Predictive Oncology Inc is blazing a path as a company that can operate through the fullness of the R&D spectrum of pharma and speed up new cancer treatments.
The Minneapolis, Minnesota-based company shared initial results in December 2021, showing that its flagship drug discovery platform powered by both artificial intelligence (AI) and large patient datasets, can predict cancer drug responses with great accuracy.
“The company’s Patient–centric Drug Discovery by Active Learning (PeDAL) platform, can make what is referred to as ‘high-confidence predictions of drug response’ and therefore improve treatment paths for cancer,” Predictive Oncology CEO J Melville Engle told Proactive.
“This enables a more informed selection of drug/tumor type combinations to increase the probability of success during drug development. Our goal is to ultimately help researchers develop better drugs that can work toward treating cancer patients, like those with ovarian cancer,” he added.
A biotech and healthcare industry visionary, Engle joined Predictive Oncology’s board in 2016, was appointed Predictive Oncology chairman in January 2020 and became CEO in March 2021. He has extensive experience in turning companies around and growing sales. As the former CEO of Dey LP, a division of Merck & Company Inc, Engle drove sales during his tenure from $250 million to over $600 million.
Discovery 21 proof-of-concept
Engle has steered Predictive Oncology to achieving a key milestone towards monetizing its unique approach to leveraging AI in oncology drug discovery.
In March this year, the firm’s Discovery 21 proof-of-concept campaign demonstrated its PeDAL platform’s ability to make accurate predictions of drug responses on ovarian tumor samples. Discovery 21 tested PeDAL’s ability to make drug response predictions, starting with an experimental space of 175 US Food and Drug Administration (FDA) approved cancer drugs and tumor samples from 130 ovarian cancers.
“The results are validated and show that drug response predictions have an accuracy of 91.8%,” said Engle. “This is statistically significant because the PeDAL platform can make high-confidence predictions of drug response enabling a more informed selection of drug/tumor type combinations to increase the probability of technical success and could lower the cost and time during development.”
The Discovery 21 team has developed a customer-facing portal to allow partners to analyze project data and outcomes in real-time. Significantly, Predictive Oncology now has a database with drug response results for 137 types of tumors and more than 15 years of tumor drug response data spanning more than 150,000 clinical cases.
“PeDAL provides an AI machine learning platform to drive experimental testing (coupled with tumor assay capabilities) to evaluate hundreds of diverse tumor samples against hundreds of drug compounds early in drug discovery,” pointed out Engle.
Predictive Oncology says that by selecting key experiments, PeDAL’s AI can make “high-confidence predictions” without testing the entire experimental space – meaning every drug with every tumor sample – thus saving valuable time and costs for the pharma industry. It inevitably helps companies select better drug compounds to move into clinical development.
Road to commercialization
Analysts at HC Wainwright have a 'Buy' rating on Predictive Oncology and $5 price target, noting that the company’s AI platform has demonstrated “promising predictive capability." Shares of Predictive Oncology recently traded at $0.71 on the tech-dominated Nasdaq.
“We look for the company to sign contracts of meaningful economic value in 2022,” said the analysts in a recent note to clients. “Look for the signing of an initial contract with an oncology-focused company to be a positive catalyst in 1H22.”
Predictive Oncology’s boss noted that the company is “actively seeking strategic partners” to add PeDAL into their oncology drug discovery programs.
“One of our goals is to provide customers with the understanding of how PeDAL’s technology delivers an AI-driven model that utilizes 150,000+ patient-unique profiles of drug responses and biomarkers,” said Engle. “This brings the patient into the heart of drug discovery and offers answers that other companies can’t.”
Stable revenue stream
In addition, Predictive Oncology is anchoring the generation of a stable stream of revenue by becoming a provider of soluble and stable formulations for proteins, including vaccines, antibodies, large and small proteins, and protein complexes.
“By establishing a Good Manufacturing Practices (GMP) facility, Predictive Oncology is able to build a cohesive set of products and services to become an end-to-end formulation and manufacturing capability to support the supply of large molecules to the drug discovery and development industry through Phase 1 clinical trials,” said Engle. “It offers solutions for customers in the discovery and development of biological products for a variety of cancers and other chronic diseases.”
Engle noted that the firm has incorporated its AI capabilities in formulation products and services to “predict optimal additive combinations” that result in “soluble and physically stable drug products,” thus optimizing the formulation process during drug development.
The zPREDICTA advantage
Engle believes that to understand how therapeutic agents behave in a patient, researchers need to use models that mimic actual human tissue. zPREDICTA’s technology is as close as it gets. Therefore, in November 2021, Predictive Oncology acquired zPREDICTA, a leading provider of tumor-specific 3D cancer cell models for in vitro testing of anti-cancer therapeutics.
The cell culture models that the company produces are disease and tissue-specific making them much better at predictive analytics than those provided by 2D testing or non-disease-specific 3D models, noted the company.
“Predictive Oncology acquired zPREDICTA to expand its product offering to 3D oncology drug discovery programs, especially with pharmaceutical and biotech companies,” said Engle.
The zPREDICTA product offers the capability to produce 3D models and generate increasing revenue. According to company projections, each new cancer project will earn approximately $1.3 million in annual revenue after one year of development at a moderate cost.
The company has had success with Multiple Myeloma cell growth, and with human 3D models of Acute Myeloid Leukemia (AML), a type of cancer of the blood and bone marrow.
Precision medicine
Precision medicine, precisely targeting drugs based on the genomic profile of the patient’s tumor, has become the goal for cancer therapy. However, there are so many mutations in a patient’s tumor, that most are not actionable with current drugs, said the company. So, the race has begun to generate more data to understand which mutations are linked to what drug.
“Predictive Oncology is moving in the direction of becoming a precision medicine company,” said Engle.
By leveraging its historical database of the drug responses of over 150,000 patient tumors and building data-driven predictive models of tumor drug response, Predictive Oncology can potentially provide “actionable insights” critical to both new drug development and individualizing patient treatment.
High growth potential
The analysts at HC Wainwright, who have a $5 price target on Predictive Oncology, have stated that the company’s potential for “accelerated revenue growth” is sorely "underappreciated."
“We believe AI models POAI generates have potential to improve treatment paths for ovarian cancer and drive the development of new therapies,” said the analysts.
The knowledge-driven company is set to be a first-mover in the artificial intelligence-powered drug discovery market that will grow to $20 billion in the next three years.
“Predictive Oncology is a compelling story,” noted Engle. “Our drug discovery products reduce the timeframe and increase the speed of the drug discovery process, increase the likelihood of drug efficacy by efficiently addressing tumor heterogeneity, and improve the diversity of the drug portfolio against a given cancer.”
Ultimately, for every drug that makes it to market, pharma companies spend nearly $2.8 billion just testing drugs that may never see the light of day. Significantly, Predictive Oncology offers a potentially swifter and economically viable development pathway.
“Drug development is a long and costly process. We seek investors that understand this timeline and can navigate the promise that AI brings to the process and patients,” concluded Engle. “Predictive Oncology is an investment in leading-edge technology which improves the cancer drug discovery and development process.”
Contact the author Uttara Choudhury at uttara@proactiveinvestors.com
Follow her on Twitter: @UttaraProactive