Two-stage Group Sequential Design Research Articles

Two-stage group-sequential designs with delayed responses – what is the point of applying corresponding methods?

BackgroundIn group-sequential designs, it is typically assumed that there is no time gap between patient enrollment and outcome measurement in clinical trials. However, in practice, there is usually a lag between the two time points. This can affect the statistical analysis of the data, especially in trials with interim analyses. One approach to address delayed responses has been introduced by Hampson and Jennison (J R Stat Soc Ser B Stat Methodol 75:3-54, 2013), who proposed the use of error-spending stopping boundaries for patient enrollment, followed by critical values to reject the null hypothesis if the stopping boundaries are crossed beforehand. Regarding the choice of a trial design, it is important to consider the efficiency of trial designs, e.g. in terms of the probability of trial success (power) and required resources (sample size and time).MethodsThis article aims to shed more light on the performance comparison of group sequential clinical trial designs that account for delayed responses and designs that do not. Suitable performance measures are described and designs are evaluated using the R package rpact. By doing so, we provide insight into global performance measures, discuss the applicability of conditional performance characteristics, and finally whether performance gain justifies the use of complex trial designs that incorporate delayed responses.ResultsWe investigated how the delayed response group sequential test (DR-GSD) design proposed by Hampson and Jennison (J R Stat Soc Ser B Stat Methodol 75:3-54, 2013) can be extended to include nonbinding lower recruitment stopping boundaries, illustrating that their original design framework can accommodate both binding and nonbinding rules when additional constraints are imposed. Our findings indicate that the performance enhancements from methods incorporating delayed responses heavily rely on the sample size at interim and the volume of data in the pipeline, with overall performance gains being limited.ConclusionThis research extends existing literature on group-sequential designs by offering insights into differences in performance. We conclude that, given the overall marginal differences, discussions regarding appropriate trial designs can pivot towards practical considerations of operational feasibility.

Sample size recalculation in three-stage clinical trials and its evaluation

BackgroundIn clinical trials, the determination of an adequate sample size is a challenging task, mainly due to the uncertainty about the value of the effect size and nuisance parameters. One method to deal with this uncertainty is a sample size recalculation. Thereby, an interim analysis is performed based on which the sample size for the remaining trial is adapted. With few exceptions, previous literature has only examined the potential of recalculation in two-stage trials.MethodsIn our research, we address sample size recalculation in three-stage trials, i.e. trials with two pre-planned interim analyses. We show how recalculation rules from two-stage trials can be modified to be applicable to three-stage trials. We also illustrate how a performance measure, recently suggested for two-stage trial recalculation (the conditional performance score) can be applied to evaluate recalculation rules in three-stage trials, and we describe performance evaluation in those trials from the global point of view. To assess the potential of recalculation in three-stage trials, we compare, in a simulation study, two-stage group sequential designs with three-stage group sequential designs as well as multiple three-stage designs with recalculation.ResultsWhile we observe a notable favorable effect in terms of power and expected sample size by using three-stage designs compared to two-stage designs, the benefits of recalculation rules appear less clear and are dependent on the performance measures applied.ConclusionsSample size recalculation is also applicable in three-stage designs. However, the extent to which recalculation brings benefits depends on which trial characteristics are most important to the applicants.

Protocol for CHAMPION study: a prospective study of maximal-cytoreductive therapies for patients with de novo metastatic hormone-sensitive prostate cancer who achieve oligopersistent metastases during systemic treatment with apalutamide plus androgen deprivation therapy

BackgroundThe proposed trial is to examine the feasibility of prostate-specific membrane antigen (PSMA) positron emission tomography/computed tomography (PET/CT)-guided cytoreduction plus apalutamide and androgen deprivation therapy (ADT) for newly diagnosed metastatic hormone-sensitive prostate cancer (mHSPC) at oligometastatic state.MethodsCHAMPION (NCT05717582) is an open-label, single-arm, phase II trial, planning to enroll newly diagnosed mHSPC cases with oligometastases (≤ 10 distant metastatic sites in conventional imaging). Patients will receive 6 cycles of apalutamide plus ADT. Patients with oligometastatic disease at PSMA PET/CT after 3 treatment cycles will receive cytoreductive radical prostatectomy. PSMA PET/CT-guided metastasis-directed external radiation therapy will be determined by the investigators. Apalutamide plus ADT will be continued for 2 weeks postoperatively. The primary endpoint is the proportion of patients with undetectable prostate-specific antigen (PSA), no disease progression, and no symptom deterioration after 6 cycles of apalutamide plus ADT. Secondary endpoints include the percentage of patients with PSA ≤ 0.2 ng/mL and oligometastases by the end of 3 treatment cycles, PSA response rate, and safety. Fleming’s two-stage group sequential design will be adopted in the study, where the null hypothesis is that the rate of patients with an undetectable PSA is ≤ 40% after 6 cycles of treatment, while the alternate hypothesis is an undetectable PSA of > 60%; with one-sided α = 0.05, power = 0.80, and an assumed dropout rate of 10%, the required number of patients for an effective analysis is 47. Enrolment in the study commenced in May 2023.DiscussionThe multi-modal therapy based on treatment response may improve the prognosis of newly diagnosed mHSPC patients with oligometastases.Trial registrationThe study is registered with Clinical Trials.Gov (NCT05717582). Registered on 8th February 2023.

