Cellworks Study Predicted Non-Response to Azacitidine in MDS Patients with 100% Accuracy Using AI-Driven Biosimulation

Analysis Uncovered Possible Mechanisms for AZA Resistance That Could Be Targeted to Induce Response

Cellworks Group, Inc., a leader in Precision Medicine and a global pioneer of Therapy Response Index (TRI) technology, announced that its AI-driven biosimulation technology predicted resistance to Azacitidine (AZA) in newly diagnosed Myelodysplastic Syndromes (MDS) patients with high accuracy. In a cohort of 37 intermediate and high-risk MDS patients, Cellworks predicted AZA non-responders with 100% accuracy.

The results from this study will be featured as a poster presentation (Abstract 3087) at the American Society of Hematology (ASH) Annual Meeting and Exposition held December 1-4, 2018 in San Diego, California. ASH attendees can also get more information by visiting Cellworks booth #201 in the San Diego Convention Center.

Read More: Interview with Victor Zabrockis, Chairman at JET8 Foundation

“Azacitidine is currently a drug of choice for most of high-risk MDS patients, although only 40-50% experience clinical improvement,” said Shireen Vali, Cellworks Co-founder & CSO. “There is a clear need for a predictive clinical decision support tool to identify MDS patients with a higher or lower likelihood of AZA response. By predicting whether an individual tumor will respond to Azacitidine, we can spare patients life-threatening toxicities and medical expenses for a therapy that has no chance of response, while patients with a high chance for response would receive maximized treatment.”

Read More: Interview with Hillary Henderson (IBM Watson Media) and Richie Hyden (IRIS.TV)

For this study, Cellworks analyzed the clinical and genomic (NGS, cytogenetics and FISH) data for a cohort of 48 Int-2 and high risk MDS patients who were treated with AZA for median of 12 (4-34) cycles. Patients were treated by AZA until progression to AML.

Cellworks created a Computational Biology Model (CBM) for 37 out of 48 patients utilizing genomic data to create a predictive workflow complemented with digital mechanistic model of AZA and other FDA approved drugs. Drugs were modeled by programming their mechanism of action on pathways and simulated individually and in combination. A disease inhibition score characterized the drug impact on inhibition of the disease phenotypes. For all AZA non-responder profiles, unique combinations were identified that could produce response.

Read More: Blockchain-Automobile Collaboration Sending Huge Clean Waves Through Green Automotive Industry in Joint Venture Agreement Between Sino United and InnoTAD Group