Notes on Test and Estimation in Comparison of Three Treatments under A Simple Carry-Over Three-Period Model

2018 ◽  
Vol 14 (1) ◽  
Author(s):  
Kung-Jong Lui
Keyword(s):  
Type I Error ◽  
Average Length ◽  
Test Procedure ◽  
Forced Expiratory Volume ◽  
Crossover Design ◽  
Type I ◽  
Test Procedures ◽  
Double Blind ◽  
Interval Estimators ◽  
Carry Over

Abstract Under the three-treatment three-period crossover design with simple carry-over effects, we derive the least-squares estimators for period effects, treatment effects and carry-over effects, as well as their covariance matrix in closed and explicit expressions. Using Monte Carlo simulation, we compare the test procedure adjusting carry-over with that ignoring carry-over with respect to Type I error and power. We further compare interval estimators adjusting carry-over with those ignoring carry-over with respect to the coverage probability and the average length. When the variation of responses within patients is small, the test procedure and interval estimators ignoring carry-over can lose accuracy in the presence of carry-over effects. When the variation of responses within patients is large, this loss of accuracy may become small or even minimal. We note that the loss of efficiency due to the adjustment of carry-over under the simple carry-over three-period crossover design is moderate, and is much less than that found for a two-period crossover design. We use the double-blind three-period crossover trial comparing formoterol solution aerosol and salbutamol suspension aerosol with a placebo for patients suffering from exercise-induced asthma on the forced expiratory volume in one second (FEV1) to illustrate the use of test procedures and interval estimators discussed here.

scholarly journals Testing for equality of distributions using the concept of (niche) overlap

2021 ◽  
Author(s):  
Judith H. Parkinson-Schwarz ◽  
Arne C. Bathke
Keyword(s):  
Basic Assumption ◽  
Type I Error ◽  
Test Procedure ◽  
Niche Overlap ◽  
Relative Effect ◽  
Type I ◽  
Underlying Distribution ◽  
Parametric Test ◽  
Higher Power ◽  
Non Parametric

AbstractIn this paper, we propose a new non-parametric test for equality of distributions. The test is based on the recently introduced measure of (niche) overlap and its rank-based estimator. As the estimator makes only one basic assumption on the underlying distribution, namely continuity, the test is universal applicable in contrast to many tests that are restricted to only specific scenarios. By construction, the new test is capable of detecting differences in location and scale. It thus complements the large class of rank-based tests that are constructed based on the non-parametric relative effect. In simulations this new test procedure obtained higher power and lower type I error compared to two common tests in several settings. The new procedure shows overall good performance. Together with its simplicity, this test can be used broadly.

scholarly journals Non-inferiority test for a continuous variable with a flexible margin in an active controlled trial : An application to the "Stratall ANRS 12110 / ESTHER" trial

2020 ◽  
Author(s):  
Arsene Sandie ◽  
Nicholas Molinari ◽  
Anthony Wanjoya ◽  
Charles Kouanfack ◽  
Christian Laurent ◽  
...  
Keyword(s):  
Confidence Interval ◽  
Active Control ◽  
Error Rate ◽  
Type I Error ◽  
Clinical Monitoring ◽  
Small Scale ◽  
Type I ◽  
Test Procedures ◽  
Power Estimate ◽  
Control Intervention

