Looking ahead in early-phase trial design to improve the drug development process: examples in oncology.

BACKGROUND: Clinical trial design must consider the specific resource constraints and overall goals of the drug development process (DDP); for example, in designing a phase I trial to evaluate the safety of a drug and recommend a dose for a subsequent phase II trial. Here, we focus on design considerations that involve the sequence of clinical trials, from early phase I to late phase III, that constitute the DDP. METHODS: We discuss how stylized simulation models of clinical trials in an oncology DDP can quantify important relationships between early-phase trial designs and their consequences for the remaining phases of development. Simulations for three illustrative settings are presented, using stylized models of the DDP that mimic trial designs and decisions, such as the potential discontinuation of the DDP. RESULTS: We describe: (1) the relationship between a phase II single-arm trial sample size and the likelihood of a positive result in a subsequent phase III confirmatory trial; (2) the impact of a phase I dose-finding design on the likelihood that the DDP will produce evidence of a safe and effective therapy; and (3) the impact of a phase II enrichment trial design on the operating characteristics of a subsequent phase III confirmatory trial. CONCLUSIONS: Stylized models of the DDP can support key decisions, such as the sample size, in the design of early-phase trials. Simulation models can be used to estimate performance metrics of the DDP under realistic scenarios; for example, the duration and the total number of patients enrolled. These estimates complement the evaluation of the operating characteristics of early-phase trial design, such as power or accuracy in selecting safe and effective dose levels.

Using Patient-Reported Outcomes in Dose-Finding Oncology Trials: Surveys of Key Stakeholders and the National Cancer Research Institute Consumer Forum.

BACKGROUND: Patient-reported adverse events may be a useful adjunct for assessing a drug's tolerability in dose-finding oncology trials (DFOT). We conducted surveys of international stakeholders and the National Cancer Research Institute (NCRI) Consumer Forum to understand attitudes about patient-reported outcome (PRO) use in DFOT. METHODS: A 35-question survey of clinicians, trial managers, statisticians, funders, and regulators of DFOT was distributed via professional bodies examining experience using PROs, benefits/barriers, and their potential role in defining tolerable doses. An 8-question survey of the NCRI Consumer Forum explored similar themes. RESULTS: International survey: 112 responses from 15 September-30 November 2020; 103 trialists [48 clinicians (42.9%), 38 statisticians (34.0%), 17 trial managers (15.2%)], 7 regulators (6.3%), 2 funders (1.8%)]. Most trialists had no experience designing (73, 70.9%), conducting (52, 50.5%), or reporting (88, 85.4%) PROs in DFOT. Most agreed that PROs could identify new toxicities (75, 67.0%) and provide data on the frequency (86, 76.8%) and duration (81, 72.3%) of toxicities. The top 3 barriers were lack of guidance regarding PRO selection (73/103, 70.9%), missing PRO data (71/103, 68.9%), and overburdening staff (68/103, 66.0%). NCRI survey: 57 responses on 21 March 2021. A total of 28 (49.1%) were willing to spend <15 min/day completing PROs. Most (55, 96.5%) preferred to complete PROs online. 61 (54.5%) trialists and 57 (100%) consumers agreed that patient-reported adverse events should be used to inform dose-escalation decisions. CONCLUSION: Stakeholders reported minimal experience using PROs in DFOT but broadly supported their use. Guidelines are needed to standardize PRO selection, analysis, and reporting in DFOT.

Quantitative MRI Differentiates Electromyography Severity Grades of Denervated Muscle in Neuropathy of the Brachial Plexus.

