Historical Clinical Trials Research Articles

Modern approaches for evaluating treatment effect heterogeneity from clinical trials and observational data.

In this paper, we review recent advances in statistical methods for the evaluation of the heterogeneity of treatment effects (HTE), including subgroup identification and estimation of individualized treatment regimens, from randomized clinical trials and observational studies. We identify several types of approaches using the features introduced in Lipkovich et al (Stat Med 2017;36: 136-196) that distinguish the recommended principled methods from basic methods for HTE evaluation that typically rely on rules of thumb and general guidelines (the methods are often referred to as common practices). We discuss the advantages and disadvantages of various principled methods as well as common measures for evaluating their performance. We use simulated data and a case study based on a historical clinical trial to illustrate several new approaches to HTE evaluation.

Brain metastases in clinical trial participants with KRAS-mutated advanced non-small cell lung cancer receiving docetaxel: Pooled data analysis

ObjectivesLimited data are available on central nervous system (CNS) efficacy with standard-of-care therapies for KRAS-mutated (KRASmut) advanced non-small cell lung cancer (NSCLC). The objective of this study was to investigate the incidence and progression of brain metastases in KRASmut advanced NSCLC treated with docetaxel using pooled data from historical clinical trials. Materials and MethodsData from phase 2/3 trials of docetaxel-containing regimens in advanced NSCLC were sourced from the Medidata platform. Analysis was restricted to stage IIIB-IV KRASmut NSCLC with disease progression after ≥ 1 systemic anticancer therapy. Participants with asymptomatic, treated, and stable brain metastases were included. Endpoints included 12-month CNS disease control rate (CNS-DCR) and CNS progression per Response Evaluation Criteria in Solid Tumors; progression-free survival (PFS); and overall survival (OS). Data were pooled and analyses stratified by baseline brain metastases status. ResultsA total of 595 participants were included in the analysis (62 [10%] with baseline brain metastases and 533 [90 %] without). Among participants with brain metastases, 17 (27.4 %) had CNS progression during docetaxel treatment and 12-month CNS-DCR was 75.8 %; 45 (8.4 %) participants without baseline brain metastases developed brain metastases during treatment. In an analysis restricted to patients with metastatic disease, outcomes with and without baseline brain metastases included: median PFS, 3.3 and 4.9 months (p < 0.005); 12-month PFS, 5 % and 16 %; median OS, 6.9 and 10.4 months (p < 0.005); and 12-month OS, 20 % and 44 %, respectively. ConclusionThese findings establish CNS progression rates with docetaxel in previously treated KRASmut advanced NSCLC and facilitate interpretation of data from ongoing randomized clinical trials of novel KRAS-targeted therapeutic strategies vs. docetaxel.

Prediction of therapeutic response and cancer outcomes in solid tumours via in silico clinical trials.

e13603 Background: Oncology therapeutic development has high failure rates with substantial costs. Advances in technology, including molecular characterisations of cancer and computational power, provide the opportunity to better model therapeutic response for existing and emerging therapies. Methods: Concr adapted Bayesian statistical principles used by astrophysicists to effectively integrate multi-dimensional data (genomic &amp; transcriptomic) for accurate modelling of cancer biology. Our proprietary computational framework was applied to predict clinical response and survival through construction of digital twins and in silico simulations of clinical trials. Predicted log odds ratios (LOR) were generated for drug response and compared to published data from 8 suitable historical clinical trials. Three distinct models underpin this framework: Drug Efficacy, Treatment Response &amp; Overall Survival (OS). Three publicly available datasets were used for training: Cancer Therapeutic Response Portal ( in vitro dose-response data), Cancer Cell Line Encyclopaedia and TCGA. Results: For all 8 clinical studies, the digital twin model accurately simulated both trial arms, compared drug efficacy across arms and predicted which treatment was most active. Unblinded evaluation: Our model accurately predicted gemcitabine would have greater clinical benefit than 5-FU (LOR -0.10, P &lt;0.0001) in metastatic pancreatic cancer. In advanced breast cancer, the model predicted docetaxel had a higher response rate than doxorubicin, and carboplatin response rates were similar to docetaxel in BRCA wild-type patients. In platinum-sensitive recurrent ovarian cancer, the model predicted cisplatin-based combination had a higher response rate than paclitaxel. Blinded evaluation: Using data for paclitaxel, the model correctly predicted that nab-paclitaxel+gemcitabine response rates were higher than gemcitabine in metastatic pancreatic cancer (predicted LOR -0.090, p = &lt;0.001). In all 3 early breast cancer studies, predictions mirrored clinical trial results, adjuvant ECF &gt; CMF, adjuvant TC &gt; AC and adjuvant tamoxifen was superior to capecitabine. The underlying pan-tumour Overall Survival model used a Random Survival Forest approach, trained on 23 cancers with predictive accuracy of 0.78 AUC ROC and C-index 0.71. To enable cohort enrichment for drug response, the model segmented patient cohorts into treatment responder and non-responder. Using this approach, the log odds increased for drug response rate for 16 of the 19 different cancers and 14 of the different 17 drugs. Conclusions: Our digital twin model represents transformative technology for clinical trial simulation with strong potential for accelerating and de-risking clinical drug development, and utility for synthetic control arms.