Abstract Background: The non-inferiority trials are becoming increasingly popular in public health and clinical research. The choice of the non-inferiority margin is the cornerstone of the non-inferiority trial. When the effect of active control intervention is unknown, it can be interesting to choose the non-inferiority margin as a function of the active control intervention effect. In this case, the uncertainty surrounding the non-inferiority margin should be accounted for in statistical tests. In this work, we explored how to perform the non-inferiority test with a flexible margin for continuous endpoint.Methods: It was proposed in this study two procedures for the non-inferiority test with a flexible margin for the continuous endpoint. The proposed test procedures are based on test statistic and confidence interval approach. Simulations have been used to assess the performances and properties of the proposed test procedures. An application was done on clinical real data, which the purpose was to assess the efficacy of clinical monitoring alone versus laboratory and clinical monitoring in HIV-infected adult patients.Results: Basically, the two proposed test procedures have good properties. In the test based on a statistic, the actual type 1 error rate estimate is approximatively equal to the nominal value. It has been found that the confidence interval level determines approximately the level of significance. The 80%, 90%, and 95%one-sided confidence interval levels led approximately to a type I error of 10%, 5% and 2.5% respectively. The power estimate was almost 100% for two proposed tests, except for the small scale values of the reference treatment where the power was relatively low when the sample sizes were small.Conclusions: Based on type I error rate and power estimates, the proposed non-inferiority hypothesis test procedures have good performance and are applicable in practice.Trial registration: The trial data used in this study was from the ”Stratall ANRS 12110 / ESTHER”, registered with ClinicalTrials.gov, number NCT00301561. Date : March 13, 2006, url : https://clinicaltrials.gov/ct2/show/NCT00301561.

Axitinib with or without dose titration for first-line metastatic renal cell carcinoma (mRCC): Unblinded results from a randomized phase II study.

2013 ◽  
Vol 31 (6_suppl) ◽  
pp. LBA349-LBA349 ◽  
Author(s):  
Brian I. Rini ◽  
Viktor Gruenwald ◽  
Mayer N. Fishman ◽  
Bohuslav Melichar ◽  
Takeshi Ueda ◽  
...  
Keyword(s):  
Drug Exposure ◽  
Type I Error ◽  
Response Rate ◽  
Objective Response ◽  
Progression Free Survival ◽  
Dose Titration ◽  
Type I ◽  
First Line ◽  
Antihypertensive Medications ◽  
Double Blind

LBA349 Background: Patients receiving the 5-mg twice daily (BID) axitinib starting dose exhibit variable drug exposure; prior pharmacokinetic analyses indicate higher exposure is associated with better outcomes in mRCC. Dose titration based on individual tolerability may optimize exposure and improve efficacy. Methods: Patients (N=213) with treatment-naïve mRCC received axitinib 5 mg BID for a 4-week lead-in period. Then, patients with 2 consecutive weeks of blood pressure ≤150/90 mmHg, no axitinib-related toxicities >grade 2, no dose reductions, and ≤2 antihypertensive medications were randomized (double-blind) to axitinib 5 mg BID + dose titration to 10 mg BID maximum with axitinib or placebo. Those not eligible for randomization continued axitinib 5 mg BID or lower. Primary endpoint was objective response rate (ORR) in randomized arms. Progression-free survival (PFS), overall survival, and safety were secondary endpoints. Assuming response rate under the null hypothesis is 0.15, this study had ≥80% power (1-sided type I error 10%) to detect a ≥25% absolute improvement in ORR with active vs placebo titration. Results: In all, 56 patients each were randomized to active and placebo titration arms, 91 were not randomized, and 10 withdrew during the lead-in period. As of Oct 12, 2012, ORR (95% confidence interval [CI]) was 54% (40–67) in the active titration arm vs 34% (22–48) in the placebo titration arm (1-sided P=0.019), and 59% (49–70) in the non-randomized arm. Median PFS (95% CI) from first dose was 14.5 mo (9.2–24.5) in the active titration arm vs 15.7 mo (8.3–19.4) in the placebo titration arm (hazard ratio favored active titration, 0.85; 95% CI, 0.54–1.35; 1-sided P=0.244), and 16.6 mo (11.2–22.5) in the non-randomized arm. Most frequent all-grade, all-causality adverse events in active titration, placebo titration, and non-randomized arms, respectively, were diarrhea (61% vs 63% vs 63%), hypertension (61% vs 43% vs 82%), and fatigue (45% vs 46% vs 54%). Conclusions: Axitinib is effective and well tolerated in first-line mRCC with prolonged median PFS in all treatment arms compared to historical controls. Axitinib dose titration significantly improved ORR vs placebo. Clinical trial information: NCT00835978.

