Are the tests overpowered or underpowered? A unified solution to correctly specify type I errors in design of clinical trials for two sample proportions.

As one of the most commonly used data types, methods in testing or designing a trial for binary endpoints from two independent populations are still being developed until recently. However, the power and the minimum required sample size comparisons between different tests may not be valid if their type I errors are not controlled at the same level. In this article, we unify all related testing procedures into a decision framework, including both frequentist and Bayesian methods. Sufficient conditions of the type I error attained at the boundary of hypotheses are derived, which help reduce the magnitude of the exact calculations and lay out a foundation for developing computational algorithms to correctly specify the actual type I error. The efficient algorithms are thus proposed to calculate the cutoff value in a deterministic decision rule and the probability value in a randomized decision rule, such that the actual type I error is under but closest to, or equal to, the intended level, respectively. The algorithm may also be used to calculate the sample size to achieve the prespecified type I error and power. The usefulness of the proposed methodology is further demonstrated in the power calculation for designing superiority and noninferiority trials.

The impact of social distancing measures on anti-JC virus serostatus changes before and during the COVID-19 pandemic in US patients with multiple sclerosis.

BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic offered an epidemiological opportunity to evaluate if isolation and masking affected John Cunningham (JC) virus transmission. OBJECTIVE: This study aimed to assess the proportion of natalizumab-treated patients who converted to a positive anti-JCV antibody serostatus before and during the pandemic. METHODS: Data from TYSABRI Outreach: Unified Commitment to Health (TOUCH) for 22,375 US patients treated with natalizumab with anti-JCV antibody records were assessed in epochs annually from 2017 to 2022. RESULTS: Pre-pandemic anti-JCV antibody serostatus change was observed for 7.4%-7.7%. During the first and second years of the pandemic, 7.3% and 7.2% of patients' serostatus changed, respectively. CONCLUSION: The proportion of patients with anti-JCV antibody serostatus change did not significantly differ during the first 2 years of the pandemic compared with prior years. In contrast to seasonal influenza, masking and social distancing had no discernable effect on JCV serostatus change.

Exploratory clinical efficacy and patient-reported outcomes from NOVA: A randomized controlled study of intravenous natalizumab 6-week dosing versus continued 4-week dosing for relapsing-remitting multiple sclerosis.

BACKGROUND: Natalizumab (TYSABRI®) 300 mg administered intravenously every-4-weeks (Q4W) is approved for treatment of relapsing-remitting multiple sclerosis but is associated with increased risk of progressive multifocal leukoencephalopathy (PML). Extended natalizumab dosing intervals of approximately every-6-weeks (Q6W) are associated with a lower risk of PML. Primary and secondary clinical outcomes from the NOVA randomized clinical trial (NCT03689972) suggest that effective disease control is maintained in patients who were stable during treatment with natalizumab Q4W for ≥12 months and who then switched to Q6W dosing. We compared additional exploratory clinical and patient-reported outcomes (PROs) from NOVA to assess the efficacy of Q6W dosing. METHODS: Prespecified exploratory clinical efficacy endpoints in NOVA included change from baseline in Expanded Disability Status Scale (EDSS) score, Timed 25-Foot Walk (T25FW), dominant- and nondominant-hand 9-Hole Peg Test (9HPT), and Symbol Digit Modalities Test (SDMT). Exploratory patient-reported outcome (PRO) efficacy endpoints included change from baseline in the Treatment Satisfaction Questionnaire for Medication (TSQM), Neuro-QoL fatigue questionnaire, Multiple Sclerosis Impact Scale (MSIS-29), EuroQol 5 Dimensions (EQ-5D-5 L) index score, Clinical Global Impression (CGI)-Improvement (patient- and clinician-assessed) and CGI-Severity (clinician-assessed) rating scales. Estimated proportions of patients with confirmed EDSS improvement were based on Kaplan-Meier methods. Estimates of mean treatment differences for Q6W versus Q4W in other outcomes were assessed by least squares mean (LSM) and analyzed using a linear mixed model of repeated measures or ordinal logistic regression (CGI-scale). RESULTS: Exploratory clinical and patient-reported outcomes were assessed in patients who received ≥1 dose of randomly assigned study treatment and had ≥1 postbaseline efficacy assessment (Q6W group, n = 247, and Q4W group, n = 242). Estimated proportions of patients with EDSS improvement at week 72 were similar for Q6W and Q4W groups (11.7% [19/163] vs 10.8% [17/158]; HR 1.02 [95% confidence interval [CI], 0.53-1.98]; P = 0.9501). At week 72, there were no significant differences between Q6W and Q4W groups in LSM change from baseline for T25FW (0.00, P = 0.975), 9HPT (dominant [0.22, P = 0.533] or nondominant [0.09, P = 0.862] hand), or SDMT (-1.03, P = 0.194). Similarly, there were no significant differences between Q6W and Q4W groups in LSM change from baseline for any PRO (TSQM, -1.00, P = 0.410; Neuro-QoL fatigue, 0.52, P = 0.292; MSIS-29 Psychological, 0.67, P = 0.572; MSIS-29 Physical, 0.74, P = 0.429; EQ-5D-5 L, 0.00, P = 0.978). For the EQ-5D-5 L, a higher proportion of Q6W patients than Q4W patients demonstrated worsening (≥0.5 standard deviation increase in the EQ-5D-5 L index score; P = 0.0475). From baseline to week 72 for Q6W versus Q4W, odds ratio (ORs) of LSM change in CGI scores did not show meaningful differences between groups (CGI-Improvement [patient]: OR [95% CI] 1.2 [0.80-1.73]; CGI-Improvement [physician]: 0.8 [0.47-1.36]; CGI-Severity [physician]: 1.0 [0.71-1.54]). CONCLUSIONS: No significant differences were observed in change from baseline to week 72 between natalizumab Q6W and Q4W groups for all exploratory clinical or PRO-related endpoints assessed. For the EQ-5D-5 L, a higher proportion of Q6W than Q4W patients demonstrated worsening.

Statistical methods for handling missing data to align with treatment policy strategy.

The International Council for Harmonization (ICH) E9(R1) addendum recommends choosing an appropriate estimand based on the study objectives in advance of trial design. One defining attribute of an estimand is the intercurrent event, specifically what is considered an intercurrent event and how it should be handled. The primary objective of a clinical study is usually to assess a product's effectiveness and safety based on the planned treatment regimen instead of the actual treatment received. The estimand using the treatment policy strategy, which collects and analyzes data regardless of the occurrence of intercurrent events, is usually utilized. In this article, we explain how missing data can be handled using the treatment policy strategy from the authors' viewpoint in connection with antihyperglycemic product development programs. The article discusses five statistical methods to impute missing data occurring after intercurrent events. All five methods are applied within the framework of the treatment policy strategy. The article compares the five methods via Markov Chain Monte Carlo simulations and showcases how three of these five methods have been applied to estimate the treatment effects published in the labels for three antihyperglycemic agents currently on the market.