Optimal two-stage designs for phase II clinical trials
Abstract
The primary objective of a phase II clinical trial of a new drug or regimen is to determine whether it has sufficient biological activity against the disease under study to warrant more extensive development. Such trials are often conducted in a multi-institution setting where designs of more than two stages are difficult to manage. This paper presents two-stage designs that are optimal in the sense that the expected sample size is minimized if the regimen has low activity subject to constraints upon the size of the type 1 and type 2 errors. Two-stage designs which minimize the maximum sample size are also determined. Optimum and “minimax” designs for a range of design parameters are tabulated. These designs can also be used for pilot studies of new regimens where toxicity is the endpoint of interest.
References (15)
- E.A. Gehan
The determination of the number of patients required in a follow-up trial of a new chemotherapeutic agent
J Chron Dis
(1961) - M.N. Chang et al.
Confidence intervals following group sequential tests
Controlled Clin Trials
(1986) - R. Simon
Design, Analysis and Reporting of Cancer Clinical Trials
- T. Colton et al.
Two-stage plans compared with fixed-sample-size and Wald SPRT plans
J Am Stat Assoc
(1976) - K. McPherson
On choosing the number of interim analyses in clinical trials
Stat Med
(1982) - C. Jennison
Efficient group sequential tests with unpredictable group sizes
Biometrika
(1987) - C. Jennison et al.
Confidence interbals for a binomial parameter following a multistage test with application to MIL-STD 105D and medical trials
Technometrics
(1983)
Cited by (3217)
Optimal timing for an accelerated interim futility analysis incorporating real world data
2024, Contemporary Clinical TrialsRandomized controlled trials include interim monitoring guidelines to stop early for safety, efficacy, or futility. Futility monitoring facilitates re-allocation of limited resources. However, conventional methods for interim futility monitoring require a trial to accrue nearly half of the outcome data to make a reliable early stopping decision, limiting its benefit. As early stopping for futility will not inflate type-I error, these analyses are an appealing venue for incorporating external data to improve efficiency.
We propose a Bayesian approach to futility monitoring leveraging real world data using Semi-Supervised MIXture Multi-source Exchangeability Models, which accounts for both measured and unmeasured differences between data sources. We implement futility monitoring using predictive probabilities and investigate the optimal timing with respect to the expected sample size under the null hypothesis. Because we only incorporate external data during the interim futility analysis the proposed design is not limited by type-I error inflation.
When the external and trial data are exchangeable, the proposed method provides a roughly 70 person reduction in expected sample size under the null. Under scenarios where exchangeability does not hold, our approach still provides a 10–20 person reduction in expected sample size under the null with about 80% power.
External data borrowing in interim futility monitoring is a promising venue to improve trial efficiency without type-I error inflation. Approaches that are acceptable to regulatory authorities and leverage the complementary strengths of real world and trial data are vital to more efficiently allocate limited resources amongst clinical trials.
Single-agent oral vinorelbine is a standard of care for hormone receptor (HR)-positive/human epidermal growth factor receptor 2 (HER2)-negative advanced breast cancer (ABC) that has progressed on endocrine therapy. Metronomic administration may offer a better balance of efficacy and safety than standard regimens, but data from previous trials are scarce.
In this open-label, multicenter, phase II trial, patients were randomized to oral vinorelbine administered on a metronomic (50 mg three times weekly) or weekly (60 mg/m2 in cycle 1, increasing to 80 mg/m2 if well tolerated) schedule. Treatment was continued until disease progression or intolerance. The primary endpoint was disease control rate (DCR, the proportion of patients with a best overall confirmed response of CR, PR, or stable disease lasting 6 months or more).
One-hundred sixty-three patients were randomized and treated. The DCR was 63.4% (95% confidence interval [CI]: 52.0–73.8) with metronomic vinorelbine and 72.8% (95% CI: 61.8–82.1) with weekly vinorelbine. Weekly vinorelbine was also associated with longer progression-free survival (5.6 vs 4.0 months) and overall survival (26.7 vs 22.3 months) than metronomic vinorelbine, but was associated with more adverse events.