A two-stage group-sequential design for delayed treatment responses with the possibility of trial restart.

Common statistical theory applicable to confirmatory phase III trial designs usually assumes that patients are enrolled simultaneously and there is no time gap between enrollment and outcome observation. However, in practice, patients are enrolled successively and there is a lag between the enrollment of a patient and the measurement of the primary outcome. For single-stage designs, the difference between theory and practice only impacts on the trial duration but not on the statistical analysis and its interpretation. For designs with interim analyses, however, the number of patients already enrolled into the trial and the number of patients with available outcome measurements differ, which can cause issues regarding the statistical analyses of the data. The main issue is that current methodologies either imply that at the time of the interim analysis there are so-called pipeline patients whose data are not used to make a statistical decision (like stopping early for efficacy) or the enrollment into the trial needs to be at least paused for interim analysis to avoid pipeline patients. There are methods for delayed responses available that introduced error-spending stopping boundaries for the enrollment of patients followed by critical values to reject the null hypothesis in case the stopping boundaries have been crossed beforehand. Here, we will discuss other solutions, considering different boundary determination algorithms using conditional power and introducing a design allowing for recruitment restart while keeping the type I error rate controlled.

Pylorus resection versus pylorus preservation in pancreatoduodenectomy (PyloResPres): study protocol and statistical analysis plan for a German multicentre, single-blind, surgical, registry-based randomised controlled trial

IntroductionPartial pancreatoduodenectomy (PD) is the treatment of choice for various benign and malignant tumours of the pancreatic head or the periampullary region. For reconstruction of the gastrointestinal passage, two stomach-preserving...

Interval and point estimation in adaptive Phase II trials with binary endpoint.

Phase II clinical trials are concerned with making decision of whether a treatment is sufficiently efficacious to be worth further investigations in late large scale Phase III trials. In oncology Phase II trials, frequentist single-arm two-stage group-sequential designs with a binary endpoint are commonly used. To allow for more flexibility, adaptive versions of these designs have been proposed. In this paper, we propose point and interval estimation for adaptive designs in which the second stage sample size is a pre-specified function of first stage's number of responses. Our approach is based on sample space orderings, from which we derive p-values, and point and interval estimates. Simulation studies show that our proposed methods perform better, in terms of bias and root mean square error, than the fixed-sample maximum likelihood estimator.

Early response rates and Curie scores at end of induction: An update from a phase II study of an anti-GD2 monoclonal antibody (mAb) with chemotherapy (CT) in newly diagnosed patients (pts) with high-risk (HR) neuroblastoma (NB).

