Stopping rules for phase II studies

On the Methodological Aspects of the Clinical Trials for COVID-19 Conducted in the First Year of the Pandemic: A Descriptive Analysis

In 2020, the whole planet was plagued by the extremely deadly COVID-19 pandemic. More than 83 million people had been infected with COVID-19 while more than 1.9 million people around the planet had died from this virus in the first year of the pandemic. From the first moment, the medical community started working to deal with this pandemic. For this reason, many clinical trials have been and continue to be conducted to find a safe and efficient cure for the virus. In this paper, we review the 96 clinical trials, registered in the ClinicalTrials.gov database, that had been completed by the end of the first year of the pandemic. Although the clinical trials contained significant heterogeneity in the main methodological features (enrollment, duration, allocation, intervention model, and masking) they seemed to be conducted based on an appropriate methodological basis.

Nonmyopic and pseudo-nonmyopic approaches to optimal sequential design in the presence of covariates

In sequential experiments, subjects become available for the study over a period of time, and covariates are often measured at the time of arrival. We consider the setting where the sample size is fixed but covariate values are unknown until subjects enrol. Given a model for the outcome, a sequential optimal design approach can be used to allocate treatments to minimize the variance of the estimator of the treatment effect. We extend existing optimal design methodology so it can be used within a nonmyopic framework, where treatment allocation for the current subject depends not only on the treatments and covariates of the subjects already enrolled in the study, but also the impact of possible future treatment assignments within a specified horizon. The nonmyopic approach requires recursive formulae and suffers from the curse of dimensionality. We propose a pseudo-nonmyopic approach which has a similar aim to the nonmyopic approach, but does not involve recursion and instead relies on simulating trajectories of future possible decisions. Our simulation studies show that, for the simple case of a logistic regression with a single binary covariate and a binary treatment, and a more realistic case with four binary covariates, binary treatment and treatment-covariate interactions, the nonmyopic and pseudo-nonmyopic approaches provide no competitive advantage over the myopic approach, both in terms of the size of the estimated treatment effect and also the efficiency of the designs. Results are robust to the size of the horizon used in the nonmyopic approach, and the number of simulated trajectories used in the pseudo-nonmyopic approach.

Year-to-year variation in attack rates could result in underpowered RSV vaccine efficacy trials

OBJECTIVE

Year-to-year variation of respiratory viruses may result in lower attack rates than expected. We aimed to illustrate the impact of year-to-year variation in attack rates on the likelihood of demonstrating vaccine efficacy (VE) .

STUDY DESIGN AND SETTING

We considered an individually randomized maternal vaccine trial against respiratory syncytial virus-associated hospitalisations (RSVH). For 10 RSVH per 1,000 infants, sample size to have 80% power for true VE of 50% and 70% was 9,846 and 4,424 participants. We reported power to show VE for varying attack rates, selected to reflect realistic year-to-year variation using observational studies. Eight scenarios including varying number of countries and seasons were developed to assess the influence of these trial parameters.

RESULTS

Including up to three seasons decreased the width of the interquartile range for power. Including more seasons concentrated statistical power closer to 80%. Least powered trials had higher statistical power with more seasons. In all scenarios, at least half of the trials had <80% power. For three-season trials, increasing the sample size by 10% reduced the percentage of underpowered trials to less than one-quarter of trials.

CONCLUSION

Year-to-year variation in RSV attack rates should be accounted for during trial design. Mitigation strategies include recruiting over more seasons, or adaptive trial designs.

An Overview of Phase 2 Clinical Trial Designs

Clinical trials are studies to test new treatments in humans. Typically, these treatments are evaluated over several phases to assess their safety and efficacy. Phase 1 trials are designed to evaluate the safety and tolerability of a new treatment, typically with a small number of patients (eg, 20-80), generally spread across several dose levels. Phase 2 trials are designed to determine whether the new treatment has sufficiently promising efficacy to warrant further investigation in a large-scale randomized phase 3 trial, as well as to further assess safety. These studies usually involve a few hundred patients. This article provides an overview of some of the most commonly used phase 2 designs for clinical trials and emphasizes their critical elements and considerations. Key references to some of the most commonly used phase 2 designs are given to allow the reader to explore in more detail the critical aspects when planning a phase 2 trial. A comparison of 3 potential designs in the context of the NRG-HN002 trial is presented to complement the discussion about phase 2 trials.

