. The primary endpoint was incidence and severity of adverse events. Secondary endpoints were the overall response rate, duration of response, relapse-free survival, progression-free survival, and overall survival. This trial is registered with ClinicalTrials.gov (NCT02746952) and is complete. FINDINGS: Between Aug 1, 2016, and June 30, 2020, 25 patients were enrolled in the study and treated with UCART19. Median duration of follow-up was 12·8 months (IQR 2·8-24·8). Median age was 37 years (IQR 28-45). 14 (56%) patients were male and 11 (44%) female. 17 (68%) patients were White, two (8%) Black, two (8%) Asian, and four (16%) from other racial or ethnic groups. Three patients developed dose-limiting toxicities (one at each dose level); one had grade 4 cytokine release syndrome and two had grade 4 prolonged cytopenias. Grade 3 or higher cytokine release syndrome was reported in six (24%) patients and grade 3 or higher neurological toxicity in one (4%) patient. Grade 3 or higher infections occurred in seven (28%) patients, and grade 4 prolonged cytopenia in four (16%) patients. Two (8%) patients developed grade 1 acute cutaneous graft-versus-host disease. 14 patients died, nine from progressive disease and five from infections or other complications, of which four were considered to be related to UCART19 or lymphodepletion, or both. After a median of follow-up of 12·8 months (IQR 2·8-24·8), overall response rate was 48% (95% CI 28-69; 12 of 25 patients), duration of response and median relapse-free survival were 7·4 months (95% CI 1·8 to not calculable), progression-free survival was 2·1 months (95% CI 1·2-2·8), and overall survival was 13·4 months (95% CI 4·8-23·0). INTERPRETATION: UCART19 had a manageable safety profile, and showed evidence of antileukaemic activity in heavily pretreated adult patients with relapsed or refractory B-cell acute lymphoblastic leukaemia. This study shows that allogeneic off-the-shelf CAR T cells can be used safely to treat patients with relapsed B-cell acute lymphoblastic leukaemia. FUNDING: Servier.

Comparison of Outcome Adjudication by Investigators and by a Central End Point Committee in Heart Failure Trials: Experience of the SHIFT Heart Failure Study.

BACKGROUND: The usefulness of adjudication by central end point committees (CECs) is poorly assessed in heart failure (HF) trials. We aimed to assess its impact on the outcome of the SHIFT trial (Systolic HF Treatment With the If Inhibitor Ivabradine Trial). METHODS: SHIFT was a randomized placebo-controlled trial investigating the effect of ivabradine in 6505 HF patients with reduced ejection fraction. Prespecified end points, reported by investigators (all cardiologists) using specific case report form pages, included all-cause and specific causes of deaths and hospitalizations. The primary end point was a composite of cardiovascular deaths or hospitalizations for worsening HF. We compared the adjudication of prespecified end points made by investigators and by the CEC. RESULTS: Investigators identified 7529 prespecified end points, 6793 of which were confirmed by the CEC: 98.1% of cardiovascular deaths, 88.6% of all hospitalizations, and 84.4% of hospitalizations for worsening HF. These differences had no meaningful impact on the study results; hazard ratio for the primary composite end point: investigators, 0.83 (95% CI, 0.76-0.91) versus CEC, 0.82 (95% CI, 0.75-0.90), with similar results for each component of the primary end point (hazard ratio of 0.92 versus 0.91 for cardiovascular death and 0.78 versus 0.74 for hospitalization for worsening HF). CONCLUSIONS: Central adjudication by a CEC in the SHIFT study confirmed most of cardiovascular deaths and worsening HF hospitalizations assessed by cardiologists and did not result in a significant change of the final result as compared to investigator judgment. In this context, the benefits of CEC in blinded HF trials should be reconsidered. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT02441218. URL: http://www.isrctn.com/ISRCTN70429960; Unique identifier: ISRCTN70429960.

Incremental benefit of drug therapies for chronic heart failure with reduced ejection fraction: a network meta-analysis.