10534 Background: We are conducting a Phase II trial of the anti-GD2 mAb hu14.18K322A given concomitantly with CT in newly diagnosed pts with HR NB. Our primary objective is to compare the response rate (RR; defined as ≥ PR) after two courses of CT with cyclophosphamide (CTX)/topotecan (TPT) and hu14.18K322A with GM-CSF and IL-2 to the RR reported by Park et al. (JCO 29:4351, 2011), using identical CT with GCSF only, in a two-stage group sequential design. Semiquantitative MIBG scoring (Curie; CS) has been shown to be a prognostic indicator of outcome in pts with HR NB (Yanik et al, J Nucl Med, 2013), particularly, those with CS > 2 at end of induction (EoI) CT have inferior outcomes. Here we update the RR and report the CS at EoI CT. Methods: Pts received induction CT (6 cycles) as described by Park et al, with the addition of hu14.18K332A (on d 2-5; 40mg/m2/d x 4d), daily sc GM-CSF and sc IL-2 (1 x 106 IU/m2/dose) qod x 6. 123I-MIBG scans and scoring (CS) were obtained on all pts at diagnosis, after second course of CT and at EoI. Results: 42 evaluable pts completed the first two courses of CT (24 male, median age 2.9 yrs (range 6 m -15.2 yrs), 36 INSS 4, 19 MYCNamplified); 40/42 evaluable pts had measurable reductions in primary tumor volume after two courses of CT (median 79%, average 69%, range +5 – 100%). Responses (≥ PR) after two courses were seen in 32/42 (76.2%; 95% CI 61.3 – 87.9%); No mAb dose-reductions were made but 20/43 (47%) pts had infusion times extended. The development of human anti-human antibody reactivity (HAHA) to hu14.18K322A in the first 22 pts is minimal. Of the 36 INSS 4 pts, 1 withdrew prior to EoI and 2 are too early. The median CS at diagnosis for the 33 stage 4 pts who have completed induction CT was 18 (range 1 – 28). At EoI the median CS was 0 (range 0 -23); 29/33 INSS 4 pts had EoI CS ≤ 2. Conclusions: The addition of hu14.18K322A to two courses of CTX/TPT significantly improves the RR compared to two courses of CTX/TPT alone (32/42 vs 12/30 as reported by Park et al,; P = 0.000004). The improved median CS of the stage 4 patients from 18 at diagnosis to 0 at EoI suggest the improvement in early RR may translate into improved EFS as well. Clinical trial information: NCT01857934.

Unbiased estimation of biomarker panel performance when combining training and testing data in a group sequential design.

Motivated by an ongoing study to develop a screening test able to identify patients with undiagnosed Sjögren's Syndrome in a symptomatic population, we propose methodology to combine multiple biomarkers and evaluate their performance in a two-stage group sequential design that proceeds as follows: biomarker data is collected from first stage samples; the biomarker panel is built and evaluated; if the panel meets pre-specified performance criteria the study continues to the second stage and the remaining samples are assayed. The design allows us to conserve valuable specimens in the case of inadequate biomarker panel performance. We propose a nonparametric conditional resampling algorithm that uses all the study data to provide unbiased estimates of the biomarker combination rule and the sensitivity of the panel corresponding to specificity of 1-t on the receiver operating characteristic curve (ROC). The Copas and Corbett (2002) correction, for bias resulting from using the same data to derive the combination rule and estimate the ROC, was also evaluated and an improved version was incorporated. An extensive simulation study was conducted to evaluate finite sample performance and propose guidelines for designing studies of this type. The methods were implemented in the National Cancer Institutes Early Detection Network Urinary PCA3 Evaluation Trial.

Stepwise two‐stage sample size adaptation

Several adaptive design methods have been proposed to reestimate sample size using the observed treatment effect after an initial stage of a clinical trial while preserving the overall type I error at the time of the final analysis. One unfortunate property of the algorithms used in some methods is that they can be inverted to reveal the exact treatment effect at the interim analysis. We propose using a step function with an inverted U-shape of observed treatment difference for sample size reestimation to lessen the information on treatment effect revealed. This will be referred to as stepwise two-stage sample size adaptation. This method applies calculation methods used for group sequential designs. We minimize expected sample size among a class of these designs and compare efficiency with the fully optimized two-stage design, optimal two-stage group sequential design, and designs based on promising conditional power. The trade-off between efficiency versus the improved blinding of the interim treatment effect will be discussed.

A phase II evaluation of aflibercept in the treatment of recurrent or persistent endometrial cancer: A Gynecologic Oncology Group study

ObjectivesAflibercept targets vascular endothelial growth factor and placental growth factor. We evaluated activity and toxicity of aflibercept in recurrent/persistent endometrial cancer patients. Biomarkers and association with clinical characteristics and outcome were explored. MethodsEligible patients had measurable disease; 1–2 prior cytotoxic regimens; performance status 0–2. Aflibercept 4mg/kg IV q14 days (28-day cycles) was administered until disease progression or prohibitive toxicity. Primary endpoints were the proportion of patients with progression-free survival at 6months (PFS6) and tumor response rate. A flexible two-stage group sequential design to detect 20% increases in the proportion of patients responding or enduring PFS6 with 90% power (α=10%) was employed. ResultsForty-nine patients were enrolled; five were excluded: wrong primary (2), second primary (1), wrong cell type (1); and never treated (1). Median age was 64 (range 48–83). Eighteen patients (41%) had two prior regimens; 27 (61%) had prior radiation. The PFS6 rate was 41%; three patients (7%, 90% CI: 2–17) had partial response. Of note, 10 patients (23%) met the PFS6 endpoint without starting a subsequent therapy; the remaining eight patients discontinued therapy for toxicity and started another therapy before 6months elapsed. Median PFS and overall survival were 2.9months and 14.6months, respectively. Significant grade 3/4 toxicities were: cardiovascular (23%/5%), constitutional (7%/0), hemorrhage (2%/5%), metabolic (7%/2%), and pain (18%/0). Two treatment-related deaths were recorded: GI perforation (1), and arterial rupture (1). FGF1 expression was associated with response. ConclusionsAflibercept met pretrial activity parameters, but was associated with significant toxicity at this dose and schedule in this population.