Randomized, double-blind, placebo-controlled, phase II trial of first-line platinum/docetaxel with or without erlotinib (E) in patients (pts) with recurrent and/or metastatic (R/M) head and neck squamous cell carcinomas (HNSCCs).

2017 ◽  
Vol 35 (15_suppl) ◽  
pp. 6017-6017 ◽  
Author(s):  
William Nassib William ◽  
Lei Feng ◽  
Merrill S. Kies ◽  
Salmaan Ahmed ◽  
George R. Blumenschein ◽  
...  
Keyword(s):  
Primary Endpoint ◽  
Type I Error ◽  
Performance Status ◽  
Local Treatment ◽  
Progression Free Survival ◽  
Type I ◽  
Phase 2 ◽  
First Line ◽  
Double Blind ◽  
Phase 2 Trial

6017 Background: In a single-arm, phase 2 study, we previously demonstrated that in pts with R/M HNSCC, cisplatin, docetaxel and E improved progression-free survival (PFS) compared to historical data (Kim et al., ASCO 2006). Herein, we evaluated this regimen in a single center, randomized, phase 2 trial. Methods: Pts with R/M HNSCC, with a performance status (PS) 0-2, were randomized (1:1) to receive up to 6 cycles of first-line chemotherapy with cisplatin 75 mg/m2 (or carboplatin AUC 6) and docetaxel 75 mg/m2 i.v. on day 1 every 21 days, plus placebo (P) vs. E 150 mg p.o. daily, followed by maintenance P or E until disease progression. The primary endpoint was PFS. With 120 pts, the study had 80% power to detect an improvement in median PFS from 3.0 to 4.9 months with a two-sided type I error rate of 0.1. Results: From 05/2010 to 07/2015, 120 pts were randomized to the P (N = 60) or E (N = 60) groups. All pts but one initiated treatment and were eligible for evaluation of the primary endpoint – 92 males; median age 62 years; 52 oropharynx, 40 oral cavity, 19 larynx, 8 hypopharynx cancer pts; 86 current/former smokers; 43 with recurrence within 6 months of completion of local treatment; 27 with prior exposure to EGFR inhibitors. Median PFS was 4.4 vs. 6.1 months for the P and E groups, respectively (hazard ratio [HR] 0.63, 95% confidence interval [CI] 0.42-0.95 months, p = 0.026). Response rates were 44% vs. 56% for P vs. E (p = 0.21). Median overall survival (OS) for P- and E-treated pts was 13.7 vs. 17.0 months (HR = 0.67, 95% CI 0.43-1.04, p = 0.07). Benefits from E on PFS and OS were more pronounced in pts with oropharyngeal tumors (p≤0.05 for interaction). In the E group, first-cycle rash grade 2-4 (34% pts) was associated with longer OS (HR = 0.40, p = 0.02). E-treated pts experienced a higher incidence of grade 3-4 adverse events (33.9 vs. 53.3%), including diarrhea (3 vs.17%), dehydration (5 vs. 15%), nausea (5 vs. 14%), rash (0 vs. 12%). Conclusions: This study met its primary endpoint. Addition of E to first-line platinum/docetaxel improved PFS and OS. This regimen may warrant further evaluation in randomized, phase 3 trials. Clinical trial information: NCT01064479.