In this randomized phase II trial, single-agent metronomic oral vinorelbine was effective and well tolerated as first-line chemotherapy for patients with HR-positive/HER2-negative ABC. Formal comparisons are not done in this phase II study and one can simply observe that confidence intervals of all endpoints overlap. When deciding for a chemotherapy after failure of endocrine therapy and CDK 4/6 inhibitors, oral vinorelbine might be an option to be given with either schedule.
EudraCT 2014-003860-19.
No randomised controlled trial has ever been done in patients with metastatic phaeochromocytomas and paragangliomas. Preclinical and first clinical evidence suggested beneficial effects of sunitinib. We aimed to evaluate the safety and efficacy of sunitinib in patients with metastatic phaeochromocytomas and paragangliomas.
FIRSTMAPPP is a multicentre, international, randomised, placebo-controlled, double-blind, phase 2 trial done at 14 academic centres across four European countries. Eligible participants were adults (aged ≥18 years) with sporadic or inherited progressive metastatic phaeochromocytomas and paragangliomas. Patients were randomly assigned (1:1) to receive either oral sunitinib (37·5 mg per day) or placebo. Randomisation was stratified according to SDHB status (mutation present vs wild type) and number of previous systemic therapies (0 vs ≥1). Primary endpoint was the rate of progression-free survival at 12 months according to real-time central review (Response Evaluation Criteria in Solid Tumours version 1.1). On the basis of a two-step Simon model, we aimed for the accrual of 78 patients, assuming a 20% improvement of the 12-month progression-free survival rate from 20% to 40%, to conclude that sunitinib is effective. Crossover from the placebo group was allowed. This trial is registered with ClinicalTrials.gov, number NCT01371201, and is closed for enrolment.
From Dec 1, 2011, to Jan 31, 2019, a total of 78 patients with progressive metastatic phaeochromocytomas and paragangliomas were enrolled (39 patients per group). 25 (32%) of 78 patients had germline SDHx variants and 54 (69%) had used previous therapies. The primary endpoint was met, with a 12-month progression-free survival in 14 of 39 patients (36% [90% CI 23–50]) in the sunitinib group. In the placebo group, the 12-month progression-free survival in seven of 39 patients was 19% (90% CI 11–31), validating the hypotheses of our study design. The most frequent grade 3 or 4 adverse events were asthenia (seven [18%] of 39 and one [3%] of 39), hypertension (five [13%] and four [10%]), and back or bone pain (one [3%] and three [8%]) in the sunitinib and placebo groups, respectively. Three deaths occurred in the sunitinib group: these deaths were due to respiratory insufficiency, amyotrophic lateral sclerosis, and rectal bleeding. Only the latter event was considered drug related. Two deaths occurred in the placebo group due to aspiration pneumonia and septic shock.
This first randomised trial supports the use of sunitinib as the medical option with the highest level of evidence for anti-tumour efficacy in progressive metastatic phaeochromocytomas and paragangliomas.
French Ministry of Health, through the National Institute for Cancer, German Ministry of Education and Research, and the German Research Foundation within the CRC/Transregio 205/2, EU Seventh Framework Programme, and a private donator grant.
Older patients with classical Hodgkin lymphoma (cHL) require more effective and less toxic therapies than younger patients. In this multicenter, prospective, phase 2 study, we investigated a new firstline therapy regimen comprising 6 cycles of prednisone (40 mg/m2, days 1-5), vinblastine (6 mg/m2, day 1), doxorubicin (40 mg/m2, day 1), and bendamustine (120 mg/m2, day 1) (PVAB regimen) every 21 days for patients with newly diagnosed cHL aged ≥61 years with an advanced Ann Arbor stage. A Mini Nutritional Assessment score ≥17 was the cutoff value for patients aged ≥70 years. The primary end point was the complete metabolic response (CMR) rate after 6 cycles. The median age of the 89 included patients was 68 years (range, 61-88 years), with 35 patients (39%) aged ≥70 years. Seventy-eight patients (88%) completed the 6 cycles. The toxicity rate was acceptable, with a 20% rate of related serious adverse events. CMR was achieved by 69 patients (77.5%; 95% confidence interval [CI], 67-86). After a median follow-up of 42 months, 31 patients progressed or relapsed (35%), and 24 died (27%) from HL (n = 11), toxicity during treatment (n = 4), secondary cancers (n = 6), or other causes (n = 3). The 4-year progression-free survival (PFS) and overall survival rates were 50% and 69%, respectively. Multivariate analysis showed that liver involvement (P = .001), lymphopenia (P = .001), CRP (P = .0005), and comedications (P = .003) were independently associated with PFS. The PVAB regimen yielded a high CMR rate with acceptable toxicity. Over long-term follow-up, survival end points were influenced by unrelated lymphoma events. This trial was registered at www.clinicaltrials.gov as #NCT02414568 and at EudraCT as 2014-001002-17.