Bacteriophage therapy in musculoskeletal infections: from basic science to clinical application.

The treatment of musculoskeletal infections (MSIs), including periprosthetic joint infection (PJI) and fracture-related infection (FRI), is often complicated by biofilm-related challenges necessitating multiple revision surgeries and incurring substantial costs. The emergence of antimicrobial resistance (AMR) adds to the complexity of the problem, leading to increased morbidity and healthcare expenses. There is an urgent need for novel antibacterial strategies, with the World Health Organization endorsing non-traditional approaches like bacteriophage (phage) therapy. Phage therapy, involving the targeted application of lytic potent phages, shows promise in the treatment of MSIs. Although historical clinical trials and recent case studies present significant milestones in the evolution of phage therapy over the past century, challenges persist, including variability in study designs, administration protocols and phage selection. Efforts to enhance treatment efficacy consist of personalized phage therapy and combination with antibiotics. Future perspectives entail addressing regulatory barriers, standardizing treatment protocols, and conducting high-quality clinical trials to establish phage therapy's efficacy for the treatment of MSIs. Initiatives like the PHAGEFORCE study and the PHAGEinLYON Clinic programme aim to streamline phage therapy, facilitating personalized treatment approaches and systematic data collection to advance its clinical utility in these challenging infections.

Augmenting external control arms using Bayesian borrowing: a case study in first-line non-smallcell lung cancer.

Aim: This study aimed to improve comparative effectiveness estimates and discuss challenges encountered through the application of Bayesian borrowing (BB) methods to augment an external control arm (ECA) constructed from real-world data (RWD) using historical clinical trial data in first-line non-small-cell lung cancer (NSCLC). Materials & methods: An ECA for a randomized controlled trial (RCT) in first-line NSCLC was constructed using ConcertAI Patient360™ to assess chemotherapy with or without cetuximab, in the bevacizumab-inappropriate subpopulation. Cardinality matching was used to match patient characteristics between the treatment arm (cetuximab+chemotherapy) and ECA. Overall survival (OS) was assessed as the primary outcome using Cox proportional hazards (PH). BB was conducted using a static power prior under a Weibull PH parameterization with borrowing weights from 0.0 to1.0 and augmentation of the ECA from a historical control trial. Results: The constructed ECA yielded a higher overall survival (OS) hazard ratio (HR) (HR=1.53; 95% CI: 1.21-1.93) than observed in the matched population of the RCT (HR=0.91; 95% CI: 0.73-1.13). The OS HR decreased through the incorporation of BB (HR=1.30; 95% CI: 1.08-1.54, borrowing weight=1.0). BB was applied to augment the RCT control arm via a historical control which improved the precision of the observed HR estimate (1.03; 95% CI: 0.86-1.22, borrowing weight=1.0), in comparison to the matched population of the RCT alone. Conclusion: In this study, the RWD ECA was unable to successfully replicate the OS estimates from the matched population of the selected RCT. The inability to replicate could be due to unmeasured confounding and variations in time-periods, follow-up and subsequent therapy. Despite these findings, we demonstrate how BB can improve precision of comparative effectiveness estimates, potentially aid as a bias assessment tool and mitigate challenges of traditional methods when appropriate external data sources are available.

Validation of computational fluid dynamics models for airway deposition with SPECT data of the same population

This study compared computational fluid dynamic (CFD) model predictions on aerosol deposition in six asthmatic patients to the in-vivo results of the same group. Patient-specific ventilation and internal air distribution were prescribed using inspiratory and expiratory CT scans of each patient, accounting for individual lobar air flow distribution. Moreover, the significant influence of realistic mouth and throat geometries on regional deposition is demonstrated. The in-vivo data were obtained from single photon emission computed tomography (SPECT) in 6 subjects with mild asthma selected from a database of historical clinical trials. The governing flow and particle tracking equations were solved numerically using a commercial CFD tool, and the modeled deposition results were compared to the SPECT data. Good agreement was found between the CFD model, applying k-ω SST turbulence model, and SPECT in terms of aerosol deposition. The average difference for the lobar deposition obtained from CFD model and SPECT/CT data was 2.1%. The high level of agreement is due to applying patient specific airway geometries and inspiratory/expiratory CT images, anatomical upper airways, and realistic airway trees. The results of this study show that CFD is a powerful tool to simulate patient-specific deposition if correct boundary conditions are applied and can generate similar information obtained with functional imaging tools, such as SPECT.