BACKGROUND: Quantitative MRI (qMRI) metrics reflect microstructural skeletal muscle changes secondary to denervation and may correspond to conventional electromyography (EMG) assessments of motor unit recruitment (MUR) and denervation. HYPOTHESIS: Differences in quantitative T2 , diffusion-based apparent fiber diameter (AFD), and fat fraction (FF) exist between EMG grades, in patients with clinically suspected neuropathy of the brachial plexus. STUDY TYPE: Prospective. POPULATION: A total of 30 subjects (age = 37.5 ± 17.5, 21M/9F) with suspected brachial plexopathy. FIELD STRENGTH/SEQUENCE: 3-Tesla; qMRI using fast spin echo (T2 -mapping), multi-b-valued diffusion-weighted echo planar imaging (for AFD), and dual-echo Dixon gradient echo (FF-mapping) sequences. ASSESSMENT: qMRI values were compared against EMG grades (MUR and denervation). qMRI values (T2 , AFD, and FF) were obtained for five regional shoulder muscles. A 4-point scale was used for MUR/denervation severity. STATISTICAL TESTS: Linear mixed models and least-squares pairwise comparisons were used to evaluate qMRI differences between EMG grades. Predictive accuracy of EMG grades from qMRI was quantified by 10-fold cross-validated logistic models. A P value < 0.05 was considered statistically significant. RESULTS: Mean (95% confidence interval) qMRI for "full" MUR were T2  = 39.40 msec (35.72-43.08 msec), AFD = 78.35 µm (72.52-84.19 µm), and FF = 4.54% (2.11-6.97%). Significant T2 increases (+8.36 to +14.67 msec) and significant AFD decreases (-11.04 to -21.58 µm) were observed with all abnormal MUR grades as compared to "full" MUR. Significant changes in both T2 and AFD were observed with increased denervation (+9.59 to +15.04 msec, -16.25 to -18.66 µm). There were significant differences in FF between some MUR grades (-1.45 to +2.96%), but no significant changes were observed with denervation (P = 0.089-0.662). qMRI prediction of abnormal MUR or denervation was strong (mean accuracy = 0.841 and 0.810, respectively) but moderate at predicting individual grades (accuracy = 0.492 and 0.508, respectively). DATA CONCLUSION: Quantitative T2 and AFD differences were observed between EMG grades in assessing muscle denervation. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 1.

Age and Sex Comparison of the Canine Supraspinatus Tendon Using Quantitative Magnetic Resonance Imaging T2 Mapping.

OBJECTIVE: The normal canine supraspinatus tendon has properties commonly attributed to damage such as core hyperintensity and increased width. Little is published regarding the normal tendon, including how senescent changes and sex differences may affect the appearance. Conventional magnetic resonance imaging (MRI) techniques provide subjective analysis of tendons based on observer assessment of signal intensity and appearance. Quantitative MRI (qMRI) techniques such as T2 mapping provide an objective comparison of collagen orientation with analysis of a decay constant, T2. This study investigates age and sex related changes in the canine supraspinatus tendon using the qMRI technique of T2 mapping. STUDY DESIGN: In this study, 34 tendons of clinically sound male and female dogs (0.6-13 years) were imaged using qMRI T2 mapping techniques. Sagittal plane T2 maps of the supraspinatus tendon were depth-normalized, and profiles compared using two separate four-parameter logistic equations describing T2 mapping profiles as sigmoidal curves. Combined parameters evaluated included range of T2 values, curve steepness, vertical curve shift, lower bound of T2, upper bound of T2 and horizontal curve shift. RESULTS: A significant reduction in the most central portion of the supraspinatus tendon was found for every increased year in age (-1.56 ± 0.47 milliseconds [-2.56, -0.56, p = 0.004]). No significant difference in curve parameters was found between sexes. CONCLUSION: The reduction in T2 with age suggests a senescent change may be anticipated with the canine supraspinatus tendon.

Implementation of platform trials in the COVID-19 pandemic: A rapid review.

MOTIVATION: Platform designs - master protocols that allow for new treatment arms to be added over time - have gained considerable attention in recent years. Between 2001 and 2019, 16 platform trials were initiated globally. The COVID-19 pandemic seems to have provided a new motivation for these designs. We conducted a rapid review to quantify and describe platform trials used in COVID-19. METHODS: We cross-referenced PubMed, ClinicalTrials.gov, and the Cytel COVID-19 Clinical Trials Tracker to identify platform trials, defined by their stated ability to add future arms. RESULTS: We identified 58 COVID-19 platform trials globally registered between January 2020 and May 2021. According to trial registries, 16 trials have added new therapies (median 3, IQR 4) and 11 have dropped arms (median 3, IQR 2.5). About 50% of trials publicly share their protocol, and 31 trials (53%) intend to share trial data. Forty-nine trials (84%) explicitly report adaptive features, and 21 trials (36%) state Bayesian methods. CONCLUSIONS: During the pandemic, there has been a surge in the number of platform trials compared to historical use. While transparency in statistical methods and clarity of data sharing policies needs improvement, platform trials appear particularly well-suited for rapid evidence generation. Trials secured funding quickly and many succeeded in adding new therapies in a short time period, thus demonstrating the potential for these trial designs to be implemented beyond the pandemic. The evidence gathered here may provide ample insight to further inform operational, statistical, and regulatory aspects of future platform trial conduct.

