Can External Control Arms Improve Clinical-Trial Diversity, Participation? | Deloitte US has been saved
Limited functionality available
By Karla Feghali, senior manager, Deloitte Consulting, LLP
A tiny pool of qualified patients can make it nearly impossible to recruit a representative sample into some clinical trials. This can lead to delayed or canceled trials, which typically come at a high price. (Clinical-trial sponsors can spend up to $100,000 to enroll each patient.) Moreover, nearly 80% of the patients who do participate in clinical trials are white, according to a multi-year study conducted by the US Food & Drug Administration. As a result, some approved therapies might not work as well on people of different races and ethnicities.
Clinical trials randomly split patients into experimental arms (those who receive the new therapy) and control arms (those who receive a placebo and/or the standard of care). Both groups of patients are subjected to the same processes, which can include frequent hospital visits, blood draws, and potentially invasive diagnostics to monitor the disease. However, control-arm patients don’t receive the new therapy. Some pharmaceutical companies are exploring ways external control arms (ECAs) could be used in place of the randomized standard-of-care control arms typically used in clinical trials.
ECAs—also known as synthetic control arms—are based on real world data (RWD) pulled from electronic health records, medical claims, wearable devices, and other sources of patient data. ECAs can help improve patient recruitment and diversity while ensuring that every patient in a trial receives the same therapy. An ECA allows investigators to simulate control populations using historical records from patients with the targeted condition. These patients are meticulously selected based on the same clinical criteria as other participants in the trial, and the treatments they received align with the care protocols under scrutiny. While ECAs won’t work for all clinical trials, they have the potential to expedite trials, reduce delays, and reduce overall costs.
COTA Q&A: Seven questions about external control arms
COTA, Inc., an oncology RWD and analytics company, works with pharmaceutical companies to establish ECAs to support clinical trials. The company was founded 10 years ago by oncologists and data scientists who wanted to improve the understanding of cancer patients and therapies. The company says it is currently involved in about 20 clinical trials that are using ECAs. ConvergeHEALTH by Deloitte and COTA have worked with mutual clients to help access data and analytics in support of ECAs.
I recently spoke with C.K. Wang, M.D., an oncologist and chief medical officer at COTA, and Viraj Narayanan, the company’s senior vice president of life sciences, about ECAs and their potential in clinical trials. Here is an excerpt from that conversation:
Karla: Only about 3% of cancer patients participate in clinical trials. Why do so few eligible patients enroll?
CK: When I was a practicing oncologist, it was always difficult to direct patients to clinical trials that contained a [placebo] control arm. Most cancer patients consider participating in clinical trials as a way to access cutting-edge therapies. But when they learn they might wind up in a control arm that does not contain the novel therapy, they lose interest. Additionally, the majority of clinical trials contain strict and extensive inclusion/exclusion criteria that make many cancer patients ineligible. Lastly, many patients do not have access to clinical trials simply because of where they live.
Karla: Targeted therapies are increasingly focused on niche cohorts, often defined by a specific genomic marker that may only exist in 5% of a particular cancer population. How does that affect clinical trials?
Viraj: The cohorting of patients has become increasingly narrow in oncology due to advances in science and targeted therapies. Targeted therapies now make up about one-third of all oncology pipelines. Those two factors have created demand for ECAs in situations where it would be difficult to enroll enough patients into a control group within a traditional clinical trial.
Karla: Attracting a diverse cohort of patients has always been a challenge for clinical-trial sponsors, particularly in oncology. Why is diversity important?
C.K. It is critical that clinical trial participants reflect the real-world patient population, because the experimental therapies, if approved, will be used in a non-selected patient population. We know that clinical trials tend to attract a relatively homogeneous patient population that is typically white, and younger and fitter than the overall patient population. Secondly, clinical trials—through their strict inclusion and exclusion criteria—often select out an ideal patient population that has a better disease make-up. The combination of these factors can result in approved therapies performing less well in the real world than in clinical trials.
Karla: How do ECAs address the lack of diversity in clinical trials?
C.K.: The diversity of the patient population in an ECA is much greater than what we typically see in clinical trials. This gives researchers a much better picture of how a patient population is treated in the real world. The socioeconomic status, ethnicity, or geography of the patients are not factored in when we construct ECAs. Instead, researchers compare a therapy against a patient sampling that reflects the real world, rather than a very controlled sampling in a clinical trial. Patients in the treatment arm might still be homogeneous but using a more representative data set to serve as the control arm can give researchers a better idea of how a drug will actually perform in the real world.
Karla: Racial biases can exist within claims data, and some pharmaceutical companies want trial participants who are likely to respond in the way they expect them to respond. How do you ensure ECAs aren’t biased?
C.K.: We are aware that all data can be biased. As a result, we try to make sure that our data sets are as representative and inclusive as possible. We encourage companies to not implement overly strict inclusion/exclusion criteria on the ECA so that they don’t create unintentional bias. Rather than excluding the records, we recommend that companies explain the limitations of the data. We also try to help them understand that our data is based on a broader set of patients and might not look like data from a typical randomized control arm.
Karla: The genetic makeup of people in clinical trials is being openly discussed among clinicians, patients, and pharmaceutical companies. But that hasn’t always been the case.
C.K.: That’s right. Several years ago, I had a middle-aged Asian patient who didn’t want the standard of care for his resected colon cancer. He wanted to know how many middle-aged Asian patients I had treated who declined chemotherapy. I didn’t know the answer, and more importantly, I couldn’t tell him how such patients fared. Worse yet, the clinical trial that established the standard of care in his scenario did not disclose the ethnicity or genetic makeup of its participants. That was the first time I gave serious thought to the lack of clinical trial diversity.
Karla: The enactment of the 21st Century Cures Act in late 2016 opened the door to life sciences companies that wanted to leverage electronic medical records data. Will the interoperability rule, which went into effect this year, help make EMR data more accessible?
Viraj: The Cures Act is the reason we are even talking about external control arms. That was the first time life sciences companies had a federal mandate to consider alternatives to traditional clinical trials.
C.K.: We suspect the new [interoperability] rule will help data extraction to be more straightforward and could lead to more complete patient records. It’s too early to know how long it will take to get there as many provider groups and networks may want to hold their information close.
Our company is trying to close the gap between clinical trial data and the RWD. We believe that there are tremendous insights hidden within each patient’s clinical documentation. The company’s mission is to ensure that everyone touched by cancer has a clear path to the right care. There is a hope that this understanding can reduce variations in treatment, improve outcomes, and reduce the cost of care.