FRAMM: Fair ranking with missing modalities for clinical trial site selection

The underrepresentation of gender, racial, and ethnic minorities in clinical trials is a problem undermining the efficacy of treatments on minorities and preventing precise estimates of the effects within these subgroups. We propose FRAMM, a deep reinforcement learning framework for fair trial site selection to help address this problem. We focus on two real-world challenges: the data modalities used to guide selection are often incomplete for many potential trial sites, and the site selection needs to simultaneously optimize for both enrollment and diversity. To address the missing data challenge, FRAMM has a modality encoder with a masked cross-attention mechanism for bypassing missing data. To make efficient trade-offs, FRAMM uses deep reinforcement learning with a reward function designed to simultaneously optimize for both enrollment and fairness. We evaluate FRAMM using real-world historical clinical trials and show that it outperforms the leading baseline in enrollment-only settings while also greatly improving diversity.

Neisseria gonorrhoeae vaccines: a contemporary overview.

Neisseria gonorrhoeae infection is an important public health issue, with an annual global incidence of 87 million. N. gonorrhoeae infection causes significant morbidity and can have serious long-term impacts on reproductive and neonatal health and may rarely cause life-threatening disease. Global rates of N. gonorrhoeae infection have increased over the past 20 years. Importantly, rates of antimicrobial resistance to key antimicrobials also continue to increase, with the United States Centers for Disease Control and Prevention identifying drug-resistant N. gonorrhoeae as an urgent threat to public health. This review summarizes the current evidence for N. gonorrhoeae vaccines, including historical clinical trials, key N. gonorrhoeae vaccine preclinical studies, and studies of the impact of Neisseria meningitidis vaccines on N. gonorrhoeae infection. A comprehensive survey of potential vaccine antigens, including those identified through traditional vaccine immunogenicity approaches, as well as those identified using more contemporary reverse vaccinology approaches, are also described. Finally, the potential epidemiological impacts of a N. gonorrhoeae vaccine and research priorities for further vaccine development are described.

OVEREXPRESSION OF CYCLIN D1 IN ORAL CANCER INCREASES TPF CHEMOTHERAPEUTIC DRUGS VIA ACTIVATING THE CASPASE-3 PATHWAY

Previous studies have indicated that induction chemotherapy can limit the possibility of early lesion eradication and lessen the severity of globally advanced tumors. This study aimed to ascertain if cyclin D1 expression in OSCC patients who were qualified for TPF induction chemotherapy might be regarded as a valid prognostic indicator for long-term outcomes. This study aimed to determine whether or not there was a connection between OSCC cell intervention and the production of cyclin D1 and chemosensitivity to TPF medicines through the apoptotic mechanism. The current work used immunohistochemistry to determine the expression of cyclin D1 in various tissues. To ascertain the survival rate of 232 people diagnosed with locally advanced oral squamous cell carcinoma (OSCC), the study utilized the log-rank test and Kaplan-Meier analysis. The participants in this historical clinical trial were observed for a mean duration of five years. Cytotoxicity assays were employed to assess the therapeutic effectiveness of TPF chemotherapy drugs. The activity of caspase-3 and PARP was evaluated in the HB96, CAL27, and HN30 cell lines following cyclin D1 inhibition, respectively. Long-term clinical outcomes were significantly better for patients with OSCC whose cancer cells expressed low amounts of cyclin D1 than those whose cells expressed high protein levels. Time to initial illness occurrence, time to initial local recurrence, time to mortality, and overall survival were among the metrics evaluated. The groups with low and high expression levels of cyclin D1 showed significantly different survival rates for disease-free survival, overall survival (OS), DMFS, local recurrence-free survival, and local recurrence-free survival (p0.001, 0.003, 0.004, and 0.001, respectively). Patients with cN2 oral squamous cell carcinoma (OSCC) with higher cyclin D1 expression had better distant metastasis-free survival (OS, P=0.024; DMFS, P=0.024) and overall survival (OS, P=0.024; DMFS, P=0.024) with TPF induction chemotherapy. However, patients with cN2 OSCC and low cyclin D1 expression did not show this survival advantage (P=0.024). Research has shown that oral squamous cell carcinoma (OSCC) is more sensitive to TPF-based chemotherapeutic drugs using caspase-3-mediated apoptosis that occurs when cyclin D1 expression rises. Based on these results, TPF induction chemotherapy may provide longer-term benefits than traditional treatment approaches for patients with cN2 OSCC and high cyclin D1 expression. Through a caspase-3-dependent mechanism, oral squamous cell carcinoma (OSCC) cell lines become more susceptible to TPF chemotherapeutic drugs when cyclin D1 is overexpressed.