Integrating pharmacists into aged care facilities to improve the quality use of medicine (PiRACF Study): protocol for a cluster randomised controlled trial

BACKGROUND

Medication management in residential aged care facilities is an ongoing concern. Numerous studies have reported high rates of inappropriate prescribing and medication use in aged care facilities, which contribute to residents' adverse health outcomes. There is a need for new models of care that enhance inter-disciplinary collaboration between residential aged care facility staff and healthcare professionals, to improve medication management. Pilot research has demonstrated the feasibility and benefits of integrating a pharmacist into the aged care facility team to improve the quality use of medicines. This protocol describes the design and methods for a cluster randomised controlled trial to evaluate the outcomes and conduct economic evaluation of a service model where on-site pharmacists are integrated into residential aged care facility healthcare teams to improve medication management.

METHODS

Intervention aged care facilities will employ on-site pharmacists to work as part of their healthcare teams 2 to 2.5 days per week for 12 months. On-site pharmacists, in collaboration with facility nurses, prescribers, community pharmacists, residents and families will conduct medication management activities to improve the quality use of medicines. Aged care facilities in the control group will continue usual care. The target sample size is 1188 residents from a minimum of 13 aged care facilities. The primary outcome is the appropriateness of prescribing, measured by the proportion of residents who are prescribed at least one potentially inappropriate medicine according to the 2019 Beers Criteria. Secondary outcomes include hospital and emergency department presentations, fall rates, prevalence and dose of antipsychotics and benzodiazepines, Anticholinergic Cognitive Burden Score, staff influenza vaccination rate, time spent on medication rounds, appropriateness of dose form modification and completeness of resident's allergy and adverse drug reaction documentation. A cost-consequence and cost-effectiveness analysis will be embedded in the trial.

DISCUSSION

The results of this study will provide information on clinical and economic outcomes of a model that integrates on-site pharmacists into Australian residential aged care facilities. The results will provide policymakers with recommendations relevant to further implementation of this model.

TRIAL REGISTRATION

ACTRN12620000430932 . Registered on 1 April 2020 with ANZCTR.

Phase II, multi-center, open-label, single-arm clinical trial evaluating the efficacy and safety of Mycophenolate Mofetil in patients with high-grade locally advanced or metastatic osteosarcoma (ESMMO): rationale and design of the ESMMO trial

BACKGROUND

Clinical outcomes of patients with osteosarcoma remain unsatisfactory, with little improvement in a 5-year overall survival over the past three decades. There is a substantial need for further research and development to identify and develop more efficacious agents/regimens in order to improve clinical outcomes of patients for whom the prognosis is unfavorable. Recently, mycophenolate mofetil, a prodrug of mycophenolic acid, has been found to have anticancer activity against osteosarcoma in both in vitro and animal experiments, so that further investigation in humans is warranted.

METHODS

A total of 27 patients with high-grade locally advanced or metastatic osteosarcoma will be enrolled into this phase II, multi-center, open-label, single-arm, two-stage clinical trial. The main objectives of this study are to determine the efficacy and safety of mycophenolate mofetil in the patients. The primary endpoint is progression-free survival at 16 weeks; the secondary endpoints include progression-free survival, overall survival, overall response rate, safety parameters, pharmacokinetic parameters, biomarkers, pain score, and quality of life. Mycophenolate mofetil at the initial dose of 5 g/day or lower will be administered for 4 cycles (28 days/cycle) or until disease progression or unacceptable toxicity. The dose of mycophenolate mofetil may be reduced by 1-2 g/day or withheld for some Grade 3 or Grade 4 toxicities whenever clinically needed. The duration of study participation is approximately 4-5 months, with a minimum of 12 study visits. If mycophenolate mofetil proves beneficial to some patients, as evidenced by stable disease or partial response at 16 weeks, administration of mycophenolate mofetil will continue in the extension period.

DISCUSSION

This trial is the first step in the translation of therapeutic potential of mycophenolate mofetil emerging from in vitro and animal studies into the clinical domain. It is designed to assess the efficacy and safety of mycophenolate mofetil in patients with high-grade locally advanced or metastatic osteosarcoma. The results will provide important information about whether or not mycophenolate mofetil is worth further development.

TRIAL REGISTRATION

This trial was prospectively registered on Thai Clinical Trials Registry (registration number: TCTR20190701001). The posted information will be updated as needed to reflect protocol amendments and study progress.