The Retail Outlet Health Kiosk Hypertension Trial (ROKHYT): Pilot Results.

BACKGROUND: Blood pressure (BP) control was only 43.7% in the National Health and Nutrition Survey (NHANES) survey in 2017-2018. Scalable, nonclinic-based strategies to control BP are needed. We therefore conducted a pilot trial of a text-messaging intervention in a national network of retail outlet health kiosks with BP devices. All study procedures were conducted remotely. METHODS: Eligible individuals (N = 140), based on average BP greater than or equal to 140/90 mm Hg at kiosks during the prior year, were randomized to intervention vs. usual care. Intervention consisted of tailored text messages providing educational information with embedded links to educational videos on topics related to BP control. BP measurements were obtained at kiosks at 3, 6, and 12 months following randomization; control was defined as BP < 140/90 mm Hg. Follow-up at 12 months was curtailed due to SARS-CoV-2. We therefore combined 12-month (N = 62) or carried forward 6-month (N = 61) data as the primary end point. RESULTS: Participants were 51.4% male, 70.7% white/Caucasian, had mean age of 52.1 years, and mean baseline BP 145.5/91.8 mm Hg. At the end point, 37.7% intervention vs. 27.4% usual care subjects achieved BP control (difference, 10.3%, 95% confidence interval -6.2%, 26.8%). In an intention-to-treat analysis with multiple imputation of missing data, 12-month BP control was 29.0% vs. 19.8% favoring intervention (difference, 9.2%. 95% confidence interval -7.3%, 25.7%); intervention vs. control differences in adjusted mean BP levels were systolic BP: -5.4 mm Hg (95% confidence interval: -13.5, 2.7) and diastolic BP: +0.6 mm Hg (95% confidence interval: -4.2, 5.4). CONCLUSIONS: These pilot results support the potential for a highly scalable text-messaging intervention to improve BP. CLINICAL TRIALS REGISTRATION: Trial Number NCT03515681.

KMDATA: a curated database of reconstructed individual patient-level data from 153 oncology clinical trials.

We created a database of reconstructed patient-level data from published clinical trials that includes multiple time-to-event outcomes such as overall survival and progression-free survival. Outcomes were extracted from Kaplan-Meier (KM) curves reported in 153 oncology Phase III clinical trial publications identified through a PubMed search of clinical trials in breast, lung, prostate and colorectal cancer, published between 2014 and 2016. For each trial that met our search criteria, we curated study-level information and digitized all reported KM curves with the software Digitizelt. We then used the digitized KM survival curves to estimate (possibly censored) patient-level time-to-event outcomes. Collections of time-to-event datasets from completed trials can be used to support the choice of appropriate trial designs for future clinical studies. Patient-level data allow investigators to tailor clinical trial designs to diseases and classes of treatments. Patient-level data also allow investigators to estimate the operating characteristics (e.g. power and type I error rate) of candidate statistical designs and methods. Database URL: https://10.6084/m9.figshare.14642247.v1.

Deviation from the Proportional Hazards Assumption in Randomized Phase 3 Clinical Trials in Oncology: Prevalence, Associated Factors, and Implications.