The Validity of a Synthetic Control Arm Derived from Historical Multiple Myeloma Clinical Trials

Introduction While randomized phase III clinical trials (RCTs) are the gold-standard for establishing drug efficacy in the United States, external control arms (ECAs) may be valuable for both early and late phase drug development. An ECA is a collection of patients with the disease of interest who were treated outside of a clinical trial of interest (ie, target trial) and whose outcomes are compared to those of the target trial patients' to evaluate comparative drug efficacy and/or safety. ECAs created through rigorous statistical non-experimental methods such as matching patient-level characteristics may provide early estimates of comparative efficacy in advance of RCTs and offset prospective enrollment of control patients in randomized studies. We evaluate the validity of using a matched ECA to estimate the comparative efficacy of an experimental drug in relapsed refractory multiple myeloma (r/r MM). Methods From the anonymized Medidata Enterprise Data Store (MEDS), we identified historic RCTs in r/r MM containing patients exposed to an experimental drug as well as an active control drug (ie, a corticosteroid). Arbitrarily, we selected one trial to serve as the “target trial”. For the remaining trials, we collapsed all corticosteroid control patients into a single pool to construct an ECA. Using propensity score (PS) optimal full matching, we identified corticosteroid ECA patients by selecting from the candidate patients those whose baseline demographic and clinical attributes were similar to patients receiving the experimental drug in the target trial. Using the Kaplan-Meier (KM) estimator and a Cox proportional hazards regression, we first compared overall survival time (OS) of patients in the experimental drug arm vs control arm in the target trial. We then compared the OS in the experimental drug arm vs the ECA. Results We found fewer than five historic clinical trials from MEDS which were open-label RCTs conducted from 2010-2017. They enrolled r/r MM patients aged ≥18 years who had received more than two prior lines of therapy, one of which was lenalidomide and bortezomib. The target trial contained 300 patients in the investigational drug arm and 152 patients in the control arm. Using the PS matching, we identified 183 patients from the control arms in pooled trials that were matched with 290 patients in the investigational drug arm in the target trial (i.e., ECA). These ECA patients were then upweighted to be matched with 290 patients in the investigational drug arm in the target trial. Full matching successfully attenuated baseline differences between the investigational drug and ECA arms. Figure 2 shows the KM curves of OS (investigational vs control) in the target trial. Median OS was 13.25 months vs. 8.52 months (log-rank p&amp;lt;0.01) and the investigational drug was associated with a 36% reduction in the hazard of death (HR 0.74, 95% CI: 0.60-0.92). Figure 2 shows the KM curves of OS, after full matching, using the ECA. Median OS was 13.51 months vs. 8.75 months (log-rank p&amp;lt;0.01) and the investigational drug was associated with a 36% reduction in the hazard of death (HR 0.74, 95% CI: 0.58-0.94). Conclusions In this methodologic validation study, we have demonstrated that it is possible to produce an ECA from historical clinical trials data using propensity score methods which replicates the comparative efficacy estimate from a target without introducing bias. That is, the ECA, which was well balanced with the investigational arm with respect to baseline patient attributes, returned nearly identical estimates of comparative efficacy as assessed by OS of both target trial control arm and ECA against the target trial experimental arm. This work illustrates that it is possible to create a valid ECA even when historical data size is limited and to do so without excessive exclusion of non-matched patients. That the treatment effect on OS estimated in comparison to the randomized control was very closely matched by that of the ECA suggests this approach to ECAs may be used to (1) estimate comparative efficacy of therapies that have only been studied in single arm settings and/or (2) augment or replace a randomized control arm in future trials when ethically or practically challenging. Figure 1A. Kaplan-Meier OS estimates of target trial experimental (N=300) vs. control arm (N=152) Figure 1B. Kaplan-Meier OS estimates of target trial experimental (N=290) vs. ECA (N=290)

Clinical Outcomes of Patients with Newly Diagnosed Diffuse Large B-Cell Lymphoma in a County Hospital System