PROCalcitonin-based algorithm for antibiotic use in Acute Pancreatitis (PROCAP): study protocol for a randomised controlled trial

BACKGROUND

Differentiating infection from inflammation in acute pancreatitis is difficult, leading to overuse of antibiotics. Procalcitonin (PCT) measurement is a means of distinguishing infection from inflammation as levels rise rapidly in response to a pro-inflammatory stimulus of bacterial origin and normally fall after successful treatment. Algorithms based on PCT measurement can differentiate bacterial sepsis from a systemic inflammatory response. The PROCalcitonin-based algorithm for antibiotic use in Acute Pancreatitis (PROCAP) trial tests the hypothesis that a PCT-based algorithm to guide initiation, continuation and discontinuation of antibiotics will lead to reduced antibiotic use in patients with acute pancreatitis and without an adverse effect on outcome.

METHODS

This is a single-centre, randomised, controlled, single-blind, two-arm pragmatic clinical and cost-effectiveness trial. Patients with a clinical diagnosis of acute pancreatitis will be allocated on a 1:1 basis to intervention or standard care. Intervention will involve the use of a PCT-based algorithm to guide antibiotic use. The primary outcome measure will be the binary outcome of antibiotic use during index admission. Secondary outcome measures include: safety non-inferiority endpoint all-cause mortality; days of antibiotic use; clinical infections; new isolates of multiresistant bacteria; duration of inpatient stay; episode-related mortality and cause; quality of life (EuroQol EQ-5D); and cost analysis. A 20% absolute change in antibiotic use would be a clinically important difference. A study with 80% power and 5% significance (two-sided) would require 97 patients in each arm (194 patients in total): the study will aim to recruit 200 patients. Analysis will follow intention-to-treat principles.

DISCUSSION

When complete, PROCAP will be the largest randomised trial of the use of a PCT algorithm to guide initiation, continuation and cessation of antibiotics in acute pancreatitis. PROCAP is the only randomised trial to date to compare standard care of acute pancreatitis as defined by the International Association of Pancreatology/American Pancreatic Association guidelines to patients having standard care but with all antibiotic prescribing decisions based on PCT measurement.

TRIAL REGISTRATION

International Standard Randomised Controlled Trial Number, ISRCTN50584992. Registered on 7 February 2018.

An introduction to multiplicity issues in clinical trials: the what, why, when and how

In clinical trials it is not uncommon to face a multiple testing problem which can have an impact on both type I and type II error rates, leading to inappropriate interpretation of trial results. Multiplicity issues may need to be considered at the design, analysis and interpretation stages of a trial. The proportion of trial reports not adequately correcting for multiple testing remains substantial. The purpose of this article is to provide an introduction to multiple testing issues in clinical trials, and to reduce confusion around the need for multiplicity adjustments. We use a tutorial, question-and-answer approach to address the key issues of why, when and how to consider multiplicity adjustments in trials. We summarize the relevant circumstances under which multiplicity adjustments ought to be considered, as well as options for carrying out multiplicity adjustments in terms of trial design factors including Population, Intervention/Comparison, Outcome, Time frame and Analysis (PICOTA). Results are presented in an easy-to-use table and flow diagrams. Confusion about multiplicity issues can be reduced or avoided by considering the potential impact of multiplicity on type I and II errors and, if necessary pre-specifying statistical approaches to either avoid or adjust for multiplicity in the trial protocol or analysis plan.

Flow diversion in the treatment of aneurysms: a randomized care trial and registry

OBJECTIVE

The Flow Diversion in the Treatment of Intracranial Aneurysm Trial (FIAT) was designed to guide the clinical use of flow diversion, an innovative method to treat intracranial aneurysms, within a care trial and to study safety and efficacy.

METHODS

FIAT, conducted in 3 Canadian hospitals, proposed randomized allocation to flow diversion or standard management options (observation, coil embolization, parent vessel occlusion, or clip placement), and a registry of non-randomized patients treated with flow diversion. The primary safety outcome was death or dependency (modified Rankin Scale score > 2) at 3 months, to be determined for all patients who received flow diversion at any time. The primary efficacy outcome was angiographic occlusion at 3–12 months combined with an independent clinical outcome.