Use of interval estimations in design and evaluation of multiregional clinical trials with continuous outcomes

2018 ◽  
Vol 28 (7) ◽  
pp. 2179-2195 ◽  
Author(s):  
Chieh Chiang ◽  
Chin-Fu Hsiao
Keyword(s):  
Clinical Trials ◽  
Sample Size ◽  
Type I Error ◽  
Interval Estimation ◽  
Error Rates ◽  
New Drugs ◽  
Sample Size Determination ◽  
Type I ◽  
Size Determination ◽  
Interval Estimators

Multiregional clinical trials have been accepted in recent years as a useful means of accelerating the development of new drugs and abridging their approval time. The statistical properties of multiregional clinical trials are being widely discussed. In practice, variance of a continuous response may be different from region to region, but it leads to the assessment of the efficacy response falling into a Behrens–Fisher problem—there is no exact testing or interval estimator for mean difference with unequal variances. As a solution, this study applies interval estimations of the efficacy response based on Howe’s, Cochran–Cox’s, and Satterthwaite’s approximations, which have been shown to have well-controlled type I error rates. However, the traditional sample size determination cannot be applied to the interval estimators. The sample size determination to achieve a desired power based on these interval estimators is then presented. Moreover, the consistency criteria suggested by the Japanese Ministry of Health, Labour and Welfare guidance to decide whether the overall results from the multiregional clinical trial obtained via the proposed interval estimation were also applied. A real example is used to illustrate the proposed method. The results of simulation studies indicate that the proposed method can correctly determine the required sample size and evaluate the assurance probability of the consistency criteria.

Exact tests in binary data under an incomplete block crossover design

2016 ◽  
Vol 27 (2) ◽  
pp. 579-592 ◽  
Author(s):  
Kung-Jong Lui ◽  
Kuang-Chao Chang
Keyword(s):  
Binary Data ◽  
Logistic Regression Model ◽  
Crossover Trial ◽  
Test Procedure ◽  
Crossover Design ◽  
Incomplete Block ◽  
Primary Dysmenorrhea ◽  
Test Procedures ◽  
Exact Test ◽  
High Doses

To improve the power of a parallel groups design and reduce the time length of a crossover trial, we may consider an incomplete block crossover design. Under a distribution-free random effects logistic regression model, we derive an exact test and a Mantel-Haenszel Type of summary test procedure for testing non-equality in binary data when comparing three treatments. We employ Monte Carlo simulation to evaluate the performance of these test procedures. We find that both test procedures developed here can perform well in a variety of situations. We use the data taken as a part of the crossover trial comparing the low and high doses of an analgesic with a placebo for the relief of pain in primary dysmenorrhea to illustrate the use of the proposed test procedures.

scholarly journals Non-Inferiority Test for a Continuous Variable with a Flexible Margin in an Active Controlled Trial : An Application to the "Stratall ANRS 12110 / ESTHER" Trial

2021 ◽  
Author(s):  
Arsene Sandie ◽  
Nicholas Molinari ◽  
Anthony Wanjoya ◽  
Charles Kouanfack ◽  
Christian Laurent ◽  
...  
Keyword(s):  
Confidence Interval ◽  
Active Control ◽  
Error Rate ◽  
Type I Error ◽  
Clinical Monitoring ◽  
Small Scale ◽  
Type I ◽  
Test Procedures ◽  
Power Estimate ◽  
Control Intervention

Abstract Background : The non-inferiority trials are becoming increasingly popular in public health and clinical research. The choice of the non-inferiority margin is the cornerstone of the non-inferiority trial. When the effect of active control intervention is unknown, it can be interesting to choose the non-inferiority margin as a function of the active control intervention effect. In this case, the uncertainty surrounding the non-inferiority margin should be accounted for in statistical tests. In this work, we explored how to perform the non-inferiority test with a flexible margin for continuous endpoint.Methods: It was proposed in this study two procedures for the non-inferiority test with a flexible margin for the continuous endpoint. The proposed test procedures are based on test statistic and confidence interval approach. Simulations have been used to assess the performances and properties of the proposed test procedures. An application was done on clinical real data, which the purpose was to assess the efficacy of clinical monitoring alone versus laboratory and clinical monitoring in HIV-infected adult patients.Results : Basically, the two proposed test procedures have good properties. In the test based on a statistic, the actual type 1 error rate estimate is approximatively equal to the nominal value. It has been found that the confidence interval level determines approximately the level of significance. The $80\%$ , $90\%$ , and $95\%$ one-sided confidence interval levels led approximately to a type I error of $10\%$ , $5\%$ and $2.5\%$ respectively. The power estimate was almost $100\%$ for two proposed tests, except for the small scale values of the reference treatment where the power was relatively low when the sample sizes were small.Conclusions : Based on type I error rate and power estimates, the proposed non-inferiority hypothesis test procedures have good performance and are applicable in practice.Trial registration : The trial data used in this study was from the "Stratall ANRS 12110 / ESTHER", registered with ClinicalTrials.gov, number NCT00301561. Date : March 13, 2006, url : https://clinicaltrials.gov/ct2/show/NCT00301561.

