2022: A Remarkable Year for External Control Arms in Clinical Trials
2022 was a remarkable year for external control arms (ECAs) in clinical trials. Developed by Medidata AI, a Synthetic Control Arm® (SCA®) is an external control arm derived from both cross-industry historical clinical trial data and real world data. In contrast to other external controls—static summary measures that do not adequately account for patient baseline difference—an SCA is constructed using carefully selected patient level data to yield a baseline composition that is statistically well balanced with the experimental arm to create an accurate synthetic control group.
From non-small cell lung cancer (NSCLC), to recurrent glioblastoma (rGBM) and ovarian cancer, Medidata AI research has shown that an SCA:
- Is comparable to randomized controlled trials (RCTs)
- Is embraced by the FDA in indications where a control group is hard to recruit or retain
- Can answer previously unanswerable research questions—particularly in early-phase clinical trials
Here’s a look back at our research publications from 2022, where Medidata AI applied innovative statistical methods to improve patient lives and accelerate drug development.
Exploring the Potential of External Control Arms Created from Patient Level Data: A Case Study in Non-small Cell Lung Cancer
Authors: Xiang Yin, PhD; Pallavi S. Mishra-Kalyan, PhD, MS; Rajeshwari Sridhara, PhD; Mark D. Stewart, PhD; Elizabeth A. Stuart, PhD; Ruthanna Davi, PhD
RCTs are the gold standard for evaluation of new medical products. But RCTs may not always be ethical or feasible—especially when studying rare or serious diseases using a control arm therapy that is inadequate.
Challenges in enrolling or maintaining a standard-of-care randomized control arm may compromise timely recruitment, retention, or compliance. Ultimately, low patient recruitment can threaten a study’s integrity, including the validity of results.
What if life sciences organizations could augment a traditional control group—allowing more patients access to potentially life-saving experimental therapy in indications where the standard of care offers little or no hope?
In this report, Ruthanna Davi, PhD, Vice President of Data Science at Medidata AI, Xiang Yin, PhD, Vice President of Statistical Innovation at Medidata AI, and colleagues from the FDA, Friends of Cancer Research, and Johns Hopkins University showed that an external control arm mimics a randomized control so that the external control could be used to augment an RCT when encountered with challenges that threaten the feasibility and reliability of a randomized controlled clinical trial.
These methods are explored in a case study in non-small cell lung cancer (NSCLC) derived from Phase 2 and 3 RCTs from 2004-2013 drawn from the Medidata data repository and Project Data Sphere (projectdatasphere.org). The case study indicated that when balanced for baseline characteristics, the overall survival estimates from the ECA were very similar to those of the target randomized control.
Learn more about our research here.
Building an External Control Arm for Development of a New Molecular Entity: An Application in a Recurrent Glioblastoma Trial for MDNA55
Authors: Antara Majumdar, PhD, MS; Ruthanna Davi, PhD; Martin Bexon, MD; Chandtip Chandhasin, PhD, MS; Melissa Coello; Fahar Merchant, PhD, MS; Nina Merchant, MS
After confirming that an SCA can provide similar estimates of overall survival as a randomized control, Medidata AI put these results to the test with Medicenna, a clinical stage immunotherapy company. Medicenna’s MDNA55 is a new targeted therapy to treat rGBM—an indication that was previously customarily studied in Phase 3 with only randomized controls.
The nature of rGBM and the unmet medical need makes it extremely difficult to recruit and retain patients for study in a clinical setting. Glioblastoma is one of the most aggressive forms of cancer with limited treatment options. In traditional RCTs, patients may hesitate to enroll due to the possibility of being placed in the control group requiring treatment with the standard of care. They may also drop out early upon learning they have not been assigned to the investigational therapy.
Medicenna faced difficulties interpreting the results of their single arm Phase 2 study without a control. The company also needed to design a Phase 3 registrational trial that could successfully recruit and retain patients to demonstrate the effects of MDNA55 for registration.
Medicenna leveraged an SCA to bolster interpretation of their Phase 2 trial and gain a precedent-setting FDA nod for a study design with a hybrid external control arm for their Phase 3 registrational trial. In Medicenna’s upcoming trial, the hybrid SCA will reduce the number of prospective control patients needed by approximately 2/3 or 100 patients. This design will give enrolled patients a greater chance to receive the experimental drug—easing recruitment, reducing control patient drop-out, and accelerating trial timelines without compromising the scientific interpretability of the trial.
Learn more about our research here.
Phase IB Trial Efficacy Estimates via a Clinical Trial Synthetic Control Arm
Authors: Xiang Yin, PhD; Ruthanna Davi, PhD; Elizabeth Lamont, MD, MS, MMSc; Premal Thaker, MD, MS; William Bradley, MD; Charles Leath, MD, MSPH; Kathleen Moore, MD; Khursheed Anwer, PhD, MBA; Lauren Musso; Nicholas Borys, MD
Medidata AI partnered with Imunon Corporation, a clinical-stage company focused on DNA-based immunotherapy, to publish findings on the use of an SCA in a completed Phase Ib ovarian cancer dose-escalating study in the American Society of Clinical Oncology (ASCO)’s journal, the Journal of Clinical Oncology – Clinical Cancer Informatics.
The research team demonstrated that comparing patients from a single-arm study to a rigorously matched external control arm composed of historic clinical trial patients can:
- Enhance understanding of treatment effects in advance of randomized clinical trials
- Inform key nodes in drug development including go/no-go decisions and subsequent follow up randomized trial design
Taken together, this work illustrates that SCAs can increase the scientific value of early-phase clinical trials.
“The Medidata AI Synthetic Control Arm provided reliable estimates of the efficacy endpoints, which allowed for a decrease in the number of patients needed to participate in the subsequent randomized Phase II trial. In addition to decreasing the burden on patients, this helped to accelerate trial timelines and decrease costs,” said Dr. Nicholas Borys, Chief Medical Officer at Imunon. “This truly important progress will lead to further insights and advancements in future trials and, in the end, help bring treatments to patients faster.”
Medidata used advanced statistical methodology to carefully match patients treated with GEN-1, a new investigational therapy, and standard of care therapy to anonymized clinical trial patients from Medidata’s extensive repository of historical clinical trials who received only standard of care therapy. Matching Imunon and anonymized Medidata patients using baseline demographic and disease characteristics, Imunon and Medidata AI found that progression-free survival was prolonged for the patients treated with GEN-1 and standard of care chemotherapy compared to well-balanced historic control patients treated with the same standard of care chemotherapy alone.
In a Phase 1B trial which typically seeks to understand associations between dose and toxicity, Medidata helped Imunon estimate a comparative treatment effect—a value that is normally not available until Phase 3. This larger effect size estimate led to a decision to pursue continued development of GEN-1, a decision to decrease the number of patients enrolled in Imunon’s subsequent randomized Phase 2 trial, and supported a positive application for Fast Track Designation from the FDA.
Learn more about our research here.
About Medidata AI
Medidata AI provides unparalleled clinical trial data, advanced analytics, and industry expertise for pharmaceutical, biotech, and medical device leaders, to help reimagine what is possible, uncover the breakthrough insights, make confident decisions, and pursue continuous innovation from R&D to commercial launch.