Leveraging historical data into oncology development programs: Two case studies of phase 2 Bayesian augmented control trial designs.

Leveraging historical data into the design and analysis of phase 2 randomized controlled trials can improve efficiency of drug development programs. Such approaches can reduce sample size without loss of power. Potential issues arise when the current control arm is inconsistent with historical data, which may lead to biased estimates of treatment efficacy, loss of power, or inflated type 1 error. Consideration as to how to borrow historical information is important, and in particular, adjustment for prognostic factors should be considered. This paper will illustrate two motivating case studies of oncology Bayesian augmented control (BAC) trials. In the first example, a glioblastoma study, an informative prior was used for the control arm hazard rate. Sample size savings were 15% to 20% by using a BAC design. In the second example, a pancreatic cancer study, a hierarchical model borrowing method was used, which enabled the extent of borrowing to be determined by consistency of observed study data with historical studies. Supporting Bayesian analyses also adjusted for prognostic factors. Incorporating historical data via Bayesian trial design can provide sample size savings, reduce study duration, and enable a more scientific approach to development of novel therapies by avoiding excess recruitment to a control arm. Various sensitivity analyses are necessary to interpret results. Current industry efforts for data transparency have meaningful implications for access to patient-level historical data, which, while not critical, is helpful to adjust for potential imbalances in prognostic factors.

Early change in tumour size predicts overall survival in patients with first-line metastatic breast cancer.

PURPOSE: Clinical trials using change in tumour size (CTS) as a primary end-point benefit from earlier evaluation of treatment effect and increased study power over progression-free survival, ultimately resulting in more timely regulatory approvals for cancer patients. In this work, a modelling framework was established to further characterise the relationship between CTS and overall survival (OS) in first-line metastatic breast cancer (mBC). METHODS: Data from three randomised phase III trials designed to evaluate the clinical benefit of gemcitabine combination therapy in mBC patients were collated. Two drug-dependent models were developed to describe tumour growth dynamics: the first for paclitaxel/gemcitabine treatment and the second for docetaxel/gemcitabine treatment. A parametric survival model was used to characterise survival as a function of CTS and baseline patient demographics. RESULTS: While the paclitaxel/gemcitabine model incorporated tumour shrinkage by both paclitaxel and gemcitabine with resistance to paclitaxel, the docetaxel/gemcitabine model incorporated shrinkage and resistance to docetaxel alone. Predictors for OS were CTS at week 8, baseline tumour size and ECOG performance status. Model predictions reveal that for an asymptomatic mBC patient with a 6-cm tumour burden, first-line paclitaxel/gemcitabine treatment offers a median OS of 28.6 months, compared to 26.0 months for paclitaxel alone. CONCLUSION: A modelling framework was established, quantitatively describing the tumour growth inhibitory effects of various gemcitabine combotherapies and the effect of the resulting CTS on survival in first-line mBC. This work further supports the use of early CTS as a go/no-go decision point during phase II clinical evaluation of treatments for mBC.

Models for change in tumour size, appearance of new lesions and survival probability in patients with advanced epithelial ovarian cancer.

AIMS: The aims of this study were (i) to develop a modelling framework linking change in tumour size during treatment to survival probability in metastatic ovarian cancer; and (ii) to model the appearance of new lesions and investigate their relationship with survival and disease characteristics. METHODS: Data from a randomized Phase III clinical trial comparing carboplatin monotherapy to gemcitabine plus carboplatin combotherapy in 336 patients with metastatic ovarian cancer were used. A population model describing change in tumour size based on drug treatment information was established and its relationship with time to appearance of new lesions and survival were investigated with time to event models. RESULTS: The tumour size profiles were well characterized as evaluated by visual predictive checks. Metastasis in the liver at enrolment and change in tumour size up to week 12 were predictors of time to appearance of new lesions. Survival was predicted based on the patient tumour size and ECOG performance status at enrolment and on appearance of new lesions during treatment and change in tumour size up to week 12. Tumour size and survival data from a separate study were adequately predicted. CONCLUSIONS: The proposed models simulate tumour dynamics following treatment and provide a link to the probability of developing new lesions as well as to survival. The models have potential to be used for optimizing the design of late phase clinical trials in metastatic ovarian cancer based on early phase clinical study results and simulation.

Phase 1/2 study of ocaratuzumab, an Fc-engineered humanized anti-CD20 monoclonal antibody, in low-affinity FcγRIIIa patients with previously treated follicular lymphoma.

