Response rate as an endpoint in clinical trials.
Explore the source record for details and available documents.
Antineoplastic Combined Chemotherapy Protocols↗
Biomedical subjects
Stephen L George
Publications and source records attributed to Stephen L George.
Processes to activate phase III clinical trials in a Cooperative Oncology Group: the Case of Cancer and Leukemia Group B.
PURPOSE: National Cancer Institute-sponsored cooperative oncology groups are major sponsors of phase III clinical trials, yet the time and steps required to design and activate such studies has not been well studied. We examine the processes and document the calendar time required to activate such studies opened by the Cancer and Leukemia Group B (CALGB). METHODS: Setup steps were documented by (1) interviewing CALGB headquarters and statistical center staff and committee chairs to discover the steps required to transit from concept development to final study activation, (2) reviewing procedure manuals, and (3) inspecting all study records, documents, and e-mails to identify any additional steps. Calendar time was collected for each major process. RESULTS: Thirteen phase III studies were activated by CALGB during the study period of May 2002 to May 2005. More than 370 distinct processes were required for study activation: 317 work steps, 42 decision points, and 29 processing loops. Sixty-three percent of the decision points were outside CALGB. The complete process map measures 243.5" x 41" in 8-point font. Median calendar days to activate a phase III study at CALGB was 580 days (range, 295 to 1,248 days) from concept approval and 784 days (range, 537 to 1,130 days) from initial conception of the study. CONCLUSION: Setup of a phase III study at a major cooperative oncology group is a complex and lengthy process, with the majority of decision points external to the cooperative group. To improve the activation process, research should to be directed toward both internal and external groups and processes.
Biomedical Research↗
P-value calculation for multistage phase II cancer clinical trials.
Due to ethical and practical issues, clinical trials are conducted in multiple stages, but the reported p-values often fail to reflect the design aspect of the trials. We investigate some approaches to p-value calculation in analyzing multi-stage Phase II clinical trials that have a binary variable, such as response, as the primary endpoint. The sample space consists of the paired outcomes of the stopping stage and the number of responses, which jointly define a complete and sufficient statistic for the true binomial proportion. Calculating a p-value requires an ordering of the paired outcomes so that outcomes more extreme than the observed can be identified. We consider the orderings based on the maximum likelihood estimator and the uniformly minimum variance unbiased estimator. We will compare, using some examples, the p-values based on these alternative orderings and the one ignoring the multistage design aspect of phase II trials.
Algorithms↗
A multiple testing procedure to associate gene expression levels with survival.
In many microarray studies the primary objective is to identify, from a large panel of genes, those which are prognostic markers of a censored survival endpoint such as time to disease recurrence or death. Often, these genes are considered prognostic in that their respective expressions are associated with the survival endpoint of interest. To assess this association requires specifying an appropriate measure of association, a suitable test statistic and, as the number of genes is large, proper handling of multiplicity issues. In this paper, we will address these issues by utilizing a general correlation measure, a non-parametric test statistic, and control of the family-wise error rate by employing permutation resampling. Comprehensive simulation studies are conducted to investigate the statistical properties of the proposed procedure. The proposed method is applied to a recently published data set on patients with lung cancer.
Adenocarcinoma↗
Sample size calculation for simulation-based multiple-testing procedures.
In this article, we present a simple method to calculate sample size and power for a simulation-based multiple testing procedure which gives a sharper critical value than the standard Bonferroni method. The method is especially useful when several highly correlated test statistics are involved in a multiple-testing procedure. The formula for sample size calculation will be useful in designing clinical trials with multiple endpoints or correlated outcomes. We illustrate our method with a quality-of-life study for patients with early stage prostate cancer. Our method can also be used for comparing multiple independent groups.
Computer Simulation↗
A general formulation for a one-sided group sequential design.
