Accelerating Precision Oncology with Artificial Intelligence

Written by: Anna Israyelyan, MD, PhD
Manager, Biopharma Solutions

Published on:
June 20th, 2017

Anna Israyelyan, MD, PhD
Dr. Israyelyan joined CGI in 2015. She has over 15 years of extensive cancer research and management experience in industry and academia covering wide range of medical areas, including basic, translational, and clinical research. Anna is an active member of the American Society of Hematology (ASH), the American Association for Cancer Research (AACR), and American Association for the Advancement of Science (AAAS). Dr. Israyelyan has a scientific advisory role with matrix management of scientific, technical, and operational aspects related to various biopharma projects and collaborations involving biomarkers in solid tumors and hematologic malignancies.

Artificial Intelligence is creating many opportunities for transformative disruption across the entire oncology value chain – from drug discovery to patient monitoring. And, CGI is leading the charge…

Artificial Intelligence – or AI – has evolved into an umbrella term of sorts to refer to the broad set of techniques and methods to train computers to learn, communicate, reason, and make decisions like humans…in fact, even better than humans. Some call it machine learning, cognitive computing, or deep learning.  Whatever your preference in buzzwords, the evolving trend around AI is undeniable…

AI is transforming all aspects of our lives with its reach into a wide variety of industries, including health care. It is estimated health AI market will register an explosive growth rate in the next 4 – 5 years, reaching $6.6B by 2021 [1]. With evolving applications like automated patient management, system-wide decision support tools, and the onslaught of new precision medicine applications, AI offers several disruptive innovations in our industry…

Why does AI matter in our business?

AI is now being applied to large volumes of data from increasingly diverse sources across the entire oncology value chain.

Here is a short list of select startup enterprises deploying AI strategies to decipher, discover, and pursue novel drug rescue and repurposing development plans:

  • Recursion Pharmaceuticals: using “computer vision” to look at cells and identify over 1,000 features to determine if a sick cell is getting healthier when exposed to 1000s of drug compounds – fetched a partnership with Sanofi-Aventis
  • Insilico Medicine: an AI service provider – or contract research organization – offering big data analysis and deep learning for in silico drug discovery and drug repurposing hypotheses to repurpose safe drugs off patent in novel indications
  • twoXAR: developed AI-based algorithms trained on billions of real world biomedical data points including gene expression measurements, protein interaction networks, and clinical records. Screened 25,000 drug candidates for liver cancer – and narrowed down to 1 – now in clinical trials…in only 4 months
  • Lantern Pharma – a CGI partner – more below…

AI algorithms can see through the millions of studies and publications, thousands of clinical trials, millions of parameters of patient medical records, while keeping all of the patient’s specifications and mutations in mind. IBM’s Watson for Oncology has an advanced ability to analyze the meaning and context of structured and unstructured data in clinical notes and reports that may be critical to selecting a treatment pathway. Then by combining attributes from the patient’s file with clinical expertise, external research and data, the program identifies potential treatment plans for a patient.


CGI’s partnership with Mendel Health:

CGI’s recent partnership with Mendel Health (Mendel.ai) will enable AI in precision medicine to drive personalized treatment and accelerate clinical trial matching for cancer care.  Mendel Health is a breakthrough artificial intelligence company focused on using the Mendel.ai deep learning engine to accelerate and constantly update clinical trial matching for patients.

According to the National Cancer Institute less than 5% of patients with cancer participate in clinical trials.  Research!America, the largest not-for-profit public education and advocacy alliance committed to making research improve health, found in a national survey of over 1,000 patients that less than 7 of 10 health care providers talk about clinical trials or research, while only 54% of patients believe that the care they are receiving is based on the best or most recent advancements from research.

CGI and Mendel will be actively working together to integrate Mendel.ai capabilities for clinicians, oncologists, pathologists, as well as cancer centers and hospitals with CGI’s disease-focused reports and testing results. CGI and Mendel will be among the first in laboratory medicine to provide real time, constant clinical trial matching that integrates biomarker and genomic data with electronic health records and clinical observations.


CGI’s collaboration with Lantern Pharma:

CGI recently announced a collaboration with Lantern Pharma, a highly innovative precision therapeutics biopharma company exclusively focused on accelerating the drug approval process utilizing AI and big data. The collaboration is focused on drug rescue and repurposing – a high-growth segment expected to significantly reduce development costs and contribute to as much as 25 to 30 percent of new therapeutic approvals [2].

The collaboration will leverage CGI’s biomarker discovery, clinical trial testing, and companion diagnostic development capabilities along with Lantern’s big data and AI-driven development approach. The initial focus of the collaboration will be discovery services and further development of Lantern’s lead compounds, including the clinical-stage drug candidate Tavocept (LP-300) for non-small cell lung cancer [3].


Parting thoughts…

What’s next?

We believe AI is going to transform the medical sector beyond recognition and will entail changes to every facet of modern health care, including full adoption of electronic medical records, FDA procedures and classifications, easy access to DNA sequencing, etc. Success hinges on a multidisciplinary approach that combines strong AI expertise, laboratory / technical skills, and present day health care ecosystem cognizance. And, by leveraging AI-enabled solutions, we believe such adoption will help drive the CGI value proposition as a critical and preferred partner in oncology diagnostics.

Stay tuned. This is the first in a series of AI-oriented editorials. We plan to cover further specific developments and more details about CGI’s AI-driven collaborations and approaches in future Shared Insights.


  1. Accenture Analysis. https://www.accenture.com/us-en/insight-artificial-intelligence-healthcare (Accessed on 6-13-2017)
  2. Persidis, A. Myths and Realities of Repositioning. Arrowhead 4th Annual Drug Repositioning, Repurposing & Rescue Conference. Chicago, IL, USA. May 27-28th, 2015
  3. Lantern Pharma Product Pipeline. http://www.lanternpharma.com/products-pipeline.php (Accessed on 6-13-2017)

Comments

comments