This phase 2 study assessed the safety and efficacy of ocaratuzumab, a humanized anti-CD20 monoclonal antibody. Fifty patients with previously treated follicular lymphoma (FL) and a low-affinity genotype of FcγRIIIa received ocaratuzumab 375 mg/m(2) weekly for 4 weeks. Grade 3/4/5 adverse events (AEs) were reported in 11/1/1 patients, respectively. Serious AEs were reported by 11/50 patients, and three discontinued due to AEs. One patient died from aspiration pneumonia due to possibly drug-related nausea and vomiting. Investigator-assessed response rate was 30% (15/50), including four complete responses (CR), three CR unconfirmed (CRu) and eight partial responses (PR). Investigator-assessed median Progression-free survivial (PFS) was 38.3 weeks. Ocaratuzumab's pharmacokinetic profile was similar to that reported for rituximab. Lymphocyte subset analysis showed significant, selective reduction of B-cells during and after ocaratuzumab treatment. Ocaratuzumab at this dose and schedule is active and well tolerated in patients with previously treated FL with low affinity FcγRIIIa genotypes. ClinTrials registry number: NCT00354926.

Randomized, double-blind, phase II trial comparing gemcitabine-cisplatin plus the LTB4 antagonist LY293111 versus gemcitabine-cisplatin plus placebo in first-line non-small-cell lung cancer.

INTRODUCTION: In this phase II study, patients with stage IIIB/IV non-small-cell lung cancer were randomly assigned (1:1:1) to receive LY293111 (200 mg twice daily [200 LY293111] or 600 mg twice daily [600 LY293111]) or placebo for 7 days, followed by concurrent cisplatin (75 mg/m2; day 1) and gemcitabine (1250 mg/m2; days 1 and 8), every 21 days.The primary endpoint was progression-free survival, (PFS), with 75% power to detect 33% improvement compared with placebo (5 months). METHODS: Of 200 randomized patients, 195 were treated. Demographics were well balanced across treatment arms: 65% of the patients were men; median age was 62 years; 85% had stage IV disease; and patients had an Eastern Cooperative Oncology Group performance status of 0 (36%) or 1 (64%). RESULTS: The most frequent study drug-related toxicities were nausea, vomiting, and fatigue. Response rates were similar across treatment arms (200 LY293111: 20%; 600 LY293111: 25%; placebo: 31%). CONCLUSIONS: Median PFS (95% confidence interval) was not significantly different across treatment arms (200 LY293111: 4.6 months [3.2-5.0]; 600 LY293111: 5.6 months [4.1-6.8]; placebo: 6.0 months [5.2-7.5]). LY293111 combined with gemcitabine-cisplatin did not increase median PFS compared with placebo plus gemcitabine-cisplatin in patients with non-small-cell lung cancer.

Use of historical control data for assessing treatment effects in clinical trials.

Clinical trials rarely, if ever, occur in a vacuum. Generally, large amounts of clinical data are available prior to the start of a study, particularly on the current study's control arm. There is obvious appeal in using (i.e., 'borrowing') this information. With historical data providing information on the control arm, more trial resources can be devoted to the novel treatment while retaining accurate estimates of the current control arm parameters. This can result in more accurate point estimates, increased power, and reduced type I error in clinical trials, provided the historical information is sufficiently similar to the current control data. If this assumption of similarity is not satisfied, however, one can acquire increased mean square error of point estimates due to bias and either reduced power or increased type I error depending on the direction of the bias. In this manuscript, we review several methods for historical borrowing, illustrating how key parameters in each method affect borrowing behavior, and then, we compare these methods on the basis of mean square error, power and type I error. We emphasize two main themes. First, we discuss the idea of 'dynamic' (versus 'static') borrowing. Second, we emphasize the decision process involved in determining whether or not to include historical borrowing in terms of the perceived likelihood that the current control arm is sufficiently similar to the historical data. Our goal is to provide a clear review of the key issues involved in historical borrowing and provide a comparison of several methods useful for practitioners.

Randomization by minimization for unbalanced treatment allocation.

Minimization is a dynamic randomization technique that has been widely used in clinical trials for achieving a balance of prognostic factors across treatment groups, but most often it has been used in the setting of equal treatment allocations. Although unequal treatment allocation is frequently encountered in clinical trials, an appropriate minimization procedure for such trials has not been published. The purpose of this paper is to present novel strategies for applying minimization methodology to such clinical trials. Two minimization techniques are proposed and compared by probability calculation and simulation studies. In the first method, called naïve minimization, probability assignment is based on a simple modification of the original minimization algorithm, which does not account for unequal allocation ratios. In the second method, called biased-coin minimization (BCM), probability assignment is based on allocation ratios and optimized to achieve an 'unbiased' target allocation ratio. The performance of the two methods is investigated in various trial settings including different number of treatments, prognostic factors and sample sizes. The relative merits of the different distance metrics are also explored. On the basis of the results, we conclude that BCM is the preferable method for randomization in clinical trials involving unequal treatment allocations. The choice of different distance metrics slightly affects the performance of the minimization and may be optimized according to the specific feature of trials.

