Trends in Applied AI and Precision Oncology

Source: allinonemovie/Pixabay

For over a century, scientists have tried to use the body’s own immune system to combat diseases. Today, there are FDA-approved drugs that achieve this endeavor due to innovation in precision medicine by pioneering scientists, and visionary leaders across disciplines in artificial intelligence, genomic, bioinformatics, cloud-computing, health care, biotechnology, and pharmaceutical industries. At this week’s American Society of Clinical Oncology (ASCO) Annual Meeting 2019 in Chicago, a number of companies are sharing how they are applying artificial intelligence (AI) to precision oncology.

Precision Medicine and Precision Oncology

Precision medicine uses the patient’s protein and genetic data to prevent, diagnose and treat diseases. An example of precision medicine is immunotherapy, where the body’s own immune system is activated to defend against disease.

Precision oncology is precision medicine to prevent, diagnose, and treat cancer. There are 15 cancers that can be successfully treated by immunotherapy, more than 1,000 immunotherapy clinical trials currently ongoing in the U.S., and more than 70 immunotherapy drugs in the clinical pipeline, according to Johns Hopkins Medicine.

At this week's ASCO 2019, Boston-based Concerto HealthAI, backed by SymphonyAI, presented a series of research abstracts demonstrating the use of AI machine learning and natural language processing to perform a retrospective analysis of 12,000 de-identified patient records in the CancerLinQ Discovery® dataset. In April 2019, Concerto HealthAI announced a collaboration with Pfizer and a month earlier in March, Concerto HealthAI signed a multi-year deal with Bristol-Myers Squibb.

Immune checkpoint inhibitors

T-cells, a type of white blood cell, have receptors that activate an immune system defense when it recognizes a foreign microorganism such as a bacteria or virus. T-cell accelerators are proteins that can activate a full-blown immune response. Other proteins may inhibit an immune response, and function as brakes on T-cells, such as T-cell proteins CTLA-4 and PD-1. An example of immunotherapy is a class of drugs called immune checkpoint inhibitors. Immune checkpoint inhibitors block the proteins that keep the T-cells from killing cancer cells.

From the Nobel Prize to Saving Lives

The 2018 Nobel Prize in Physiology or Medicine was awarded jointly to James P. Allison, PhD, a professor at the University of Texas MD Anderson Cancer Center, and Tasuku Honjo, MD, a professor at Kyoto University, for the discovery of cancer treatment by inhibiting negative immune regulation. Allison used antibodies against CTLA-4 to block the functioning of that brake so that the body’s immune system would attack cancer cells. Honjo used a PD-1 blockade to unleash the body’s immune system to attack cancer.

The first FDA approval for a cancer drug based on the tumor’s genetic change, regardless of cancer type, was in 2017 for the immunotherapy PD-1 inhibitor drug from Merck & Co. called Keytruda (pembrolizumab)—for both adults and children with advanced solid tumors with specific genetic changes of either a mismatch repair deficiency (dMMR) or high microsatellite instability (MSI-H). Prior to 2017, Keytruda was FDA approved for various cancer types: advanced melanoma in 2014, advanced non-small cell lung cancer in 2015, recurrent or metastatic head and neck squamous cell carcinoma in 2016, and metastatic non-small cell lung cancer in 2016. Former U.S. President Jimmy Carter was treated with Keytruda, a checkpoint inhibitor, in 2015.

Another PD-1 inhibitor drug approved last year is Opdivo (nivolumab) by Bristol-Myers Squibb Company Inc. to treat metastatic small cell lung cancer (SCLC) with “progression after platinum-based chemotherapy and at least one other line of therapy.”

PD-1 inhibitor Libtayo by Regeneron Pharmaceuticals, Inc. was granted FDA approval in 2018 for “metastatic cutaneous squamous cell carcinoma (CSCC) or locally advanced CSCC who are not candidates for curative surgery or curative radiation.”

Record-Setting Personalized Medicine FDA Approvals

In 2018 a record-setting 42 percent of all new drug approvals by the FDA were personalized medicine approvals, according to a recent Personalized Medicine Coalition (PMC) report. It was also a landmark year for precision medicine as the FDA approved a second cancer drug based on biomarker rather than tumor type, Vitrakvi (larotrectinib) according to the same report.

“This new site-agnostic oncology therapy isn’t specific to a cancer arising in a particular body organ, such as breast or colon cancer,” stated Scott Gottlieb, M.D., the 23rd Commissioner of Food and Drug Administration during the period 2017-2019, in a 2018 FDA news report. “Its approval reflects advances in the use of biomarkers to guide drug development and the more targeted delivery of medicine. Its approval reflects advances in the use of biomarkers to guide drug development and the more targeted delivery of medicine. We now have the ability to make sure that the right patients get the right treatment at the right time. This type of drug development program, which enrolled patients with different tumors but a common gene mutation, wouldn’t have been possible a decade ago because we knew a lot less about such cancer mutations. Using our breakthrough therapy designation and accelerated approval processes, we support innovation in precision oncology drug development and the evolution of more targeted and effective treatments for cancer patients. This is especially true when it comes to pediatric cancers.”

Artificial Intelligence Essential Reads

AI Performs at Human Levels on Theory of Mind Tests

AI Innovation: Who or What Gets the Credit?

AI in Oncology Immunotherapy Research

Artificial intelligence is being used in immunotherapy research for PD-1 inhibitors. For example, on June 2, 2019 at ASCO, Lunit Inc. presented a study using AI deep learning to develop an AI score to analyze scanned images of immune-tumor microenviroments in order to assess the predictive value of AI versus PD-L1. Lunit Inc. is an award-winning venture-backed AI software company founded in 2013 based in Seoul, South Korea, that provides advanced medical imaging analytics and imaging biomarkers. In 2017 Lunit Inc. was named as one of the world’s top 100 AI startups by CB Insights and one of the world’s top five “Social Impact” AI startups by NVIDIA.

Precision medicine is making significant inroads lately. With the convergence of applied genomics, artificial intelligence and biomedicine, new treatments are emerging that are more targeted towards fighting cancer, offering hope for a better future through innovative technologies.

Copyright © 2019 Cami Rosso All rights reserved.