Insilico Medicine, a pioneer in AI-driven drug discovery, has made a significant breakthrough. The company has successfully predicted the outcomes of Phase II to Phase III clinical trials using its innovative AI tool, inClinico.
Approximately 90% of drug development failures occur at the clinical stage, leading to trillions of dollars in losses and years of wasted effort. To combat this, Insilico has developed inClinico, a generative AI platform designed to predict the outcomes of Phase II clinical trials.
The platform leverages various engines that utilize gen AI and multimodal data, including text, omics, clinical trial design, and small molecule properties. It has been trained on over 55,600 unique Phase II clinical trials from the past seven years.
Insilico's researchers have developed a clinical trial probability model that demonstrated an impressive 79% accuracy when validated against real-world trials in the prospective validation set where measurable outcomes were available.
AI revolutionizing drug development
The research, published in the Clinical Pharmacology and Therapeutics journal, highlights the transformative potential of AI in drug development and investment decision-making.
Insilico revealed that the AI engines used in this study have been integrated into the inClinico system, which is designed to predict clinical trial outcomes. This integration is a crucial part of the Medicine42 clinical trials analysis and planning platform.
“AI provides a significant advantage in processing complex data and identifying patterns,” said Alex Zhavoronkov, founder and CEO of Insilico Medicine. “We used machine learning and AI to build models based on various data points related to successful and failed drugs. These models were then integrated into our prediction engine inClinico. For every evaluated Phase II trial, inClinico generates a probability of success for proceeding to Phase III.”
Zhavoronkov revealed that the validation studies were conducted both internally and in collaboration with pharmaceutical companies and financial institutions, demonstrating the robustness of the inClinico platform. On a quasi-prospective validation dataset, the platform achieved an impressive ROC AUC score of 0.88, a measure of its ability to discriminate between success and failure in clinical trial transitions.
The company claims that the platform’s accurate predictions were tested with a date-stamped virtual trading portfolio, resulting in a 35% return on investment (ROI) over nine months, making it a valuable tool for investors seeking critical technical due diligence insights.
Leveraging generative AI for drug development and discovery
Insilico’s Zhavoronkov revealed that his research group created the initial dataset of Phase II clinical trial data from 55,653 trials sourced from clinicaltrials.gov and various other public sources, including pharma press releases and publications.
This data was meticulously labeled, annotated, and linked together by biomedical experts, a discriminative transformer, and a generative large language model.
A transformer system then mapped these trials to drugs and diseases using a state-of-the-art natural language processing (NLP) pipeline based on the Drug and Disease Interpretation Learning with Biomedical Entity Representation Transformer (DILBERT), which was published at the ECIR 2021 conference.
Zhavoronkov pointed out that the pharma industry has traditionally relied on fundamental academic research and serendipity to generate new ideas and hypotheses. However, the high failure rate indicates that the complexity of diseases and biological mechanisms make it exceedingly challenging to identify successful targets for treating diseases, especially novel targets.
Revealing insights, potential treatments
Zhavoronkov believes that incorporating AI into the analysis of large, diverse datasets can reveal insights about disease mechanisms and potential treatments that may not be evident to humans. PandaOmics is a part of the inClinico and assimilates vast amounts of data from clinical trials, drugs, and disease information to predict the likelihood of success or failure during the Phase II to Phase III transition.
PandaOmics utilizes various data types such as omics data, grants, clinical trials, compounds, and publications to analyze and produce a ranked list of potential targets specific to a disease of interest.
“PandaOmics is a knowledge graph for target identification through which our generative AI platform can find connections between clinical trial success or failure, disease conditions, and drug attributes that might elude human scientists,” Zhavoronkov told VentureBeat. “Using this data, we built our model for predicting the Phase II clinical trial probability of success, defined as the transition of drug-condition pair from Phase II to Phase III.”
Enhanced predictive capabilities
Insilico Medicine has been training inClinico on clinical trials, drugs, and diseases since 2014, said Zhavoronkov, who emphasized that by combining multimodal LLMs and other gen AI technologies, the company has significantly enhanced its predictive capabilities.
As a result, inClinico now serves as a tool to guide companies in directing their research funds and expertise toward programs with the highest likelihood of success while enabling them to capture and utilize valuable information from programs that have faced setbacks.