RESULTS

Of 112 participating patients recruited between May 2, 2011, and February 25, 2015, 78 were randomized (39 in each arm), and 34 received flow diversion within the registry. The study was halted due to safety concerns. Twelve (16%) of 75 patients (95% CI 8.9%–26.7%) who were allocated to or received flow diversion at any time were dead (n = 8) or dependent (n = 4) at 3 months or more, crossing a predefined safety boundary. Death or dependency occurred in 5 (13.2%) of 38 patients randomly allocated and treated by flow diversion (95% CI 5.0%–28.9%) and in 5 (12.8%) of 39 patients allocated to standard treatment (95% CI 4.8%–28.2%). Efficacy was below expectations of the trial hypothesis: 16 (42.1%) of 38 patients (95% CI 26.7%–59.1%) randomly allocated to flow diversion failed to reach the primary outcome, as compared with 14 (35.9%) of 39 patients allocated to standard treatment (95% CI 21.7%–52.9%).

CONCLUSIONS

Flow diversion was not as safe and effective as hypothesized. More randomized trials are needed to determine the role of flow diversion in the management of aneurysms.

Clinical trial registration no.: NCT01349582 (clinicaltrials.gov)

Statistical Considerations in Clinical Trial Design of Immunotherapeutic Cancer Agents

The classical model for identification and clinical development of anticancer agents was based on small molecules, which were often quite toxic. Early studies in small groups of patients would seek to identify a maximum tolerated dose and major dose-limiting toxicities. Tumor response (shrinkage) would be assessed after a minimum number of doses in phase II testing. The decision to take the drug into the randomized phase III clinical setting was usually based on the proportion and duration of objective tumor responses, along with overall survival compared with historical controls. Immune-oncologics that are designed to fight cancer by direct CD8(+) T-cell priming and activation or by blocking a negative regulatory molecule have a number of sharp distinctions from cytotoxic drugs. These include cytoreductive effects that may be very different in timing of onset from traditional chemotherapy and the potential for inducing long-term durable remissions even in heavily pretreated patients with metastatic disease. In this paper we review the different classes of immune-oncologic drugs in clinical development with particular attention to the biostatistical challenges associated with evaluating efficacy in clinical trials. Confronting these issues upfront is particularly important given the rapidly expanding number of clinical trials with both monotherapy and combination trials in immunooncology.

Do single-arm trials have a role in drug development plans incorporating randomised trials?

Often, single-arm trials are used in phase II to gather the first evidence of an oncological drug's efficacy, with drug activity determined through tumour response using the RECIST criterion. Provided the null hypothesis of 'insufficient drug activity' is rejected, the next step could be a randomised two-arm trial. However, single-arm trials may provide a biased treatment effect because of patient selection, and thus, this development plan may not be an efficient use of resources. Therefore, we compare the performance of development plans consisting of single-arm trials followed by randomised two-arm trials with stand-alone single-stage or group sequential randomised two-arm trials. Through this, we are able to investigate the utility of single-arm trials and determine the most efficient drug development plans, setting our work in the context of a published single-arm non-small-cell lung cancer trial. Reference priors, reflecting the opinions of 'sceptical' and 'enthusiastic' investigators, are used to quantify and guide the suitability of single-arm trials in this setting. We observe that the explored development plans incorporating single-arm trials are often non-optimal. Moreover, even the most pessimistic reference priors have a considerable probability in favour of alternative plans. Analysis suggests expected sample size savings of up to 25% could have been made, and the issues associated with single-arm trials avoided, for the non-small-cell lung cancer treatment through direct progression to a group sequential randomised two-arm trial. Careful consideration should thus be given to the use of single-arm trials in oncological drug development when a randomised trial will follow.

Changing the Paradigm for the Treatment and Development of New Therapies for FSGS

Focal segmental glomerulosclerosis (FSGS) is a renal pathology finding that represents a constellation of rare kidney diseases, which manifest as proteinuria, edema nephrotic syndrome, hypertension, and increased risk for kidney failure. Therapeutic options for FSGS are reviewed displaying the expected efficacy from 25 to 69% depending on specific therapy, patient characteristics, cost, and common side effects. This variability in treatment response is likely caused, in part, by the heterogeneity in the etiology and active molecular mechanisms of FSGS. Clinical trials in FSGS have been scant in number and slow to recruit, which may stem, in part, from reliance on classic clinical trial design paradigms. Traditional clinical trial designs based on the "learn and confirm" paradigm may not be appropriate for rare diseases, such as FSGS. Future drug development and testing will require novel approaches to trial designs that have the capacity to enrich study populations and adapt the trial in a planned way to gain efficiencies in trial completion timelines. A clinical trial simulation is provided that compares a classical and more modern design to determine the maximum tolerated dose in FSGS.

