Medidata AI Presents CAR-T Research at the 65th American Society of Hematology Annual Meeting

Medidata Solutions is excited to present three posters at the 65th American Society of Hematology (ASH) annual meeting, Dec 9–12 in San Diego, CA. The research highlights the clinically impactful results made possible by our academic-industry collaborations, and the need to continue advancing the field’s understanding of the CAR-T (and broader immunotherapy) landscape collectively.

This work has been done as a part of the newly-launched Medidata Research Alliance, which brings together academic and industry key opinion leaders to broaden the collaborative reach of this scientific research and its impact on patients.

Phosphorus Disruption is Associated with the Incidence and Severity of Neurotoxicity Symptoms in CD19-Targeted CAR T-Cell Therapy: A Pooled Clinical Trial Analysis

Authors: Penelope Lafeuille, MS¹ ; Jack Pengfei Tang, BSc² ; Sheila Diamond, MS, CGC¹ ; Alexander Socolov, MBAn¹ ; Jacob Aptekar, MD, PhD¹ ; Theodore Nowicki, MD, PhD² 

Affiliations: ¹Medidata, a Dassault Systèmes company, New York, NY; ²UCLA

Hypophosphatemia occurs when phosphorus levels in the blood are low, and can be caused by increased cell metabolic activity. This neurologically presents in a similar manner to immune effector cell-associated neurotoxicity syndrome (ICANS), which is an adverse event associated with CAR-T cell therapy. Through a research collaboration with Theodore Scott Nowicki, MD, PhD (UCLA), Medidata AI explored the correlation between serum electrolytes and the risk of ICANS in CAR-T clinical trial patients.

After analyzing over 500 patients diagnosed with either r/r B-ALL or non-Hodgkin's lymphoma, and either having experienced ICANS or not, CRS and electrolyte disturbances were statistically significant between the two groups. Findings suggest that monitoring phosphorus levels might be beneficial as a biomarker for ICANS. Additionally, supplementing phosphorus may help to mitigate the incidence of ICANS in some of these CAR-T patients. 

Date: December 10, 2023, 6–8 PM PST

See the poster.

Significant Cytokine Release Syndrome Risk Model with T-Cell Engaging Therapies 

Authors: Pénélope Lafeuille¹ , William A. Blumentals² , Claire Brulle-Wohlhueter² , Weixi Chen¹ , Chao Sang¹ , Sydney Manning¹ , Silvy Saltzman¹, Jan Canvin² , Susan Richards² , Cris Kamperschroer² , Giovanni Abbadessa² , Aniketh Talwai¹ , Caroline Der-Nigoghossian¹ , Yahav Itzkovich¹ , Vibhu Agarwal¹ , Rahul Jain¹ , Tanmay Jain¹ , Jacob Aptekar¹ , Stephen Grupp³ 

Affiliations: ¹Medidata Solutions, New York City, NY, USA; ²Sanofi, Cambridge, MA, USA; ³Children's Hospital of Philadelphia, Philadelphia, PA, USA

Cytokine release syndrome (CRS) is an adverse event that can occur in patients who receive T-cell engaging (TCE) therapies such as BiTEs and DARTs. CRS happens when your body’s immune system fights back against foreign cells infused in the body. The challenge with CRS is that it emerges a few days after infusion and the severity varies from patient to patient. More importantly, it has been historically very difficult to predict which patients will have CRS. While mild CRS leads to symptoms such as fever, in severe cases, CRS can lead to organ failure or even death.

Our project involved mining patient-level data from past clinical trials of T-cell engager therapies and building a model to predict which patients are at high risk of severe CRS by looking at their baseline characteristics, disease history, prior treatments and other pre-infusion variables such as lab analytes and vital signs. The goal of this research is to enable physicians and investigators to identify which patients are at high risk of CRS before they are treated with T-cell engager therapies, so that they can be treated effectively and safely.

Date: December 10, 2023, 6–8 PM PST

See the poster.

Machine Learning-based Decision Tree for Identifying and Grading Severity of ICANS Based on Neurological Adverse Events in Patients Treated with CD19 CAR T-Cell

Authors: Alexandru Socolov, MBAn¹, Penelope Lafeuille, MS¹, Sheila Diamond, MS, CGC¹, Eric Yang, PhD¹, Francesco Maura, MD², Jacob Aptekar, MD, PhD¹ , Esther Nie, MD, PhD³

Affiliations: ¹Medidata, a Dassault Systemes company, New York, NY; ²Sylvester Comprehensive Cancer Center, University of Miami, Miami, FL; ³Department of Neurology, Stanford University, Stanford, CA

Managing ICANS (neurotoxicity) as a side effect associated with CAR-T therapies remains a significant challenge. Through a research collaboration with Esther Nie, MD, PhD (Stanford) and Francesco Maura, MD (UMiami), Medidata AI applied machine learning with clinical expertise to develop a decision tree for clinical management of ICANS for patients undergoing CAR-T therapy. After analyzing ~400 patients who had neurological adverse events after CAR-T therapy, the team was able to accurately predict, with machine learning, the severity of ICANS in these patients. This type of clinical decision support tool has the power to help physicians make critical decisions earlier on about their patients. Because of the potential severity of ICANS, having even earlier knowledge about how a patient might react to a treatment can be crucial to improving outcomes. 

Date: December 11, 2023, 6–8 PM PST

See the poster.

Medidata AI is able to leverage RWD and HCTD, as well as clinical insights and machine learning, in order to provide insights that have proven crucial to patient safety and outcomes, especially in the CAR-T space. More in-depth research will be presented on all three of these findings by Medidata AI at the 65th American Society of Hematology Annual Meeting December 9–12 in San Diego, California.