Introduction: Advances in the treatment of diffuse large B-cell lymphoma (DLBCL) have improved survival, with Surveillance, Epidemiology, and End Results (SEER) Program data showing 5-year overall survival (OS) of 64.7% (SEER*Explorer, 2023). However, insurance status in the US is related to DLBCL survival (Han et al., Cancer 2014), and real-world data for patients who are uninsured or who may have socioeconomic barriers to care remain limited. We performed a retrospective cohort study to examine outcomes for patients with newly diagnosed DLBCL treated in a large safety-net hospital system in the third most populous county in the US. Methods: Patients aged 18 years and older diagnosed with DLBCL between January 2011 and June 2022 and treated at Harris Health System, Houston, TX, were included in the analysis. Demographic, disease, treatment, response, and follow-up data were abstracted from the electronic medical record. US Census Block Group FIPS codes derived from geocoded home address at the time of initial diagnosis were used to compute the Area Deprivation Index (ADI). Eligibility for cellular therapies including hematopoietic stem cell transplantation (HSCT) and chimeric antigen receptor T-cell therapy (CAR-T) was determined by the abstractors. Time-to-event analysis was performed using the Kaplan-Meier method, and Cox proportional hazards regression was used to examine factors associated with event-free survival (EFS) and OS. Results: Among 461 patients with newly diagnosed DLBCL (Table 1), the median age was 53 years (range, 18-93), 75% were uninsured, and 80% were in the most disadvantaged ADI national quartiles, defined as at or above the fiftieth percentile. Most patients had stage III or IV disease and an Eastern Cooperative Oncology Group Performance Status (ECOG) of 0 to 1 at the time of diagnosis. One hundred ninety-eight patients (44%) had poor-risk disease by the Revised International Prognostic Index (R-IPI). Seventy-five (16%) had a history of HIV infection. Common first-line treatments included R-CHOP ( n = 210, 46%) and R-EPOCH ( n = 183, 40%); 18 received no therapy (3.9%). The median time from diagnosis to treatment initiation (DTI) was 18 days (range, 0-246); 76% completed therapy. Of 425 evaluable patients, 288 (68%) achieved a complete response (CR) and 44 (10%) had a partial response (PR); of these, 74 (22%) relapsed. Twenty-five (58%) of 43 patients eligible for cellular therapies at relapse received them: 14 (56%) underwent autologous HSCT, 8 (32%) received CAR-T, and 3 (12%) had both. Median follow-up time for the cohort was 48.8 months (range, 0.2-147). By the time of analysis, 167 patients (36%) had developed refractory/relapsed disease, and 148 (32%) had died. At 1, 2, and 5 years, EFS was 72%, 64%, and 58%, and OS was 82%, 76%, and 69% (Figure 1). At 1, 2 and 5 years, EFS was 77%, 69%, and 64% and OS was 85%, 79%, and 75% for patients with germinal center B-cell-like (GCB) DLBCL, whereas EFS was 68%, 56%, and 50% and OS was 80%, 73%, and 62% for those with non-GCB DLBCL. In multivariable Cox regression models, predictors of OS were poor-risk disease by R-IPI (HR 6.08 compared to very-good-risk disease, 95% CI 1.02-36.09), ECOG (HR 2.10 for ECOG ≥2 compared to 0-1, 95% CI 1.12-3.95), and normalized LDH (HR 1.07 per fold increase, 95% CI 1.03-1.11). Race/ethnicity, insurance status, and DTI were not associated with significant differences in OS. However, receiving fewer than planned cycles of first-line therapy was associated with higher mortality (HR 2.00, 95% CI 1.12-3.56), as was not attaining CR by the end of first-line therapy (HR 7.90, 95% CI 3.65-17.12). Conclusion: This is one of the largest real-world studies to date examining outcomes of patients with newly diagnosed DLBCL at a safety-net hospital system. Overall, patients had shorter EFS and OS compared to historical clinical trials but had similar OS to SEER estimates. In this health system with robust financial assistance programs and access to cellular therapies for some patients, uninsured status was not associated with inferior outcomes. Addressing barriers to care may improve outcomes for uninsured patients in other settings.

BIOS-01. EFFICACY AMONG THERAPEUTICALLY EFFECTIVE ABSORBED DOSE PATIENTS IN RESPECT-GBM PHASE I OF RHENIUM-186 NANOLIPOSOME (186RNL) IN RECURRENT GLIOBLASTOMA (RGBM) COMPARED TO AN EXTERNAL CONTROL ARM (ECA)