“The ability of inClinico to predict the successful Phase II to Phase III transition drugs, even without prior information related to the clinical relevance of the drug’s action of disease, validates the generative AI models and their ability to build on existing data to predict outcomes for diseases where fewer data is available,” Zhavoronkov explained. “The more data it has, and the more successful outcomes, the better AI becomes at accurate prediction.”
What’s next for Insilico?
Zhavoronkov expressed strong encouragement regarding the findings, while also acknowledging their basis within a limited dataset. He firmly believes that the system’s sophistication and precision will continuously improve over time, driven by a surge in data and reinforcement, including insights from Insilico’s internal pipeline programs — three of which (for idiopathic pulmonary fibrosis, cancer, and COVID-19) have successfully advanced to clinical trials.
Insilico projects that approximately 20 to 25% of trials can be predictably assessed using the inClinico tool with meaningful accuracy. The company aspires to expand its capabilities further, leveraging new laboratory robotics advancements to predict success rates for combination therapies and facilitate the selection of the most effective combinations for targeted therapies.
“We integrate cutting-edge technological breakthroughs into our platform, incorporating AI-powered robotics, AlphaFold, and quantum computing,” Zhavoronkov explained. “My grand goal is to see this tool deployed extensively because broader usage will drive further improvement. We employ an approach called Reinforcement Learning from Expert Feedback (RLEF), where the tool’s accuracy improves with the insights we receive from analysts using it for predictions. Currently, we can only predict small molecule first-in-class single-agent targeted therapeutics.”
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.
Title: The High Predictive Accuracy of Insilico Medicine’s Generative AI Tool, InClinico, in Clinical Trial Outcomes
Introduction
In recent years, the world of biotechnology has witnessed massive strides in leveraging innovative technological solutions to revolutionize medical research processes. One of the notable players steering transformative advancements in this space is Insilico Medicine. It has made its imprint with the development of a groundbreaking tool known as InClinico. This article aims to explore InClinico's high predictive accuracy in clinical trial outcomes, reshaping the landscape of modern healthcare.
InSilico Medicine’s Generative AI, InClinico
Insilico Medicine, a global leader in end-to-end artificial intelligence for target discovery, small molecule chemistry, and clinical development, has developed a generative artificial intelligence tool, InClinico. This world-class tool is designed to simulate human clinical trials, radically reducing the time and cost traditionally associated with clinical trials.
InClinico's High Predictive Accuracy
One of the primary reasons for the rising prominence of InClinico is its high predictive accuracy. Traditionally, clinical trials are prolonged, costly, and beset with a high rate of failure due to unpredicted responses in humans. InClinico stands out by displaying an impressive degree of accuracy in predicting clinical trial outcomes. This revolutionary tool utilizes sophisticated AI algorithms to effectively model and predict how varying pharmacological agents will interact with the human body.
The performance of InClinico is rooted in its cutting-edge generative adversarial network (GAN) learning structure. This AI framework enables the machine learning model to predict potential medical phenomena based on the vast data inputs from previously conducted studies.
The vital implication of its high predictive accuracy lies not merely in cost and time savings, but significantly, in improving patient safety. In reaching a successful clinical outcome, avoiding potential adverse events is absolutely critical, and InClinico proves instrumental in predicting such occurrences.
Impact on Clinical Trials
The introduction of InClinico to clinical trials has had a revolutionary impact. Its technology bypasses the traditional linear path, allowing for parallel testing of potential therapeutics, which considerably expedites the trial process. Moreover, with its high predictive accuracy, medical researchers can effectively anticipate clinical outcomes, which allows for modification and advancement through the various trial phases.
Simulated clinical trials provided by InClinico’s AI serve not only to predict the outcome of a trial but also help design a more optimized, patient-centered approach. Its ability to synthesize and analyze vast datasets enables the development of precision medicine, as different populations and patients can respond uniquely to the same treatment.
Conclusion
InClinico’s pivotal role in reshaping the clinical trial landscape serves to highlight the transformative impact of AI technology in modern healthcare. The tool’s high accuracy in predicting clinical trial outcomes illustrates the immense potential of AI to not only streamline clinical trials, but also to significantly enhance patient safety. By merging data analysis, innovation, and next-generation technology, Insilico Medicine continues to drive advancements, positioning itself at the forefront of AI-assisted medical research. However, looking to the future, it is essential to continually refine and validate AI tools like InClinico for there to be sustainable, long-term success in this revolutionary era of healthcare.