In Vivo Measurements of Tumor Metabolism and Growth after Administration of Enzastaurin Using Small Animal FDG Positron Emission Tomography.

Background. The use of 2-[(18)F]fluoro-2-deoxy-D-glucose ([(18)F]FDG) may help to establish the antitumor activity of enzastaurin, a novel protein kinase C-beta II (PKC-betaII) inhibitor, in mouse xenografts. Methods. The hematologic cell line RAJI and the solid tumor cell line U87MG were each implanted in NOD/SCID mice. Standard tumor growth measurements and [(18)F]FDG PET imaging were performed weekly for up to three weeks after tumor implantation and growth. Results. Concomitant with caliper measurements, [(18)F]FDG PET imaging was performed to monitor glucose metabolism. Heterogeneity of glucose uptake in various areas of the tumors was observed after vehicle or enzastaurin treatment. This heterogeneity may limit the use of [(18)F]FDG PET imaging to measure enzastaurin-associated changes in xenograft tumors. Conclusion. [(18)F]FDG PET imaging technique does not correlate with standard caliper assessments in xenografts to assess the antitumor activity of enzastaurin. Future studies are needed to determine the use of [(18)F]FDG PET imaging in preclinical models.

Phase I pharmacokinetic and pharmacodynamic study of the oral protein kinase C beta-inhibitor enzastaurin in combination with gemcitabine and cisplatin in patients with advanced cancer.

PURPOSE: Enzastaurin targets the protein kinase C and phosphatidylinositol 3-kinase/AKT pathways to reduce tumor angiogenesis and cell proliferation and to induce cell death. A phase I trial was conducted to evaluate the feasibility of combining enzastaurin with gemcitabine and cisplatin. EXPERIMENTAL DESIGN: Patients with advanced cancer received a 14-day lead-in treatment with oral enzastaurin followed by subsequent 21-day cycles of daily enzastaurin, gemcitabine on days 1 and 8, and cisplatin on day 1. Enzastaurin doses were escalated between 350 mg once daily to 500 mg twice daily, whereas gemcitabine doses were either 1,000 or 1,250 mg/m(2) and cisplatin doses were either 60 or 75 mg/m(2). Circulating endothelial cell numbers and CD146 and CD133 mRNA expression were evaluated as pharmacodynamic markers. RESULTS: Thirty-three patients (median age, 58 years) were enrolled in seven dose levels. The maximum tolerated dose was not identified. Two dose-limiting toxicities (grade 2 QT interval corrected for heart rate prolongation and grade 3 fatigue) were reported. Other toxicities included grade 3/4 neutropenia (3 of 6 patients), thrombocytopenia (1 of 6 patients), grade 3 leukopenia (2 patients), and fatigue (5 patients). Enzastaurin twice daily (> or =250 mg) resulted in more discontinuations and low-grade toxicities. In the combination, enzastaurin exposures decreased slightly but remained above the target of 1,400 nmol/L, whereas gemcitabine/cisplatin exposures were unaltered. Three patients (9.1%) had partial responses and 13 (39.4%) had stable disease. Measurement of circulating endothelial cell numbers and CD146 and CD133 mRNA expression did not contribute to decision-making on dose escalation. CONCLUSIONS: Recommended phase II dose is 500 mg enzastaurin once daily, 1,250 mg/m(2) gemcitabine, and 75 mg/m(2) cisplatin. This regimen is well tolerated with no significant alterations in the pharmacokinetic variables of any drug.

Phase II study of enzastaurin, a protein kinase C beta inhibitor, in patients with relapsed or refractory diffuse large B-cell lymphoma.

PURPOSE: Protein kinase C beta (PKCbeta) was identified by gene-expression profiling, preclinical evaluation, and independent immunohistochemical analysis as a rational therapeutic target in diffuse large B-cell lymphoma (DLBCL). We conducted a multicenter phase II study of a potent inhibitor of PKCbeta, enzastaurin, in patients with relapsed or refractory DLBCL. PATIENTS AND METHODS: Enzastaurin was taken orally once daily until disease progression or unacceptable toxicity occurred. Study end points included freedom from progression (FFP) for > or= two cycles (one cycle = 28 days), objective response, and toxicity. RESULTS: Fifty-five patients (median age, 68 years) were enrolled. Patients had received a median number of two prior therapies (range, one to five); six patients relapsed after high-dose therapy and autologous stem-cell transplantation. Only one grade 4 toxicity (hypomagnesemia) occurred. Grade 3 toxicities included fatigue (n = 2), edema (n = 1), headache (n = 1), motor neuropathy (n = 1), and thrombocytopenia (n = 1). No grade 3 or 4 neutropenia occurred. No deaths or discontinuations due to toxicity were reported. Fifteen patients completed less than one cycle of therapy. Twelve of 55 patients (22%; 95% CI, 13% to 46%) experienced FFP for two cycles, and eight patients remained free from progression for four cycles (15%; 95% CI, 6% to 27%). Four patients (7%; 95% CI, 2% to 18%), including three complete responders and one patient with stable disease, continue to experience FFP 20+ to 50+ months after study entry. CONCLUSION: Treatment with enzastaurin was well-tolerated and associated with prolonged FFP in a small subset of patients with relapsed or refractory DLBCL. Further studies of enzastaurin in DLBCL are warranted.