BACKGROUND: A major contribution to the statistical literature on group sequential designs was provided by Pampallona and Tsiatis who developed closed form functions that can be used to iteratively calculate the boundary points of a family of popular group sequential designs. A related area of interest is the use of conditional probability calculations to make interim decisions in stochastic curtailment procedures. PURPOSE: The purpose of the paper is to develop group sequential designs based on conditional probabilities, to compare our results to the general closed form family of designs developed by Pampallona and Tsiatis, and to relate these to commonly used stochastic curtailment procedures. METHODS: The problem and its solution are formulated and derived mathematically. A graphical interpretation of the results provides the reader with an alternative mechanism to understand the results and their significance. RESULTS: One-sided group sequential design boundary points, as closed form functions, are derived from conditional probability statements. These conditional probability statements can be interpreted as the probability, at the final analysis, of reversing the conclusion reached at an interim state. Under mild constraints, these boundary points are identical to the Pampallona and Tsiatis boundary points. At any interim stage when a boundary point is attained or surpassed we suggest a graphical approach to examine the conditional probability of reversing the interim decision at the final stage versus a range of possible parameter values. For stochastic curtailment procedures, we recommend relaxing (increasing) the conditional probability levels to at least 0.50 so that early stopping is at least as likely as for the O'Brien-Fleming procedure. LIMITATIONS: The results are limited to one-sided group sequential designs. CONCLUSIONS: Conditional probabilities of reversing interim decisions provides a useful concept to develop group sequential designs and to evaluate stochastic curtailment procedures.
Breast Neoplasms↗
Sequential multiagent chemotherapy is not superior to high-dose cytarabine alone as postremission intensification therapy for acute myeloid leukemia in adults under 60 years of age: Cancer and Leukemia Group B Study 9222.
The Cancer and Leukemia Group B (CALGB) study 9222 tested the hypothesis that treatment intensification of acute myeloid leukemia (AML) in first remission with multiple chemotherapy agents is superior to high-dose cytarabine (HiDAC) alone. We enrolled 474 patients younger than 60 years old with untreated de novo AML. Daunorubicin and cytarabine resulted in complete remission (CR) in 342 patients (72%), and 309 of these patients were randomized to receive one of 2 different intensification regimens. The first regimen consisted of 3 courses of HiDAC. The second regimen consisted of one course of HiDAC, a second course with etoposide and cyclophosphamide, and a third course with diaziquone and mitoxantrone. After a median follow-up time of 8.3 years, the median survival for all randomized patients was 2.8 years (95% CI, 1.9-6.8 years). There was no difference in disease-free survival (DFS) between the 2 regimens (P = .66). The median DFS was 1.1 years (95% CI, 0.9-1.7 years) for patients receiving HiDAC and 1.0 year (95% CI, 0.9-1.3 years) for those receiving multiagent chemotherapy. Cytogenetics was the only pretreatment characteristic prognostic for DFS, but there was no evidence of a differential treatment effect within cytogenetic risk groups. Toxicity was greater with multiagent chemotherapy. These 2 postremission regimens produced similar outcomes.
Acute Disease↗
Dose escalation studies of cytarabine, daunorubicin, and etoposide with and without multidrug resistance modulation with PSC-833 in untreated adults with acute myeloid leukemia younger than 60 years: final induction results of Cancer and Leukemia Group B Study 9621.
PURPOSE: P-glycoprotein (Pgp) is strongly inhibited by PSC-833. A chemotherapy dose-escalation study was performed with PSC-833 in patients younger than 60 years with untreated acute myeloid leukemia. Clinical rather than pharmacokinetic end points were used to develop two induction therapies containing drugs susceptible to Pgp-mediated efflux and associated with comparable toxicities at the maximum-tolerated doses. PATIENTS AND METHODS: A total of 410 patients were enrolled. Fifteen induction regimens containing variable doses of daunorubicin (DNR) and etoposide (ETOP) and fixed doses of cytarabine were evaluated with (ADEP) or without (ADE) a fixed dose of PSC-833. RESULTS: Doses selected for phase III testing were DNR 90 mg/m(2) and ETOP 100 mg/m(2) in ADE, and DNR and ETOP each 40 mg/m(2) in ADEP. Intolerable mucosal toxicity occurred at higher doses of ADEP. Although the design of this study precludes direct comparisons, there was an apparent advantage for receiving ADEP with respect to disease-free and overall survival in patients < or = 45 years old, despite the significantly lower doses of DNR and ETOP given in ADEP compared with ADE. CONCLUSION: A large clinical data set was used to develop induction regimens containing two drugs susceptible to Pgp-mediated efflux, with and without an inhibitor of Pgp function. The chosen doses have comparable antileukemia activity and toxicity, making them suitable for use in a phase III comparative study of induction chemotherapy for patients with acute myeloid leukemia younger than 60 years. That trial will also clarify whether patients < or = 45 years old are especially likely to benefit from Pgp inhibition during induction therapy.