Flexible designs for phase II comparative clinical trials involving two response variables

The aim of phase II clinical trials is to determine whether an experimental treatment is sufficiently promising and safe to justify further testing. The need for reduced sample size arises naturally in phase II clinical trials owing to both technical and ethical reasons, motivating a significant part of research in the field during recent years, while another significant part of the research effort is aimed at more complex therapeutic schemes that demand the consideration of multiple endpoints to make decisions. In this paper, our attention is restricted to phase II clinical trials in which two treatments are compared with respect to two dependent dichotomous responses proposing some flexible designs. These designs permit the researcher to terminate the clinical trial when high rates of favorable or unfavorable outcomes are observed early enough requiring in this way a small number of patients. From the mathematical point of view, the proposed designs are defined on bivariate sequences of multi-state trials, and the corresponding stopping rules are based on various distributions related to the waiting time until a certain number of events appear in these sequences. The exact distributions of interest, under a unified framework, are studied using the Markov chain embedding technique, which appears to be very useful in clinical trials for the sample size determination. Tables of expected sample size and power are presented. The numerical illustration showed a very good performance for these new designs.

Inhaled PGE1 in neonates with hypoxemic respiratory failure: two pilot feasibility randomized clinical trials

Background

Inhaled nitric oxide (INO), a selective pulmonary vasodilator, has revolutionized the treatment of neonatal hypoxemic respiratory failure (NHRF). However, there is lack of sustained improvement in 30 to 46% of infants. Aerosolized prostaglandins I₂ (PGI₂) and E₁ (PGE₁) have been reported to be effective selective pulmonary vasodilators. The objective of this study was to evaluate the feasibility of a randomized controlled trial (RCT) of inhaled PGE₁ (IPGE₁) in NHRF.

Methods

Two pilot multicenter phase II RCTs are included in this report. In the first pilot, late preterm and term neonates with NHRF, who had an oxygenation index (OI) of ≥15 and <25 on two arterial blood gases and had not previously received INO, were randomly assigned to receive two doses of IPGE₁ (300 and 150 ng/kg/min) or placebo. The primary outcome was the enrollment of 50 infants in six to nine months at 10 sites. The first pilot was halted after four months for failure to enroll a single infant. The most common cause for non-enrollment was prior initiation of INO. In a re-designed second pilot, co-administration of IPGE₁ and INO was permitted. Infants with suboptimal response to INO received either aerosolized saline or IPGE₁ at a low (150 ng/kg/min) or high dose (300 ng/kg/min) for a maximum duration of 72 hours. The primary outcome was the recruitment of an adequate number of patients (n = 50) in a nine-month-period, with fewer than 20% protocol violations.

Results

No infants were enrolled in the first pilot. Seven patients were enrolled in the second pilot; three in the control, two in the low-dose IPGE₁, and two in the high-dose IPGE₁ groups. The study was halted for recruitment futility after approximately six months as enrollment targets were not met. No serious adverse events, one minor protocol deviation and one pharmacy protocol violation were reported.

Conclusions

These two pilot RCTs failed to recruit adequate eligible newborns with NHRF. Complex management RCTs of novel therapies for persistent pulmonary hypertension of the newborn (PPHN) may require novel study designs and a longer period of time from study approval to commencement of enrollment.

Trial registration: ClinicalTrials.gov

Pilot one: NCT number: 00598429 registered on 10 January 2008. Last updated: 3 February 2011.

Pilot two: NCT number: 01467076 17 October 2011. Last updated: 13 February 2013.

Electronic supplementary material

The online version of this article (doi:10.1186/1745-6215-15-486) contains supplementary material, which is available to authorized users.

Bayesian stopping guidelines for heart valve premarket approval studies

OBJECTIVES

The Data Monitoring Committee (DMC) for the premarket approval (PMA) study of a new heart valve prosthesis convenes periodically to review the accumulating results of the study, and determines, among other things, whether there is enough concern with safety to stop the study. Their deliberations are largely subjective, based on their combined experience and expertise, but an objective aid to evaluating complication rates, usually called a stopping rule, is desirable.