Abstract BACKGROUND ReSPECT-GBM is a Phase 1/2 study of 186RNL administered by convection enhanced delivery (CED) in patients with rGBM. In Phase I, 15 of 24 patients have achieved a therapeutic absorbed dose defined as &amp;gt; 100 Gy. Based on prior work establishing the comparability of a bevacizumab (Bev) comparator to inform early-phase CED study results, OS of the 15 TAD patients were compared to a Bev ECA. METHODS Recent Bev-treated rGBM clinical trial patients meeting the same eligibility criteria as patients in the ReSPECT-GBM study were identified from the Medidata Enterprise Data Store of aggregated and deidentified patient-level historical clinical trial data. Propensity score (PS) methods were used to select the ECA and align the baseline characteristics of the Bev patients to the ReSPECT TAD cohort. OS estimates were generated using the weighted Kaplan-Meier method and compared with a weighted 2-sample 2-sided log-rank test at α = 0.05 based on a data cut-off of December 2022. RESULTS Baseline characteristics of the ReSPECT TAD and ECA patients were well-aligned after PS modeling. Although not achieving statistical significance due to small sample sizes (median OS 12.2 vs 10.1 months, hazard ratio 0.63, 95% CI: 0.26, 1.52), comparison of OS between the ReSPECT TAD and ECA patients demonstrate consistently longer OS at each 6-month interval: 6 months: 85.1% vs 73.3%, 12 mos: 51.1% vs 30.3%, 18 months: 30.6% vs 25.4%, 24 months: 30.6% vs 3.5%. CONCLUSIONS Single-arm studies can be challenging to interpret. By comparing the ReSPECT-GBM outcomes to an ECA of similar patients with balanced baseline characteristics, we can better appreciate the treatment effect of 186RNL administered via CED. The current results are encouraging, and updated results reflecting current follow-up will be presented at the meeting.

Pharmacometrics enhanced Bayesian borrowing approach to improve clinical trial efficiency: Case of empagliflozin in type 2 diabetes.

We report use of a pharmacometrics enhanced Bayesian borrowing (PEBB) approach to leverage historical clinical trial data on a drug product to build models, project the outcome of future clinical trials, and borrow information from these projections to improve the efficiency of future target trials. This design takes a two-stage approach. First, a design phase is performed before target trial data are available to determine the operating characteristics and an appropriate tuning parameter that will be used in the subsequent analysis phase of a chosen target trial. Second, once the target trial data are available, the analysis phase is performed with the determined tuning parameter. This step is where borrowing is applied from these projections to inform the results for the target trial. To illustrate how a PEBB could improve the efficiency of clinical trials, we apply our design to trials with empagliflozin for treating patients with type 2 diabetes. We performed a retrospective evaluation applying the method to a phase III target trial and a hypothetical smaller trial. The type I error could be kept below 10% while increasing the trial power and effective sample size. Our findings suggest that a PEBB has the potential to increase the power of clinical trials, while controlling for type I error, by leveraging the information from previous trials through population pharmacokinetic/pharmacodynamic modeling and simulation.

Clinical characterization of patients with recurrent glioblastoma in trials involving CED and non-CED treatment.

e18845 Background: Glioblastoma (GBM) is the most common and aggressive of the primary malignant brain tumors in adults. Bevacizumab (beva) remains the only currently approved therapeutic for recurrent GBM (rGBM) patients who progress on prior therapy. Despite advances in care, rGBM has a median expected progression-free survival (PFS) of &lt; 4 months and overall survival (OS) of ~8 months. NCCN guidelines [Version 2.2022], and other oncology medical societies unanimously recommend patients seek treatment by entering a clinical trial. Since the late 1990’s a number of studies have explored the use of convection-enhanced delivery (CED) of drugs, a well-accepted and widely published delivery option for the treatment of rGBM. This analysis compares the demographics, medical history and survival outcome of patients enrolled into recently completed clinical trials with beva monotherapy to those entered into CED studies. Methods: Data for this analysis was sourced from the Medidata Enterprise Data Store (MEDS), comprised of over 28,000 historical clinical trials for de-identified aggregate patient-level analyses and from the study and patient-level data in the CED studies referenced in D’Amico [J Neurooncol 2021]. Aggregate summary statistics comparing beva and CED patients were based on combined study- and patient-level data using weighted (by sample size) means, incidence and overall survival (OS) rates and median of medians. Results: 163 rGBM beva patients were identified from MEDS. The CED cohort included 636 patients from MEDS and studies referenced in D’Amico. Cohort comparisons (beva vs CED) found median age: 56 vs 55 years; Caucasian race: 91% vs 93%; male sex: 68% vs 65%; not Hispanic or Latino ethnicity: 90% vs 94%; median OS: 7.9 vs 8.4 months. Over the last 2 decades there was a 23% increase in patients with ECOG Performance Status = 0 and 25% decrease in tumor area size at study enrollment, whereas there was no clear temporal OS trend. Conclusions: Patient demographics and most tumor attribute variables were similarly distributed between the beva and CED cohorts. While there is some indication that recent CED studies have enrolled healthier patients with smaller tumors; as expected, this does not appear to have translated to a consistent survival benefit.

