How to use RWD for good with Najat Khan, Ph.D

Check out how Janssen’s R&D Data Science team is using RWD and AI to bring more targeted treatments to patients, saving lives.
Authors Ryan Fukushima
COO, Tempus

Najat Khan, Ph.D.
Chief Data Science Officer and Global Head, Strategy and Operations, Janssen Research & Development

Ryan Fukushima, COO of Tempus, met with Najat Khan, Ph.D., Chief Data Science Officer and Global Head, Strategy and Operations, R&D, The Janssen Pharmaceutical Companies of Johnson & Johnson, to talk about how Janssen’s R&D Data Science team is using real-world data (RWD) and artificial intelligence (AI) to better understand diseases and the patients impacted by them, with the goal of developing more targeted treatments. Read below to learn more about how Janssen is leveraging the power of data science and digital health to improve patient outcomes.

Questions and responses have been edited for clarity and length.


Ryan: How have real-world data-enabled algorithms helped R&D thus far?

Dr. Khan: We already have targeted therapies for bladder cancer patients with mutations such as FGFR alterations. However, the challenge is that most patients don’t have their tumors sequenced today, delaying access to targeted treatments.

To address this problem, Janssen looked at routinely collected histopathology data and built an algorithm that aims to predict the presence of tumor mutations from the digitized histopathology images – and we’re now deploying these algorithms to guide patient screening efforts for our clinical trials. If you can screen patients for a trial efficiently, you’ll enroll more of the right patients in the right studies.


Ryan: How else have real-world data-enabled algorithms helped R&D thus far?

Dr. Khan: Patients with pulmonary hypertension are often misdiagnosed many times over extended periods of time. Think about what that does to a patient: the uncertainty when you can’t breathe properly and you aren’t receiving the care you need.

We worked with partners to develop an algorithm, which recently received FDA Breakthrough Device designation, that aims to detect subtleties in ECGs – tests frequently done within six months of the onset of symptoms, like shortness of breath – that could suggest a patient has pulmonary hypertension. Earlier detection could avoid a several-year delay in diagnosis and potentially result in earlier access to treatments.

We’re still going through the validation phases of this work. Rigor is important. But this is the kind of difference and change that we’re looking to make for science and for patients.


Ryan: How is real-world data accelerating patient recruitment?

Dr. Khan: One of the challenges with precision medicine is enrolling the right patients into the right trials efficiently. Today, it typically takes months to open a trial site. However, what if researchers could look at real-world data to anticipate when – and where – patients might be eligible for certain trials? That’s a patient-centric model – using real-world data to rapidly open sites where patients are through a ‘just in time’ approach. This is also helping us increase trial diversity, so that the treatments we develop can benefit more patients.


Listen to the full conversation here