METHODS

The US Food and Drug Administration has designated objective performance criteria (OPC) for 7 heart valve complications. At the end of the PMA study, when approximately 800 patient-years have been accumulated, the complication rates must compare favorably with the OPC. Given the results to date at an interim review of the data, we use a Bayesian approach to compute the probability of passing the OPC test by the end of study.

RESULTS

We provide a method that the DMC can use to predict the probability of passing the OPC test for each complication, and a graphical aid for each number of events, observed at 100 patient-year intervals.

CONCLUSIONS

Although the DMC ultimately uses combined experience and expertise to make the decision to stop a PMA valve study, we have provided an objective assessment of the probability of the valve ultimately passing the OPC test to aid in making that decision.

An evaluation of a Simon 2-Stage phase II clinical trial design incorporating toxicity monitoring

Phase II clinical trials are usually designed to measure efficacy but patient safety is also a very important aspect. Previous authors suggested a methodology that allows one to monitor the cumulative number of toxic events after each patient is treated, which is also known as continuous toxicity monitoring. In this work we describe how to combine the continuous toxicity monitoring methodology with the Simon 2-Stage design for response. Then we investigate through simulation the combined procedure's type I and type II error rates under various combinations of design parameters. We include the underlying relationship between toxicity and response in our examination of the error rates.

A phase II study of combination chemotherapy in early relapsed epithelial ovarian cancer using gemcitabine and pegylated liposomal doxorubicin

OBJECTIVE

Treatment of epithelial ovarian cancer patients relapsing with a short treatment-free interval (TFI) after prior chemotherapy is unsatisfactory. This phase II trial evaluated the activity and feasibility of pegylated liposomal doxorubicin (PLD) plus gemcitabine in this setting.

METHODS

Patients who had received prior platinum and paclitaxel with a TFI 0-12 months received PLD 25 mg/m(2)day 1 plus gemcitabine 800 mg/m(2)day 1, 8 every 21 days. Gemcitabine was dose escalated to 1000 mg/m(2)day 1, 8 from course 2 in the absence of grade 3/4 toxicity. The primary endpoint was progression-free survival (PFS). Patients were stratified according to response to primary chemotherapy.

RESULTS

Seventy-nine patients (n=26 with CR on prior chemotherapy and TFI 6-12 months; n=20 with CR and TFI 0-6 months; n=33 with PR/SD and TFI 0-12 months) were enrolled. The median age was 59 years (range 31-77 years), and 33 patients had received ≥ 2 prior treatments. A median of five courses was delivered per patient (total 389 courses). Gemcitabine was dose escalated in 124 courses and reduced in 105 courses. No PLD dose reductions occurred. Grade 3/4 toxicities were febrile neutropenia (n=4), PPE (n=2), and mucositis (n=2). One toxic death occurred (pneumonitis/alveolitis). Responses were complete in 5.1%, partial in 27.9%, and stable disease in 55.7%. Median OS and PFS were 12.5 and 6.4 months, respectively.

CONCLUSIONS

The PLD-gemcitabine combination is an effective and well-tolerated salvage treatment for relapsed epithelial ovarian cancer and is a valid candidate for evaluation in a phase III trial.

Combining proof-of-concept with dose-finding: utilization of adaptive designs in migraine clinical trials

There is an obvious need to improve clinical trial designs with respect to efficiency, duration and the number of patients recruited. Adaptive (flexible) designs may be valuable in this respect. We simulated the properties of a two-stage adaptive proof-of-concept and dose-finding trial design in adult migraine patients with moderate to severe headache, with or without aura. We also assessed the usefulness of a combined Bayesian and frequentist approach in the estimation of the probability of success of subsequent Phase III studies. Applying such an innovative approach would result in a reduction of the required sample size by 30 patients and no prolongation of the trial duration. The probability of success in Phase III is > 81%. An innovative adaptive design can facilitate testing of investigational migraine medications by reducing patient numbers and improving predictivity of success in Phase III.

