Hemanth Kanakamedara, Senior Director, Janssen Pharmaceutical Companies, Johnson & Johnson

We strive to better understand immune-mediated inflammatory diseases (IMIDs) and develop new solutions for patients’ unmet therapeutic needs. Specifically for patients with rare diseases and psoriatic arthritis (PsA), rheumatoid arthritis (RA), and inflammatory bowel disease (IBD), including Crohn’s disease (CD) and ulcerative colitis (UC) – Janssen Immunity Academia and R&D data science teams use real-world data (RWD) to improve the drug development lifecycle.

Apply machine learning and artificial intelligence (AI) to real-world data (RWD), such as government claims, electronic health records, laboratory data, and disease registries, to generate evidence for diagnosis, prognosis, and etiology. This is important for providing relevant context for proper use of novel and existing therapies. Such real-world evidence (RWE) provides greater awareness of patient care needs, patient outcomes, gaps in current treatment options, and visions for new research areas. Specific areas of impact are listed below.

RWD has proven invaluable for biomarker identification and development

RWD, powered by data science, helps us better understand the diseases we work with and the patients affected by them. Janssen uses RWD to create detailed phenotypic profiles. This allows a comprehensive analysis of patient clinical characteristics and immune disorders. This is not only an evaluation of what is in the patient’s chart, but also an evaluation of the lab and images. This information will help develop more accurate disease classification systems using AI and NLP, putting us one step closer to precision medicine.¹ This has helped us identify new disease biomarkers and find promising compounds that target them to advance clinically.

Using data science and RWD to design smarter, more efficient, and more representative clinical trials

Knowledge of the natural history of disease is important for drug development. We utilize RWD to inform the design of intervention studies, including inclusion/exclusion criteria, diagnostic criteria, appropriate follow-up, assumptions to strengthen the study, and other design elements.

RWD is of particular importance for drug development in very rare immune disorders such as fetal and neonatal hemolytic disease (HDFN). HDFN is a condition that occurs during pregnancy when maternal red blood cells or blood group antibodies cross the placenta and cause destruction of red blood cells in the fetus. Randomized controlled trials are often impractical and unethical in such populations. As a result, such patient populations have historically been underserved by conventional clinical development programs. Enroll such patients in a single-arm study with a real-world, externally controlled arm using rigorous methods to control for confounding.

Fundamental to successful treatment of immune disorders is recruiting patients for our studies who reflect the same characteristics as in the real world. Diversity, equity, and inclusion and active patient recruitment built into study inclusion/exclusion criteria are critical to addressing the needs of all patients and ensuring access to innovative therapies. Allows for the ability to improve. Without including all patient subpopulations with immune disorders, it is difficult for researchers to fully understand disease progression and response to treatment in important patient subgroups.

This is especially important in underrepresented and understudied populations. We are constantly asking ourselves whether our sites are well-located and whether we are making clinical trials accessible to all populations of patients with immune disorders. With this in mind, Janssen says he will apply AI and machine learning to his RWD to identify where pockets of patients with rare or hard-to-diagnose diseases lie, and research site placement. It helps us to let you know. You have previously participated in a clinical trial and participated in research.

Because we know that diseases and drugs can affect people differently based on their race and ethnicity, it is important to match clinical trial registries with patient population demographics. A simple yet impactful decision, such as placing a clinical trial site in an accessible location within a historically underserved community, can make a big difference in the ability to reach a representative population. , make sure you learn all about how our new treatments are addressing unmet medical needs across all races, ethnicities and genders.

Integrate digital, real-world endpoints into your exam

Understanding how improvements are seen and measured in real-world clinical settings is key to advancing outcomes for patients with diseases such as CD and other IMIDs. Critical RWD, such as endoscopic videos and histology slides that use computer vision algorithms to measure disease severity, have been incorporated into CD clinical trials to compare standard clinical trial results with real-world clinical settings. It serves as a bridge between the measurements evaluated in

The RWE approach enables the collection of more comprehensive data so that the results of randomized controlled trials (RCTs) can be contextualized to questions about diagnosis, prognosis, and disease etiology. Answers to these questions are also important to clarify the value of changing health conditions.

Comparative study of effects after product launch

Tokenized RWE can also help generate evidence of medical resource utilization and other real-world outcomes during and after clinical trials. Post-launch, monitor product efficacy and safety through RWD analysis. We emulate pragmatic RCTs of approved treatments to alleviate the limitations of traditional case-control designs.²This type of comparative evidence generation is important to inform the actual effectiveness of our treatments.

RWE plays an increasingly important role throughout the product lifecycle in our immunological studies. For more information on our research, please visit our Immunology and R&D Data Science sites.

References

1. Wen, C., othersDeep Phenotyping: Embracing Complexity and Temporality – Towards Scalability, Portability, and Interoperability. J Biomed Inform2020;105:103433. https://doi.org/10.1016/j.jbi.2020.103433
2. Miguel A. Hernan, James M. Robbins.Using big data to emulate target trials when randomized trials are not availableAmerican Journal of Epidemiology, Vol. 183, No. 8, Apr. 15, 2016, pp. 758-764https://doi.org/10.1093/aje/kwv254

About the author:

Hemanth Kanakamedala is Senior Director of Immunology at Janssen R&D Data Sciences. His expertise lies in inferring causality using observed non-randomized data. His work focuses on externally controlled intervention trials, emulating randomized trials using observational data, and integrating patient-centric digital health endpoints in trials. Prior to joining Janssen, he spent 10 years supporting the design and execution of Phase 1-3 randomized controlled trials and non-interventional trials. He holds a degree in mathematics and statistics from the University of Massachusetts Amherst.