Acute Disease↗
Strength of accumulating evidence and data monitoring committee decision making.
The data monitoring committee (DMC) is a vital component of a randomized clinical trial. Its responsibilities include stopping the trial early for extreme results. The decision to stop the trial must be based on a careful synthesis of statistical methodology and clinical judgment. It is critical to ensure the validity of this complex process. In this paper we present results of a survey of 21 DMC members conducted to investigate how they evaluate accumulating evidence. The results indicate that some DMC members may be over-interpreting developing trends in the data.
Antineoplastic Agents↗
Managing accrual in cooperative group clinical trials.
Explore the source record for details and available documents.
Clinical Trials as Topic↗
Should statisticians reporting to data monitoring committees be independent of the trial sponsor and leadership?
It has long been a fundamental principle of clinical trials that interim comparative data should be kept confidential, with such data accessible only to a small number of individuals responsible for its analysis and monitoring. The rationale for keeping investigators and sponsors blinded to interim data has been extensively discussed, but the possible conflicts of interest that could arise for the statistician who performs the analysis of the interim data and presents it to a data monitoring committee has received little attention. We describe these potential conflicts, and the advantages and disadvantages of approaches that might be taken to minimize them. We have invited commentary on this issue from several statisticians with substantial experience in clinical trials and interim data monitoring.
Clinical Trials Data Monitoring Committees↗
Satisfaction of the uncertainty principle in cancer clinical trials: retrospective cohort analysis.
OBJECTIVE: To assess whether publicly funded adult cancer trials satisfy the uncertainty principle, which states that physicians should enroll a patient in a trial only if they are substantially uncertain which of the treatments in the trial is most appropriate for the patient. This principle is violated if trials systematically favour either the experimental or the standard treatment. DESIGN: Retrospective cohort study of completed cancer trials, with randomisation as the unit of analysis. SETTING: Two cooperative research groups in the United States. STUDIES INCLUDED: 93 phase III randomised trials (103 randomisations) that completed recruitment of patients between 1981 and 1995. MAIN OUTCOME MEASURES: Whether the randomisation favoured the experimental treatment, the standard treatment, or neither treatment; effect size (outcome of the experimental treatment compared with outcome of the standard treatment) for each randomisation. RESULTS: Three randomisations (3%) favoured the standard treatment, 70 (68%) found no significant difference between treatments, and 30 (29%) favoured the experimental treatment. The average effect size was 1.20 (95% confidence interval 1.13 to 1.28), reflecting a slight advantage for the experimental treatment. CONCLUSIONS: In cooperative group trials in adults with cancer, there is a measurable average improvement in disease control associated with assignment to the experimental rather than the standard arm. However, the heterogeneity of outcomes and the small magnitude of the advantage suggest that, as a group, these trials satisfy the uncertainty principle.
Clinical Trials, Phase III as Topic↗
Admissible two-stage designs for phase II cancer clinical trials.