Bayesian design and conduct of phase II single-arm clinical trials with binary outcomes: a tutorial

The aim of phase II single-arm clinical trials of a new drug is to determine whether it has sufficient promising activity to warrant its further development. For the last several years Bayesian statistical methods have been proposed and used. Bayesian approaches are ideal for earlier phase trials as they take into account information that accrues during a trial. Predictive probabilities are then updated and so become more accurate as the trial progresses. Suitable priors can act as pseudo samples, which make small sample clinical trials more informative. Thus patients have better chances to receive better treatments. The goal of this paper is to provide a tutorial for statisticians who use Bayesian methods for the first time or investigators who have some statistical background. In addition, real data from three clinical trials are presented as examples to illustrate how to conduct a Bayesian approach for phase II single-arm clinical trials with binary outcomes.

Phase II multi-step planning methods in oncology: Comparison, recommendations and potential applications

Phase II clinical trials in oncology are commonly used to determine whether a new treatment has a sufficient response rate to be compared with the best standard therapy in phase III. The multi-step planning methods such as Fleming's procedure and the triangular test adapted for phase II trials were elaborated to terminate phase II trials early. We compared these methods considering a fixed number of steps (4 or 5) on their statistical properties: overall type II error and observed type II error, average number of subjects necessary to conclude, probability to conclude at each step and maximum number of subjects necessary to conclude. Fleming's procedure has an actual power similar to the theoretical power for low to high response rate. In contrast, triangular test has an actual power different from the theoretical power in a few situations: it was very high in case of a very low minimum response rate and very low (<20%) in case of a very high minimum response rate (>80%). In these situations, Fleming's procedure and triangular test were not compared and triangular test cannot be recommended. For intermediate response rate, triangular test required a lower average number of subjects to conclude, a larger number of subjects at each step and thus a larger maximum number of subjects. It provided a larger probability to conclude during the first steps. These advantages should be balanced with the risk to have to include a larger number of patients. Multi-step methods, when correctly used are useful for cytotoxic development when response rate is the end point. They can also be used in trials where toxicity is the end point, and could be of great interest for cytostatic development for example with biological surrogate endpoints. An example using real phase II data is also presented.

Supplementary analysis of probabilities at the termination of a group sequential phase II trial

We consider estimation of various probabilities after termination of a group sequential phase II trial. A motivating example is that the stopping rule of a phase II oncologic trial is determined solely based on response to a drug treatment, and at the end of the trial estimating the rate of toxicity and response is desirable. The conventional maximum likelihood estimator (sample proportion) of a probability is shown to be biased, and two alternative estimators are proposed to correct for bias, a bias-reduced estimator obtained by using Whitehead's bias-adjusted approach, and an unbiased estimator from the Rao-Blackwell method of conditioning. All three estimation procedures are shown to have certain invariance property in bias. Moreover, estimators of a probability and their bias and precision can be evaluated through the observed response rate and the stage at which the trial stops, thus avoiding extensive computation.

Assessing abuse liability in clinical trials

In this article, the use of data collected in registration-focused clinical trials to provide information concerning abuse liability of compounds under development is discussed. Registration-focused trials are limited by the small and select sample chosen for study participation and design constraints. However, most compounds under development are first administered to humans during the conduct of registration-focused trials, so this presents an opportunity to collect potentially important information about the effects of drugs in humans. At present, information concerning subjective effects and symptoms associated with drug discontinuation are not collected systematically. Reports are generally considered as adverse events (AEs) and are recorded on the case report forms (CRFs) in the investigators own language. There is generally no rating of drug "liking". The authors suggest strategies that could be implemented in registration-focused clinical trials to improve the information gathered with regard to subjective effects, abuse liability and discontinuation-emergent symptoms. Importantly, there remains much groundwork to be done in developing and validating appropriate assessment instruments and determining "threshold" levels for concern. Under the best of circumstances, registration-focused clinical trials have limited potential to detect abuse liability because of the small number of patients seen and the exclusion of many subjects who might be particularly vulnerable to the abuse of marketed compounds (i.e. individuals with substance use disorders). In cases where there are reasons to suspect that a drug under development has abuse potential, systematic exploration with a series of studies specifically designed to assess abuse potential must be conducted.

Learning from previous responses in phase I dose-escalation studies

Dose escalation in phase I studies is generally performed on the basis of clinical experience and judgement. In this paper some of the statistical approaches that have been proposed for the formalization of the procedure are described. Apart from the use of the Continual Reassessment Method in oncology studies, such formal methods have received little implementation. The purpose of presenting them here is to promote their further exploration and appropriate implementation. Certain limitations are discussed, which will be best overcome by collaboration between clinical pharmacologists and statisticians.