Single-center randomized trial of T-reg graft alone vs T-reg graft plus tacrolimus for the prevention of acute GVHD
2024, Blood AdvancesAllogeneic hematopoietic cell transplantation (HCT) is a curative therapy for hematological malignancies for which graft-versus-host disease (GVHD) remains a major complication. The use of donor T-regulatory cells (Tregs) to prevent GVHD appears promising, including in our previous evaluation of an engineered graft product (T-reg graft) consisting of the timed, sequential infusion of CD34+ hematopoietic stem cells and high-purity Tregs followed by conventional T cells. However, whether immunosuppressive prophylaxis can be removed from this protocol remains unclear. We report the results of the first stage of an open-label single-center phase 2 study (NCT01660607) investigating T-reg graft in myeloablative HCT of HLA-matched and 9/10-matched recipients. Twenty-four patients were randomized to receive T-reg graft alone (n = 12) or T-reg graft plus single-agent GVHD prophylaxis (n = 12) to determine whether T-reg graft alone was noninferior in preventing acute GVHD. All patients developed full-donor myeloid chimerism. Patients with T-reg graft alone vs with prophylaxis had incidences of grade 3 to 4 acute GVHD of 58% vs 8% (P = .005) and grade 3 to 4 of 17% vs 0% (P = .149), respectively. The incidence of moderate-to-severe chronic GVHD was 28% in the T-reg graft alone arm vs 0% with prophylaxis (P = .056). Among patients with T-reg graft and prophylaxis, CD4+ T-cell–to–Treg ratios were reduced after transplantation, gene expression profiles showed reduced CD4+ proliferation, and the achievement of full-donor T-cell chimerism was delayed. This study indicates that T-reg graft with single-agent tacrolimus is preferred over T-reg graft alone for the prevention of acute GVHD. This trial was registered at www.clinicaltrials.gov as #NCT01660607.
Efficacy of the Rehabilitation Planning Consult for Survivors of Head and Neck Cancer: A Phase 2 Randomized Controlled Trial
2024, International Journal of Radiation Oncology Biology PhysicsSurvivors of head and neck cancer may have significant lasting impairments and poor access to rehabilitation. To address this, our group developed and evaluated a rehabilitation planning consult (RPC). The RPC is conducted through an initial consultation and a single follow-up session with a rehabilitation professional. During the initial consultation, rehabilitation needs are determined and the survivor sets individualized goals and plans. They then implement their plans independently and are facilitated to evaluate and modify plans as necessary during the follow-up session.
We used a waitlist control design to compare the proportion of participants attaining a minimally importantly different change in quality of life (QOL) on the Short Form 36 Physical Health Summary Score from baseline to 3 months after study enrollment, between patients randomized to receive (n = 77) or wait 14 ± 3 weeks to receive (n = 76) the RPC. Additional outcomes included goal attainment indicators measured using the Brief Rehabilitation Assessment for Survivors of Head and Neck Cancer (BRASH).
Of 153 participants recruited, 95 (62%) completed the intervention; 57 were in the immediate (RPC) group and 38 were in the waiting list control (WLC) group. No significant between-group differences were seen in the proportion of patients achieving a minimally important improvement (2.5 units) on the Physical Health Summary Score from baseline to 3 months after recruitment. No between-group differences were seen on any secondary QOL indicators. Among the 67 (RPC n = 42, WLC n = 22) participants who set individualized rehabilitation goals, BRASH scores on goal performance and satisfaction with goal performance were significantly better in the RPC group.
Our results suggest that the RPC may provide benefit in patients’ individualized domains of choice among those who set goals, without affecting overall QOL. Future work could refine the subset of patients who benefit and explore the optimal timing and intensity of the intervention.