AIMS: A network meta-analysis (NMA) of all recommended drug groups for the treatment of heart failure with reduced ejection fraction (HFrEF), including their combinations, was performed to assess the relative efficacy and incremental benefit. METHODS AND RESULTS: A search was made in biomedical databases for randomized controlled trials published between 1987 and 2017 on angiotensin-converting enzyme inhibitors (ACEIs), beta-blockers (BBs), angiotensin receptor blockers (ARBs), mineralocorticoid receptor antagonists (MRAs), ivabradine (IVA), or angiotensin receptor-neprilysin inhibitors (ARNI). A total of 58 relevant trials were identified. The relative efficacy of each treatment group (or combination) in terms of all-cause mortality, cardiovascular mortality, all-cause hospitalizations and hospitalizations for heart failure, per patient-year of follow-up, were combined in a random-effects Bayesian NMA. The pairwise comparison between each regimen and for each outcome was estimated. The NMA was dominated by 15 large-scale trials with between 1984 and 18 898 patient-years of follow-up. Combinations of drug groups showed incremental benefits on outcomes over single groups. The most effective combinations were ARNI+BB + MRA and ACEI+BB + MRA + IVA, showing reductions in all-cause mortality (vs. placebo) of 62% and 59%, respectively; hazard ratios were 0.38 [credible interval (CrI) 0.20-0.65] and 0.41 (CrI 0.21-0.70); and in all-cause hospitalizations with reductions of 42% for both. These two combinations were also the most effective for the other outcomes studied. CONCLUSION: Our analysis shows that the incremental use of combinations of disease-modifying therapies has resulted in the progressive improvement in mortality and hospitalization outcomes in HFrEF. Our findings support the current guideline recommendations.

Effect of ivabradine in patients with heart failure with preserved ejection fraction: the EDIFY randomized placebo-controlled trial.

AIMS: This randomized, double-blind, placebo-controlled trial assessed whether heart rate (HR) reduction with ivabradine improves cardiac function in heart failure with preserved ejection fraction (HFpEF). METHODS AND RESULTS: The prEserveD left ventricular ejectIon fraction chronic heart Failure with ivabradine studY (EDIFY) included 179 patients in New York Heart Association (NYHA) classes II and III, in sinus rhythm, with HR of ≥70 b.p.m., NT-proBNP of ≥220 pg/mL (BNP ≥80 pg/mL) and left ventricular ejection fraction of ≥45%. Ivabradine (or placebo) was titrated to 7.5 mg b.i.d. Patients were followed for 8 months on the change and assessed for three co-primary endpoints: echo-Doppler E/e' ratio, distance on the 6-min walking test (6MWT), and plasma NT-proBNP concentration. At baseline, median E/e' was 12.8 [interquartile range (IQR): 9.9-16.3], median distance on the 6MWT was 320 m (IQR: 247-375 m), and median NT-proBNP was 375 pg/mL (IQR: 253-701 pg/mL). Baseline median HR was 75 b.p.m. (IQR: 70-107 b.p.m.). A total of 171 patients (87 in the ivabradine group, 84 in the placebo group) were evaluated for treatment efficacy. After 8 months of treatment, findings showed a median change in HR of -13.0 b.p.m. (IQR: -18.0 to -6.0 b.p.m.) in the ivabradine group and -3.5 b.p.m. (IQR: -11.5 to 3.0 b.p.m.) in the placebo group [estimated between-group difference: 7.7 b.p.m.; 90% confidence interval (CI) -10 to -5.4; P < 0.0001]. No evidence of improvement was found in any of the three co-primary endpoints. There was almost no change in median E/e' in either of the two groups [median change: +1.0 (IQR: -0.8 to 2.9) in the ivabradine group; -0.6 (IQR: -2.2 to 1.4) in the placebo group; estimated between-group difference: 1.4, 90% CI 0.3-2.5; P = 0.135]. There were no meaningful changes in the other co-primary endpoints and no apparent trends. There was no significant safety concern. CONCLUSIONS: In patients with HFpEF, HR reduction with ivabradine did not improve outcomes. These findings do not support the use of ivabradine in HFpEF.