Summary Workshop Report: Facilitating Oral Product Development and Reducing Regulatory Burden Through Novel Approaches to Assess Bioavailability/Bioequivalence

This summary workshop report highlights presentations and over-arching themes from an October 2011 workshop. Discussions focused on best practices in the application of biopharmaceutics in oral drug product development and evolving bioequivalence approaches. Best practices leverage biopharmaceutic data and other drug, formulation, and patient/disease data to identify drug development challenges in yielding a successfully performing product. Quality by design and product developability paradigms were discussed. Development tools include early development strategies to identify critical absorption factors and oral absorption modeling. An ongoing theme was the desire to comprehensively and systematically assess risk of product failure via the quality target product profile and root cause and risk analysis. However, a parallel need is reduced timelines and fewer resources. Several presentations discussed applying Biopharmaceutics Classification System (BCS) and in vitro-in vivo correlations in development and in post-development and discussed both resource savings and best scientific practices. The workshop also focused on evolving bioequivalence approaches, with emphasis on highly variable products (HVDP), as well as specialized modified-release products. In USA, two bioequivalence approaches for HVDP are the reference-scaled average bioequivalence approach and the two-stage group-sequential design. An adaptive sequential design approach is also acceptable in Canada. In European Union, two approaches for HVDP are a two-stage design and an approach to widen C (max) acceptance limits. For some specialized modified-release products, FDA now requests partial area under the curve. Rationale and limitations of such metrics were discussed (e.g., zolpidem and methylphenidate). A common theme was the benefit of the scientific and regulatory community developing, validating, and harmonizing newer bioequivalence methodologies (e.g., BCS-based waivers and HVDP trial designs).

A phase I study with an expanded cohort to assess the feasibility of intravenous paclitaxel, intraperitoneal carboplatin and intraperitoneal paclitaxel in patients with untreated ovarian, fallopian tube or primary peritoneal carcinoma: A Gynecologic Oncology Group study

ObjectiveTo define the maximum tolerated dose (MTD) and assess the feasibility of intravenous (IV) paclitaxel, intraperitoneal (IP) carboplatin, and IP paclitaxel in women with newly diagnosed Stages II–IV ovarian, fallopian tube, or primary peritoneal carcinoma. MethodsPatients received escalating doses of paclitaxel IV and carboplatin IP on day 1 and paclitaxel IP 60mg/m2 on day 8. A standard 3+3 design was used in the escalation phase. A two-stage group sequential design with 20 patients at the MTD was used in the feasibility phase. Patient-reported neurotoxicity was assessed pre and post treatment. ResultsPatients were treated with paclitaxel 175mg/m2 IV and carboplatin IP from AUC 5–7 on day 1 and paclitaxel 60mg/m2 IP on day 8. The MTD was estimated at carboplatin AUC 6 IP and 25 patients enrolled at this dose level. Within the first 4cycles, seven (35%) of twenty evaluable patients had dose-limiting toxicities (DLTs) including grade 4 thrombocytopenia (1), grade 3 neutropenic fever (3), >2week delay due to ANC recovery (1), grade 3 LFT (1), and grade 3 infection (1). De-escalation to paclitaxel 135mg/m2 IV was given to improve the safety. After six evaluable patients completed 4cycles without a DLT, bevacizumab was added and six evaluable patients completed 4cycles with one DLT (grade 3 hyponatremia). ConclusionsPaclitaxel at 175mg/m2 IV, carboplatin AUC 6 IP day 1 and paclitaxel 60mg/m2 IP day 8 yield 18–56% patients with DLTs. The tolerability of the regimen in combination with bevacizumab was indicated in a small cohort.

A randomized crossover trial of venlafaxine (V) versus gabapentin (G) for hot flashes (HF) in breast cancer survivors.