A randomized double blind phase I-II study to determine the tolerability/efficacy of two different doses of lenalidomide (L), CC- 5013, in biochemically relapsed (BR) prostate cancer (PC) patients (pts) (M0) after local treatment (LT)

2009 ◽  
Vol 27 (15_suppl) ◽  
pp. 5130-5130
Author(s):  
V. J. Sinibaldi ◽  
M. A. Carducci ◽  
S. Moore-Cooper ◽  
B. George ◽  
S. Denmeade ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Type I Error ◽  
Local Treatment ◽  
Hepatic Function ◽  
Dose Effect ◽  
Progression Rate ◽  
Type I ◽  
Antitumor Effects ◽  
Double Blind ◽  
Pooled Data

5130 Background: BR following LT is common in PC with no defined standard treatment. Lenalidomide (L) is an immunomodulatory agent with anti-angiogenic and direct antitumor effects. Methods: This trial was designed to evaluate a dose-effect relationship of L in BR PC. Pts were randomized to either 5 or 25 mg/day(d), PO, d 1–21 (28-d cycles); then stratified by PSADT (< 3, 3–8.9, ≥ 9 mos), LT and prior ADT. Eligible pts had: rising PSA (≥1 ng/mL), M0 disease, testosterone > 150 ng/mL, adequate bone marrow, renal, and hepatic function. Baseline and Q 2 mos PSA's were processed after Q 6 mos of L, along with CT and bone scan. Toxicity exams were Q mo. Primary endpoints are safety and progression after 6 mo of L (defined by a confirmed ↑ in PSA > 25% over the baseline value or mets). Secondary endpoints are changes of slopes in PSA related to pharmacokinetics (pk). A sample size of 30 pts/arm provides an 85% power to detect a PSA progression rate of 40% (compared to 80% predicted ) with a Type I error = 0.05 (Fishers exact test). Results: 59 pts were entered July 20, 2006-December 31, 2008. Pooled data from the 2 arms: median: age 64 (50–81), ECOG PS 0, baseline PSA 9.3 ng/ml (1.3–92.8 ng/ml). 16 pts had PSADT <3 mos, 26 from 3–8.9 mos, and 17 ≥ 9 mos. Median: F/U on all 59 pts is 351 + d (9 +-887+d); # cycles = 6 (1–30). Thus far, 44/59 pts completed 6 cycles of L (1 had PD, 6 stopped L due to toxicity, 8 too early). 22 /44 who completed 6 mos of L remained on L > 6 mos ( 7+-30+ mos); including 7 pts ≥ 24 mos. Of 44 pts, blinded evaluation of PSA's at 6 mos: 4 pts had ≥ 50% ↓, 22 had stable PSA,17 had PD, 1 too early . Rash was DLT. Other Gr toxicities: appendicitis, abd pain, neck pain, venous thrombolic disease, fatigue, pruritus. Conclusions: Preliminary data prior to unblinding the study treatment arms, from pooled data, suggest that L may be administered > 6 mos with acceptable toxicity, and is associated with PSA declines and long term stabilization in pts with BR. Supported by a grant from Celgene Corporation. Data coordination infrastructure is supported by the Prostate Cancer Foundation and The James Stine research fund. [Table: see text]