A new approach for estimating and quantifying disease‐modification in Alzheimer’s disease clinical trials

AbstractBackgroundThe standard methodology for estimating treatment effects in Alzheimer’s disease clinical trials has remained unchanged for decades. With an increasing focus on disease‐modifying treatments, there are opportunities for developing more powerful methods for detecting such types of treatment effects and quantifying them in more interpretable ways.MethodsThe Progression Model for Repeated Measures (PMRM) was developed as an alternative to the Mixed Model for Repeated Measures (MMRM). The assumptions and flexibility of the PMRM are like the MMRM except treatment effects are handled differently (Figure 1). The PMRM estimates treatment effects as time‐delays/slowing of progression relative to the control group while the MMRM estimates treatment effects as point differences/reduction in decline on the outcome scale at a given visit. Because the unit of the treatment effect is shared with the time variable, PMRMs can simultaneously handle multiple outcomes in a natural and efficient manner. The power and interpretability of the PMRM was compared to the MMRM and other methods in a range of scenarios based on simulated and historical clinical trial data.ResultsThe PMRM was found to produce accurate and interpretable estimates of treatment effects across a range of simulated scenarios. Using CDR‐SB as an outcome, the PMRM produced the overall highest power to detect treatment effects across all simulated scenarios compared to all studied approaches. When treatment effects were simulated as time delays or slowing of progression, the PMRM offered substantial power gains compared to the MMRM. Using CDR‐SB, ADAS‐cog and ADCS‐ADL as multivariate outcomes, the PMRM was found to outperform models that combined the information in the outcome using the global test statistic and iADRS.ConclusionThe PMRM offers advantages in terms of interpretation and meaningful quantification of treatment effects, also in situations where the average decline observed in the control group is small. PMRMs can seamlessly include multiple outcome measures to estimate a single treatment effect and it was found that PMRMs offered substantially increased power to detect slowing of disease progression compared to conventional methods using both single and multiple outcome measures.

Bayesian Method of Borrowing Study-Level Historical Longitudinal Control Data for Mixed-Effects Models with Repeated Measures.

Bringing historical control information into a new trial appropriately holds the promise of more efficient trial design with more accurate estimates, increased power, and fewer patients allocated to inefficacious control group, provided the historical control data are sufficiently similar to the concurrent control. Interest has been growing over the past few decades in leveraging historical clinical trial on the control arm. However, most of the current historical borrowing methods focus on incorporating patient-level historical control information at only one time point. In this work, we propose a Bayesian hierarchical Mixed effect Models for Repeated Measures to incorporate aggregated study-level longitudinal historical control estimates into the concurrent trial that collected repeated longitudinal data. The simulation study demonstrates that, as compared to one time point data analysis approach, leveraging longitudinal historical control data produces greater power enhancement and mitigates the power loss when the missing data under missing at random mechanism is present. Our work also helps fill the gap of lack of methods borrowing historical longitudinal control data from the published summarized estimates when patient-level control data are not available.

How granularity of data matters in understanding and accelerating racial diversity in U.S. clinical trials.

88 Background: Historically, US clinical trials have been shown to recruit disproportionately large percentages of White patients, raising concerns about the generalizability of clinical trial results to underrepresented racial minority patient populations. Because of this, the FDA has recently reiterated its guidance stating that clinical trials need to reflect the demographic distribution of the US. In this research, we sought to understand how participation of Black patients varies by therapeutic area and geography within the US. Methods: Studying patient-level clinical trial data from an industry-leading historical clinical trial data repository of over 8 million patients from 27,000 clinical trials, we assessed the racial composition of US interventional trials across indications from 2010 to 2021 which encompassed 433,822 clinical trial participants across 2,997 trials. We also analyzed participants’ racial composition within the subset of trials from three distinct therapeutic areas (i.e., oncology n = 118,194, cardiovascular n = 12,281, central nervous system n = 35,533) and distinct sites over the same period. Results: The racial distribution of clinical trial participants in the US across all therapeutic areas was 78% White, 15% Black, and 3% Asian. Within the three distinct therapeutic areas, the proportion of Black clinical trial participants varied, with oncology trials reporting 8.5% Black participants, cardiovascular trials reporting 15.3% Black participants, and central nervous system trials reporting 19.9% Black participants. Black participation also varied within a therapeutic area, depending on the indication, as well as by site, with individual sites contributing differently to the racial diversity of trials. Conclusions: Our analysis shows that US interventional trials enroll 15% Black participants, appearing consistent with the 2020 US Census, which estimates that 14.6% of Americans are Black. However, more granular analyses at the level of therapeutic area, indication and site suggest substantial variation in Black participation in clinical trials including a gap in Black representation in oncology trials vs. US Census estimates. This research suggests that aggregate estimates of racial enrollment may mask dramatic variation by other factors like granularity of disease area and geographic location. The study of historic clinical trial data may yield useful insights for accelerating diverse representation in clinical trials.

Reassessment of Historical Clinical Trials Supports the Effectiveness of Phage Therapy.