9023 Background: Nonhormonal pharmacologic interventions are recommended for the treatment of HF in breast cancer survivors. Antidepressants (SSRI/SNRI) and G have been shown to be effective, however, it is still unclear which agent is best. Methods: A randomized crossover trial compared 4 weeks of V (37.5 mg × 7 days followed by 75 mg/day) and G (300 mg daily × 3 days, 300 mg BID × 3 days then 300 mg TID). Eligibility included postmenopausal women with at least 14 bothersome HF per week for the prior month. A 2-week baseline period was followed by the first treatment period. Crossover to the alternative treatment occurred following a 2 week dose tapering/wash out period. The primary endpoint was patient preference. Secondary endpoints included change from baseline in HF score (frequency × severity) per a prospective daily HF diary, change in HF frequency, QOL, mood and toxicity. Toxicities were evaluated with a self-report questionnaire and NCI common terminology criteria. A two-stage group sequential design was used with the first analysis pre-planned at 66 patients and the second (if needed) at 131, allowing for drop out/indecision rate of 25%. The design had 80% power to detect a shift from equal preference of ≥ 15 % (2 sided type 1 error at 5%). Results: Data analysis after 66 patients showed the pre-defined stopping rule was satisfied. The study arms were well balanced. Of 56 patients with a preference (8 dropped out and 2 had no preference), 18 preferred G and 38 preferred V (32% and 68% respectively; p = 0.01). Both V and G reduced HF scores (66% reduction) and frequency (55% reduction) similarly and significantly. Satisfaction with the preferred treatment was high. V was associated with more appetite loss, nausea and constipation, and less sleeplessness, nervousness, and negative mood changes than at baseline (all p ≤ 0.03) whereas G was associated with more dizziness and increased appetite, and less sleeplessness (all p ≤ 0.004). Conclusions: Breast cancer survivors prefer V over G for the treatment of HF. Both agents are effective in the management of HF based on recorded diary experience. Preference appears to be driven by the balance between effectiveness and tolerance. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration AstraZeneca, GlaxoSmithKline, Novartis, Roche Amgen, AstraZeneca, GlaxoSmithKline, Novartis, Roche AstraZeneca

Optimal Two-Stage Designs to Evaluate a Series of New Agents or Treatments

Recent developments in cancer vaccines and other research programs have prompted the need for screening a large number of agents (or treatments) to identify promising ones for further clinical investigation. This article proposes optimal two-stage group sequential designs that minimize the expected number of patients exposed to inactive agents. The proposed three-outcome, two-stage designs include some region of uncertainty which enables factors such as cost of the agent and convenience of administration to be used in the decision making when the observed response rate falls into the uncertainty region. In the special case of two-outcome designs, our proposal extends Simon’s optimal two-stage design and other existing screening trial designs that use available distributional information on response rates. It does so by allowing early termination of the trial when the observed response rate is below a certain threshold and by providing more choices of error rates to control. A computer algorithm was developed to effectively calculate the three-outcome and/or two-outcome optimal designs. It is demonstrated via numerical studies and a cancer vaccine trial.

Two-Stage Group Sequential Designs for Comparative Clinical Trials

Group sequential designs are widely used in Phase II clinical trials, which are usually undertaken to evaluate the response probability of specific treatment regimen. In most randomized clinical trials with sequential patient entry, fixed sample size design is unjustified on ethical grounds and sequential designs are often impractical. However group sequential designs are generally more practical and they provide much of the saving possible from sequential designs. Optimal restricted two-stage design is the simplest form of a group sequential design. In this study, group sequential design obtained by α*(t) functions characterized using the type-I error probability and optimal restricted two stage design has been compared for the cases that the group sizes are equal. Furthermore, their efficiency regarding fixed sample size design has been calculated and the results have been discussed.

Optimal two-stage group-sequential designs

We derive optimal two-stage adaptive group-sequential designs for normally distributed data which achieve the minimum of a mixture of expected sample sizes at the range of plausible values of a normal mean. Unlike standard group-sequential tests, our method is adaptive in that it allows the group size at the second look to be a function of the observed test statistic at the first look. Using optimality criteria, we construct two-stage designs which we show have advantage over other popular adaptive methods. The employed computational method is a modification of the backward induction algorithm applied to a Bayesian decision problem.