Paracetamol and Pain Modulation by TRPV1, UGT2B15, SULT1A1 Genotypes: A Randomized Clinical Trial in Healthy Volunteers

Pain Medicine ◽  
2019 ◽  
Vol 21 (4) ◽  
pp. 661-669 ◽  
Author(s):  
Gisèle Pickering ◽  
Isabelle Creveaux ◽  
Nicolas Macian ◽  
Bruno Pereira
Keyword(s):  
Genetic Polymorphism ◽  
Pain Relief ◽  
Healthy Volunteers ◽  
Type I Error ◽  
Pain Treatment ◽  
Pain Modulation ◽  
University Hospital ◽  
Type I ◽  
Double Blind ◽  
Paracetamol Metabolism

Abstract Background The influence of the genetic polymorphism of enzymes and receptors involved in paracetamol metabolism and mechanism of action has not been investigated. This trial in healthy volunteers investigated the link between paracetamol pain relief and the genetic polymorphism of 23 enzymes and receptors. Design This randomized double-blind crossover controlled pilot study took place in the Clinical Pharmacology Department, University Hospital, Clermont-Ferrand, France. Forty-seven Caucasian volunteers were recruited. The trial consisted of two randomized sessions one week apart with oral paracetamol or placebo, and pain changes were evaluated with mechanical pain stimuli. The genetic polymorphism of 23 enzymes and receptors was studied, and correlations were made with pain relief. All tests are two-sided with a type I error at 0.05. Results Paracetamol was antinociceptive compared with placebo (222 ± 482 kPaxmin vs 23 ± 431 kPaxmin; P = 0.0047), and the study showed 30 paracetamol responders and 17 paracetamol nonresponders. Responders were characterized by TRPV1rs224534 A allele, UGT2B15rs1902023 TT genotype, and SULT1A1rs9282861 GG genotype (P &lt; 0.05 for all). These findings confirm for the first time the involvement of a specific TRPV1 rs224534 variant in paracetamol antinociception. They also reveal a new antinociceptive role for specific variants of hepatic phase II enzymes associated with paracetamol metabolism. Conclusions The study warrants larger clinical trials on these potential genomic markers of paracetamol analgesia in patients. Confirmation of the present findings would open the way to effective individualized pain treatment with paracetamol, the most commonly used analgesic worldwide.

scholarly journals A Monte Carlo Study of an Iterative Wald Test Procedure for DIF Analysis

2016 ◽  
Vol 77 (1) ◽  
pp. 104-118 ◽  
Author(s):  
Mengyang Cao ◽  
Louis Tay ◽  
Yaowu Liu
Keyword(s):  
Monte Carlo ◽  
Type I Error ◽  
Test Procedure ◽  
Monte Carlo Study ◽  
Error Rates ◽  
Type I ◽  
Iterative Approach ◽  
Response Options ◽  
Type I Error Rates ◽  
Dichotomous Data

This study examined the performance of a proposed iterative Wald approach for detecting differential item functioning (DIF) between two groups when preknowledge of anchor items is absent. The iterative approach utilizes the Wald-2 approach to identify anchor items and then iteratively tests for DIF items with the Wald-1 approach. Monte Carlo simulation was conducted across several conditions including the number of response options, test length, sample size, percentage of DIF items, DIF effect size, and type of cumulative DIF. Results indicated that the iterative approach performed well for polytomous data in all conditions, with well-controlled Type I error rates and high power. For dichotomous data, the iterative approach also exhibited better control over Type I error rates than the Wald-2 approach without sacrificing the power in detecting DIF. However, inflated Type I error rates were found for the iterative approach in conditions with dichotomous data, noncompensatory DIF, large percentage of DIF items, and medium to large DIF effect sizes. Nevertheless, the Type I error rates were substantially less inflated in those conditions compared with the Wald-2 approach.

Sign in / Sign up

Export Citation Format

Share Document