AI Unlocks Secrets of Autoimmune Disease Progression

FEATURED RESEARCH

AI Unlocks Secrets of Autoimmune Disease Progression

Autoimmune diseases, like rheumatoid arthritis (RA) and lupus, often begin with a preclinical stage characterized by silent biomarkers and mild symptoms that may linger for years.

For instance, antibodies for RA can be detected up to five years before symptoms appear. Around 8% of Americans live with autoimmune diseases, and most are women.

Identifying progression risks early could significantly improve outcomes.

A smarter, more accurate model: Researchers from Penn State College of Medicine developed the Genetic Progression Score (GPS) to predict disease progression more precisely.

By integrating genetic data from biobanks and case-control studies, GPS improves prediction accuracy by 25% to 1,000% compared to older models. This method leverages AI techniques like transfer learning to adapt models from related tasks, even with smaller datasets.

Real-world impact: The GPS model outperformed 20 other approaches, predicting RA and lupus progression using real-world data from the Vanderbilt University biobank and validating results with the NIH’s All of Us biobank.

High GPS scores consistently indicated higher risks of progression. This enables targeted monitoring, early interventions, and personalized treatments, minimizing irreversible damage.

Beyond prediction, GPS could streamline clinical trial recruitment, focusing on patients most likely to benefit from therapies. This research opens doors to better care, not just for autoimmune conditions but potentially for other diseases too.

For more details: Full Article

Brain Booster

Select the right answer! (See explanation below)

Opinion and Perspectives

DRUG DEVELOPMENT

AI Challenges the $2 Billion Drug Development Problem

The promise of AI in drug development is hard to ignore. With drugs taking 10-15 years and $1-2 billion to reach the market, any tool that saves time and money grabs attention. But is AI the solution everyone hopes for?

What AI can do: AI has been applied across the drug development pipeline—identifying targets, screening molecules, predicting toxicity, and optimizing clinical trials. Between 2010 and 2022, 20 AI-focused startups discovered 158 drug candidates. Impressively, 15 of these reached clinical trials, with some completing preclinical testing in just 30 months, compared to the typical 3-6 years.

The challenge: Despite these successes, AI hasn’t addressed the core issue: the 90% failure rate of drugs in clinical trials. Unlike fields like image recognition, drug development struggles with small, low-quality datasets. Subtle differences in a molecule’s structure can drastically change its behavior in the body, making predictions difficult.

The real opportunity: AI has the potential to tackle key failure points: dosage, safety, and efficacy. Using features like binding specificity, drug concentration in tissues, and structural properties, machine learning models could better predict a drug's chances of success. Testing these features in ultra-low-dose trials could cut costs and refine candidate selection.

Here at AI in Lab Coat, we’re also very interested in the potential success of AI-developed drugs. We’ll be tracking to see which ones will be a commercial success and how they stack up against the traditional drug development cycle. Look for it soon!

For more details: Full Article

Top Funded Startups

Exclusive Funding Report

The One Healthcare AI Deal That Defined December’s Funding Trends

The end of the year tends to quiet things down, and December was no exception for healthcare AI. Without the few mega deals, the funding pool would have looked slim.

Access Our Report: Full Review

Product Pipeline

ALZHEIMER’S TREATMENT

The AI-Driven Drug Giving Hope to Alzheimer’s Patients

IGC Pharma’s CALMA trial focuses on IGC-AD1, a cannabinoid-based therapy targeting agitation in Alzheimer’s patients. Agitation, marked by restlessness, aggression, and emotional instability, accelerates cognitive decline and places significant strain on caregivers.

IGC-AD1 was developed using AI to identify potential drug candidates with anti-neuroinflammatory properties, enabling a more targeted approach to treatment.

By advancing this innovative therapy, IGC Pharma aims to provide a fast-acting solution that addresses a critical unmet need in Alzheimer’s care, potentially improving outcomes for millions of patients and their caregivers.

Completion of the trial is anticipated in 2025.

For more details: Full Article

Policy and Ethics

AI DRUG DEVELOPMENT

FDA Sets the Gold Standard for AI in Drug Development

The FDA’s new guidance takes an important step in making sure AI models used in drug development are reliable and trustworthy. A key focus is on clearly defining how an AI model will be used, so its safety and performance can be properly assessed.

With over 500 AI-related submissions reviewed since 2016, the FDA stresses early conversations with developers to tackle challenges like bias, transparency, and reliability.

This effort aims to strike a balance—supporting innovation while protecting patient safety. The agency is inviting public feedback to help shape these recommendations before finalizing them.

For more details: Full Article

Byte-Sized Break

📢 Three Things AI Did This Week

• Getty Images and Shutterstock announced a $3.7 billion merger to create a stock-image powerhouse aimed at addressing AI-driven industry challenges. [Link]

• Novo Nordisk expanded its Valo Health partnership, committing $190M upfront and up to $4.6B overall to apply AI to 20 programs targeting obesity and Type 2 diabetes. [Link]

• SportsLine AI has analyzed the 2025 NFL Wild Card Weekend odds, providing predictions for game outcomes, including against-the-spread, over-under, and money-line picks. [Link]

Have a Great Weekend!

Trivia Answer: B) 25%

Around 23-25% of people have typically given up on their resolutions by the second week of January. The struggle is real, as making lasting changes often clashes with old habits and daily routines. If you’re still going strong, give yourself a pat on the back—you’re already ahead of many!