Multicenter phase II study of a cremophor-free polymeric micelle-formulated paclitaxel in patients (pts) with metastatic breast cancer (MBC)

10520 Background: The results of a phase I trial of a novel taxane, a cremophor-free polymeric micelle-formulated paclitaxel (Genexol-PM), were previously reported (Clin Cancer Res 10:3708, 2004). The dose limiting toxicities were neuropathy, myalgia, and neutropenia. Notably, acute hypersensitivity reaction was not observed. Methods: This was a single arm phase II study performed at 6 major hospitals in Korea to evaluate the efficacy and safety of Genexol-PM as a 1st- or 2nd-line therapy for MBC. Genexol-PM with no premedication was given at 300 mg/m2 i.v. over 3 hr on day 1 every 3 weeks until disease progression or intolerability. Two-stage group sequential design was used. If more than 14 responses were attained in the first 30 pts, early termination was possible. Results: Forty-three pts with MBC were enrolled from 7/2004 to 12/2004. Median follow-up duration was 7.0 mo. Thirty-seven of 41 pts excluding 2 pts who did not receive Genexol-PM received Genexol-PM as a first-line therapy. Nineteen pts received prior anthracycline containing regimen in adjuvant (15) and in metastatic (4) settings. No pts were anthracycline-resistant. Median age was 48 years (range; 30 to 69 years) with ECOG PS 0 or 1. A total of 331 cycles were administered, with median of 8 per patient (range; 1 to 16). Of 39 pts evaluable for efficacy, there were 5 CRs /19 PRs with response rate of 61.5%, 13 SDs and 2 PDs. Median response duration was not reached (range; 1.9+ to 10.0+ mo). Median TTP and OS were 9.8 and 11.2 mo, respectively. G3 non-hematologic toxicities included peripheral sensory neuropathy (51.2%) and myalgia (2.4%). Hematologic toxicities were G3/G4 neutropenia (51.2%, 17.1%, respectively) and G1/2 thrombocytopenia (22.0%). Seventeen pts (41.5%) required dose reductions due to G3 neuropathy. No febrile neutropenia was observed. Eight pts developed hypersensitivity reactions with G3 in 2 pts. Mean dose intensity (DI) was 87.4 mg/m2/wk (range; 27.7 to 116.7) at 87.4% of planned DI. Conclusions: Genexol-PM at 300 mg/m2 appears a promising agent with high response rate for anthracycline sensitive MBC. Further trials at a different dose schedule or duration of delivery, or in combination with other drugs are warranted. Supported by Samyang Co, Korea. No significant financial relationships to disclose.

A Phase II study of St. John's Wort for smoking cessation

To examine the feasibility and efficacy of St. John's Wort (SJW) for smoking cessation. This one-arm Phase II study utilized an exact two-stage group sequential design with a 1-week run-in period between the start of SJW treatment and the designated quit date. A total of 37 smokers (ages 18-65 years, smoking > or = 10 cigarettes/day) were started on SJW. Thirteen failed to make a verified quit attempt on the predesignated date and were taken off study resulting in 24 evaluable subjects. Smokers completed clinic visits at a cancer center with interval telephone calls and mailings. Standardized SJW, 450 mg capsules taken orally twice daily along with cessation counseling messages. Subjects completed validated surveys and a focused physical examination at baseline. Evaluable subjects were defined as those subjects who made a confirmed quit attempt on their "quit date" 1 week following initiation of SJW. Smoking status was determined through self-report and bioverification using carbon monoxide (CO) testing. Among evaluable subjects, the 12-week quit rate was 37.5% (9/24). Quitters had no significant change in weight from baseline to 12-weeks cessation. Use of SJW was generally well tolerated. Based upon these results (which suggest that SJW may be effective in maintaining smoking cessation) and the high compliance and few AEs, we conclude that SJW demonstrates feasibility for use in smoking cessation. If SJW proves to be effective in larger controlled studies, it could represent a less expensive, more readily accessible and well-tolerated agent to promote tobacco cessation.

Internally Adaptive Designs for Parallel Group Trials

This paper discusses the necessity of developing designs for clinical studies that make possible an adaptation of the design parameters to findings at planned interim analyses. The adaptation should 1) provide robustness against wrong initial assumptions and 2) reduce the number of patients needed in general without sacrificing basic scientific and statistical principles. Such designs will be called internally adaptive. As an example, a two-stage design is proposed that, in contrast to usual two-stage group sequential designs, makes possible the adaptation of sample size for the final analysis to the observed treatment difference at the interim analysis. A three-stage design that combines this procedure with an interim estimate of variability proves to have a power that is robust against initial underestimation of variability and to reduce expected sample size substantially relative to an adaptive procedure proposed by Wittes and Brittain (1).

The analysis of multiple end points with a two-stage group sequential design

The analysis of multiple end points with a two-stage group sequential design