In a typical two-stage design for a phase II cancer clinical trial for efficacy screening of cytotoxic agents, a fixed number of patients are initially enrolled and treated. The trial may be terminated for lack of efficacy if the observed number of tumour responses after the first stage is too small, thus avoiding treatment of patient with inefficacious regimen. Otherwise, an additional fixed number of patients are enrolled and treated to accumulate additional information on efficacy as well as safety. The minimax and the so-called 'optimal' designs by Simon have been widely used, and other designs have largely been ignored in the past for such two-stage cancer clinical trials. Recently Jung et al. proposed a graphical method to search for compromise designs with features more favourable than either the minimax or the optimal design. In this paper, we develop a family of two-stage designs that are admissible according to a Bayesian decision-theoretic criterion based on an ethically justifiable loss function. We show that the admissible designs include as special cases the Simon's minimax and the optimal designs as well as the compromise designs introduced by Jung et al. We also present a Java program to search for admissible designs that are compromises between the minimax and the optimal designs.
Bayes Theorem↗
Impact of race on survival in men with metastatic hormone-refractory prostate cancer.
OBJECTIVES: To determine whether blacks with hormone-refractory prostate cancer have shorter survival compared with whites with the same disease. METHODS: Data from eight multicenter trials (four Phase II and four randomized Phase III studies) conducted by the Cancer and Leukemia Group B were combined. Eligible patients had progressive prostate cancer after androgen deprivation therapy (with documented castration levels of testosterone), an Eastern Cooperative Oncology Group performance status of 0 to 2, and adequate hematologic, renal, and hepatic function. The proportional hazards model was used to assess the prognostic importance of race, adjusting for important factors. All statistical tests were two-sided. RESULTS: Of the 1183 patients, 15% were blacks, 45% of patients had a Gleason sum of 8 or greater, and the median age was 71 years. Of the 1183 patients, 35% had measurable disease and 89% had an Eastern Cooperative Oncology Group performance status of 0 to 1. Blacks were younger, had a shorter interval between diagnosis and study entry, and had greater prostate-specific antigen levels, lower hemoglobin levels, and a lower likelihood of prior prostatectomy than whites. The median survival was 15 months (95% confidence interval 12 to 18) for blacks compared with 14 months (95% confidence interval 13 to 15) for whites. In a multivariate analysis, adjusting for age, performance status, presence of visceral disease, hemoglobin, Gleason sum, prostate-specific antigen level, alkaline phosphatase, lactate dehydrogenase, and years since diagnosis, the hazard ratio was 0.85 (95% confidence interval 0.71 to 1.02, P = 0.08) for blacks compared with whites. CONCLUSIONS: No statistically significant difference was found in overall survival between blacks and whites with metastatic hormone-refractory prostate cancer.
Adenocarcinoma↗
Data monitoring and large apparent treatment effects.
Explore the source record for details and available documents.
Clinical Trials Data Monitoring Committees↗
The Society for Clinical Trials supports United States legislation mandating trials registration. Position paper.
The official position of the Society for Clinical Trials is to support legislation in the United States and internationally that mandates registration of all controlled clinicals trials at or before enrollment of the first participant. The major trial sponsor would be responsible for ensuring contribution of trial information. The mechanism for registering trials would be through research ethics review boards, and registration would be required for ethics approval and before trial initiation. Standardized data and a unique identification number would be available for each registered trial.
Controlled Clinical Trials as Topic↗
Medulloblastoma and birth date: evaluation of 3 U.S. datasets.
Studies from Norway and Japan have found a higher incidence of medulloblastoma related to births that occur in the fall. The authors sought further evidence concerning this association. For 122 patients in a Duke University database and 90 patients from the Central Cancer Registry of North Carolina, the frequency distribution of birth dates by month was statistically significantly different from the expected North Carolina distribution (p = 0.04 and 0.06). For 75 patients from California Surveillance, Epidemiology, and End Results (SEER) data, the frequency distribution of birth dates by month was marginally different from the expected U.S. distribution (p = 0.14). For 922 patients from national SEER data, the frequency distribution of birth dates by month was not statistically significantly different from the expected U.S. distribution (p = 0.54). Subgroup analysis suggests seasonality of birth dates is most significant for patients aged 5-14 yr diagnosed with medulloblastoma.
Adolescent↗
Selection bias, phase II trials, and the FDA accelerated approval process.
Explore the source record for details and available documents.
Antineoplastic Agents↗