Phage therapy has become a hot topic in medical research due to the increasing prevalence of antibiotic-resistant bacteria strains. In the treatment of bacterial infections, bacteriophages have several advantages over antibiotics, including strain specificity, lack of serious side effects, and low development costs. However, scientists dismissed the clinical success of early clinical trials in the 1940s, slowing the adoption of this promising antibacterial application in Western countries. The current study used statistical methods commonly used in modern meta-analysis to reevaluate early 20th-century studies and compare them with clinical trials conducted in the last 20 years. Using a random effect model, the development of disease after treatment with or without phages was measured in odds ratios (OR) with 95% confidence intervals (CI). Based on the findings of 17 clinical trials conducted between 1921 and 1940, phage therapy was effective (OR = 0.21, 95% CI = 0.10 to 0.44, P value < 0.0001). The current study includes a topic review on modern clinical trials; four could be analyzed, indicating a noneffective therapy (OR = 2.84, 95% CI = 1.53 to 5.27, P value = 0.0009). The results suggest phage therapy was surprisingly less effective than standard treatments in resolving bacterial infections. However, the results were affected by the small sample set size. This work also contextualizes the development of phage therapy in the early 20th century and highlights the expansion of phage applications in the last few years. In conclusion, the current review shows phage therapy is no longer an underestimated tool in the treatment of bacterial infections.

Analysis of clinical trial registry entry histories using the novel R package cthist.

Historical clinical trial registry data can only be retrieved by manually accessing individual clinical trials through registry websites. This limits the feasibility, accuracy and reproducibility of certain kinds of research on clinical trial activity and presents challenges to the transparency of the enterprise of human research. This paper presents cthist, a novel, free and open source R package that enables automated scraping of clinical trial registry entry histories and returns structured data for analysis. Documentation of the implementation of the package cthist is provided, as well as 3 brief case studies with example code.

Abstract 1025: Phase Ib trial single-arm efficacy estimates via comparison to a historical clinical trial synthetic control arm

Abstract Background: Innovations in data science and trial design have catalyzed novel clinical research methods which may inform the future success of new experimental therapies more efficiently than previously believed. To guide development of the novel immune agent GEN-1, we compared endpoints experienced by patients from a recent neoadjuvant single-arm phase Ib study of GEN-1 plus standard chemotherapy to those of similar historical clinical trial patients in receipt of standard chemotherapy alone. Methods: To compare safety and efficacy endpoints following first-line neoadjuvant weekly GEN-1 immunotherapy and carboplatin and paclitaxel to standard first-line neoadjuvant carboplatin and paclitaxel in women with advanced ovarian cancer, we first applied key OVATION-1 trial (NCT02480374) inclusion and exclusion criteria to the Medidata Enterprise Data Store (MEDS) data to identify candidate historical clinical trial patients for comparison. We standardized and integrated patient-level MEDS data (N=41) from distinct phase I-III trials (enrollment years 2015-2016) with patient-level OVATION-1 data (N=18). Standard propensity score methods were used to identify MEDS patients who appeared similar to OVATION-1 patients to create a synthetic control arm (SCA). Results: Fifteen OVATION-1 patients (15/18, 83%) were matched to 15 (37%, 15/41) MEDS historical trial control patients. Matching attenuated pre-existing differences in attributes between the OVATION-1 and MEDS patients. The median progression-free survival time was not reached by the OVATION-1 group and was 15.8 months for the SCA. The hazard of progression for the OVATION-1 group relative to the SCA was 0.53 (95% CI 0.16, 1.73). Fourteen of 15 OVATION-1 patients (93.3%) and 15 of the SCA patients (100%, 15/15) had at least one MedDRA toxicity. Compared to SCA patients, OVATION-1 patients had a slightly higher incidence in nausea (OVATION-1 73.3%; SCA 53.3%), fatigue (OVATION-1 73.3%; SCA 33.3%), anorexia (OVATION-1 46.7%; SCA 13.3%), chills (OVATION-1 26.7%; SCA 6.7%), and infusion-related reaction (OVATION-1 26.7%; SCA 0%). Conclusions: The comparison of patient endpoints from a single-arm phase Ib trial to a historical clinical trial SCA provided informative and relatively reliable estimates of efficacy endpoints which were used to inform GEN-1’s expected effect study size in the phase II setting. This information led to a decrease in the number of planned patients for the subsequent randomized phase II trial. More broadly, this approach supports the ability of historical clinical trial patient comparisons to inform drug development via trial design, something which may further increase the scientific value of early phase trials. Citation Format: Xiang Yin, Ruthanna Davi, Elizabeth B. Lamont, Premal H. Thaker, William H. Bradley, Charles A. Leath, Kathleen M. Moore, Khursheed Anwer, Lauren Musso, Nicholas Borys. Phase Ib trial single-arm efficacy estimates via comparison to a historical clinical trial synthetic control arm [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1025.