Phase I dose escalation and pharmacokinetic study of enzastaurin, an oral protein kinase C beta inhibitor, in patients with advanced cancer.

PURPOSE: This phase I study was conducted to determine the recommended dose of enzastaurin, an oral protein kinase C beta (PKCbeta) inhibitor, for phase II trials. Secondary objectives were maximum-tolerated dose (MTD), pharmacokinetics (PK), toxicity, and response. PATIENTS AND METHODS: Patients at least 18 years of age with advanced cancer and an Eastern Cooperative Oncology Group performance status of 0 or 1 lower received enzastaurin orally once daily at a starting dose of 20 mg. Dose escalation proceeded using a modified Simon design. RESULTS: All 47 patients enrolled (mean age, 58 years) received at least one dose of enzastaurin, with a median of two cycles (range, one to 17 cycles). Prevalent malignancies were lung (n = 10) and head and neck cancers (n = 9). Although no MTD was identified up to 700 mg/d, 525 mg was chosen as the recommended dose, and 12 additional patients were accrued at that level. Three dose-limiting toxicities (QTc changes) occurred: one at the 700-mg dose (patient discontinued), and two in the expansion cohort at the 525-mg dose. Total analytes (enzastaurin and its metabolites) exposure increased with increasing doses up to 240 mg, and appeared to plateau at 525 and 700 mg. Grade 1 chromaturia, fatigue, and other GI toxicities were the most common, while no clinically significant grade 3/4 toxicities occurred. Two deaths, unrelated to enzastaurin, occurred. Twenty-one patients (45%) achieved stable disease (SD) for two to 16 cycles. CONCLUSION: On the basis of plasma exposures and safety data, enzastaurin 525 mg once daily is the recommended phase II dose. Enzastaurin is well tolerated up to 700 mg/d. Evidence of early activity was seen with significant stable disease.

Phase I trial of the cryptophycin analogue LY355703 administered as an intravenous infusion on a day 1 and 8 schedule every 21 days.

The cryptophycin analogue LY355703 is a potent inhibitor of microtubule polymerization that displays in vitro and in vivo activity in cell lines and tumor xenografts displaying the multidrug-resistant phenotype. In a Phase I trial, 25 patients received LY355703 as a 2-h i.v. infusion on day 1 and day 8 repeated every 3 weeks. Doses were escalated from 0.1 to 2.22 mg/m2 using a modified continual reassessment method. Neurological toxicity was found to be dose-limiting at 1.84 and 2.22 mg/m2. Among four patients treated at these doses, two had grade 4 constipation/ileus, one with severe myalgias, and one had grade 3 motor neuropathy. These findings were reversible. The 1.5 mg/m2 dose level was well tolerated. An amended twice-weekly schedule was pursued in 11 patients in an attempt to improve dose intensity and avoid dose-limiting neurotoxicity. Doses of >0.75 mg/m2 on a day 1, 4, 8, and 11 schedule every 21 days were not tolerated as a result of nausea/constipation, suggesting that LY335703 toxicity is not schedule dependent and is related to cumulative dose. LY355703 plasma concentrations measured by liquid chromatography with tandem mass spectrometry were evaluated using a population pharmacokinetic model. LY355703 was eliminated rapidly with a short terminal half-life that ranged from 0.8 to 3.9 h. Interpatient variability with respect to plasma clearance and volume of distribution, including covariates, was moderate at 32% and 39%, respectively. Maximum plasma concentration and area under the plasma concentration-time curve were linear over the dose range studied. A patient with non-small cell lung cancer previously treated with taxanes experienced a partial response lasting 4 months, and five patients had stable disease lasting > or =3 months. LY355703 at a dose of 1.5 mg/m2 is recommended for Phase II evaluation on a days 1 and 8 schedule. Twice-weekly dosing did not allow improvement in dose intensity or tolerability.