PURPOSE: Deviations from proportional hazards (DPHs), which may be more prevalent in the era of precision medicine and immunotherapy, can lead to underpowered trials or misleading conclusions. We used a meta-analytic approach to estimate DPHs across cancer trials, investigate associated factors, and evaluate data-analysis approaches for future trials.Experimental Design: We searched PubMed for phase III trials in breast, lung, prostate, and colorectal cancer published in a preselected list of journals between 2014 and 2016 and extracted individual patient-level data (IPLD) from Kaplan-Meier curves. We re-analyzed IPLD to identify DPHs. Potential efficiency gains, when DPHs were present, of alternative statistical methods relative to standard log-rank based analysis were expressed as sample-size requirements for a fixed power level. RESULTS: From 152 trials, we obtained IPLD on 129,401 patients. Among 304 Kaplan-Meier figures, 75 (24.7%) exhibited evidence of DPHs, including eight of 14 (57%) KM pairs from immunotherapy trials. Trial type [immunotherapy, odds ratio (OR), 4.29; 95% confidence interval (CI), 1.11-16.6], metastatic patient population (OR, 3.18; 95% CI, 1.26-8.05), and non-OS endpoints (OR, 3.23; 95% CI, 1.79-5.88) were associated with DPHs. In immunotherapy trials, alternative statistical approaches allowed for more efficient clinical trials with fewer patients (up to 74% reduction) relative to log-rank testing. CONCLUSIONS: DPHs were found in a notable proportion of time-to-event outcomes in published clinical trials in oncology and was more common for immunotherapy trials and non-OS endpoints. Alternative statistical methods, without proportional hazards assumptions, should be considered in the design and analysis of clinical trials when the likelihood of DPHs is high.

To randomize, or not to randomize, that is the question: using data from prior clinical trials to guide future designs.

BACKGROUND: Understanding the value of randomization is critical in designing clinical trials. Here, we introduce a simple and interpretable quantitative method to compare randomized designs versus single-arm designs using indication-specific parameters derived from the literature. We demonstrate the approach through application to phase II trials in newly diagnosed glioblastoma (ndGBM). METHODS: We abstracted data from prior ndGBM trials and derived relevant parameters to compare phase II randomized controlled trials (RCTs) and single-arm designs within a quantitative framework. Parameters included in our model were (i) the variability of the primary endpoint distributions across studies, (ii) potential for incorrectly specifying the single-arm trial's benchmark, and (iii) the hypothesized effect size. Strengths and weaknesses of RCT and single-arm designs were quantified by various metrics, including power and false positive error rates. RESULTS: We applied our method to show that RCTs should be preferred to single-arm trials for evaluating overall survival in ndGBM patients based on parameters estimated from prior trials. More generally, for a given effect size, the utility of randomization compared with single-arm designs is highly dependent on (i) interstudy variability of the outcome distributions and (ii) potential errors in selecting standard of care efficacy estimates for single-arm studies. CONCLUSIONS: A quantitative framework using historical data is useful in understanding the utility of randomization in designing prospective trials. For typical phase II ndGBM trials using overall survival as the primary endpoint, randomization should be preferred over single-arm designs.

The clinical trials landscape for glioblastoma: is it adequate to develop new treatments?

Background: There have been few treatment advances for patients with glioblastoma (GBM) despite increasing scientific understanding of the disease. While factors such as intrinsic tumor biology and drug delivery are challenges to developing efficacious therapies, it is unclear whether the current clinical trial landscape is optimally evaluating new therapies and biomarkers. Methods: We queried ClinicalTrials.gov for interventional clinical trials for patients with GBM initiated between January 2005 and December 2016 and abstracted data regarding phase, status, start and end dates, testing locations, endpoints, experimental interventions, sample size, clinical presentation/indication, and design to better understand the clinical trials landscape. Results: Only approximately 8%-11% of patients with newly diagnosed GBM enroll on clinical trials with a similar estimate for all patients with GBM. Trial duration was similar across phases with median time to completion between 3 and 4 years. While 93% of clinical trials were in phases I-II, 26% of the overall clinical trial patient population was enrolled on phase III studies. Of the 8 completed phase III trials, only 1 reported positive results. Although 58% of the phase III trials were supported by phase II data with a similar endpoint, only 25% of these phase II trials were randomized. Conclusions: The clinical trials landscape for GBM is characterized by long development times, inadequate dissemination of information, suboptimal go/no-go decision making, and low patient participation.