The OSCAR Study: Unlocking the Power of Real-World Evidence in Cancer-Associated Thrombosis

Craig Coleman, PharmD

RWE is significantly impacting the scientific community by providing practical and relevant data in real-world settings. We interviewed Craig Coleman, PharmD, Professor at the University of Connecticut, and the Observational Study in Cancer-Associated Thrombosis for Rivaroxaban (OSCAR) study's principal investigator, to understand the significance and benefits of it. The OSCAR study examined the efficacy and safety of rivaroxaban and apixaban in treating Cancer-Associated Thromboembolism in active cancer patients. The study's results, which are posted on clinicaltrials.gov, will be discussed to illuminate RWE research's challenges and opportunities.

Moe Alsumidaie: What was the purpose of the Observational Study in Cancer-Associated Thrombosis for Rivaroxaban (OSCAR)?

Craig Coleman: (OSCAR), is a retrospective, head-to-head analysis, which shows XARELTO® (rivaroxaban) as effective and safe in treating Cancer-Associated Thromboembolism (CAT) as apixaban. Researchers evaluated these agents in a population where current guidelines suggest we should use them. The study assessed the time to first composite event of recurrent Venous Thromboembolism (VTE) or any bleeding resulting in hospitalization at a minimum follow-up of 3 months.

The (OSCAR) study included a cohort of 2,437 patients 18 years of age and older with CAT for whom guidelines as alternatives to low molecular weight heparin endorse Direct Oral Anticoagulants (DOACs). Patients with active brain, lung, breast, and prostate cancer, among others, experienced a hospital, emergency department, or observation unit admission for VTE or a Pulmonary Embolism (PE) event.

MA: From an endpoint standpoint, what benefits did the study offer to the industry?

CC: We wanted to evaluate the data in a population to which current guidelines suggest we should use them.Another primary goal was to conduct the study in a methodologically accurate way to demonstrate to clinicians whether there was a difference between rivaroxaban and apixaban regarding safety and efficacy. Using a non-inferiority design, we showed a hazard ratio of 0.87, the upper limit of our confidence intervals of 1.27, well below that threshold of 1.5—something that I was very proud of in terms of the study design. Unlike most real-world evidence studies, we ensured that the OSCAR U.S. study was designed rigorously so that we could draw accurate conclusions from our results.

MA: Could you share details about the data, research findings, and uncovered results?

CC: The primary objective of this study was to evaluate the effectiveness and safety of rivaroxaban versus apixaban, two oral factor-tending inhibitors commonly used in the United States and the United Kingdom, alongside many other countries. These patients have either been recently diagnosed with cancer or undergoing active treatment for specific cancers, which puts them at a very high risk compared to a normal population for developing these Venous Thromboembolic events. The study found that when you compared the efficacy and safety of these two agents in a population of patients with active cancer who have experienced Cancer-Associated Thrombosis, patients in need of anticoagulation with rivaroxaban or apixaban, there is no significant difference between those two agents.

Another important aspect of this study is that we excluded cancer types that current clinical guidelines have suggested that rivaroxaban and apixaban should not be used. This differs from other randomized control trials, as they did not have any exclusion criteria.

MA: What was your role in gathering the data for the OSCAR study (U.S.), specifically around some of the challenges you faced during the data-gathering process?

CC: OSCAR is a real-world study group conducted in three countries—the United States, the United Kingdom, and Sweden. As principal investigator of the US portion of OSCAR, I utilized US data specifically. One of the biggest challenges I encountered was working with multiple researchers from different countries using different data sets. Collectively, we needed to standardize the data and clearly understand the statistical analysis plan. We worked to ensure we evaluated the same research question in different countries using a different data set to demonstrate consistency across the board. The more significant challenge of trying to put this together was to keep it as organized as possible.

MA: How did you overcome those challenges?

CC: The team needed to collaborate with the principal investigators from the other OSCAR program countries to develop a common protocol with a clear standard of data definitions, what’s included or excluded, and how the outcomes are defined.

MA: Can you tell me abou the datasets used in the OSCAR study?

CC: As a real-world evidence study, we used preexisting datasets. In the US, we used the US Optimal Electronic Health Record Data Set. In the UK and Sweden, they used their national data set. The datasets are very similar, but there are important differences in data availability and the number of patients. We had to look closely to make sure the studies were comparable.

MA: We have seen an industry demand, particularly from the FDA, to expand the scope of research, whereby minority inclusion is becoming imperative in research. How did OSCAR ensure that minorities were included?

CC: The inclusion of minorities within research is essential as we aim for high external validity for our results and to form concrete analysis, interpretation, and takeaways that can be extrapolated to our diverse population within the United States and, ultimately, across the world. Having a representative sample is particularly important for this research as we intended to study (if any) variance in effect against the variables of ethnicity, race, gender, and sex.

The inclusion of minorities is undoubtedly reflected in the OSCAR program. When conducting a comparative effectiveness study, looking at various treatment strategies, you want to consider all variables and get as granular as possible within research findings, making inclusion critical to valid inferences from the data.

MA: Why is it important to report the study’s results on clinicaltrials.gov?

CC: I want to highlight an interesting aspect of this study: how transparent this real-world evidence study was. We registered this real-world OSCAR US study on the clinicaltrials.gov website.

There, we published our protocol so everyone could see what we were doing. Therefore, when we released the results and presented them at the American Society of Hematology annual meeting, people were able to go back to that protocol and say, “look, yes, they did what they said they were going to do from day one. They didn't tweak the methodology to show what they wanted to show essentially”. We followed the methodology and protocol we laid out from the beginning. This is not as common as I would like to see in medical literature for real-world evidence studies.

Moe Alsumidaie, MBA, MSF, is a thought leader and expert in the application of business analytics toward clinical trials, and Editorial Advisory Board member for and regular contributor to Applied Clinical Trials.