Innovations in Drug Development

Stacie Calad-Thomson started her talk on the impact of #AI in #drugdevelopment, highlighting that while AI provides frameworks and insights, it's not a silver bullet. AI is a game-changer, significantly expediting the drug discovery process. A journey from traditional drug discovery to AI-driven applications: The Problems: Traditional drug discovery challenges, including a high failure rate, data silos, complexity of biology, and inefficient design-make-test cycles. The process takes a decade and costs $billions. 1- Target Identification: Companies like Recursion use lab #automation and multi-omics (#invivomics, #phenomics, #metabolomics, #proteomics, #transcriptomics, #genomics) data to map biological relationships and run 2 Million experiments weekly. They leverage NVIDIA GPUs and recently secured a $50M investment, and acquired Cyclica & Valence Labs. By integrating AI, bridge the gap between #Biotech and #Techbio. 2- HIT Screening: Open source tools like AlphaFolio have revolutionized drug discovery by predicting 3D protein structures, enabling rapid in silico screening and precise target design. 3- Lead ID: BenevolentAI generative molecular design and active learning facilitate the rapid identification of potent drug candidates, such as #Percipinib, Eli Lilly and Company, for COVID treatments. 4- Lead Optimization: Exscientia combines generative molecular design with active learning for multi-parameter optimization, streamlining drug development. 5- Preclinical: Exscientia's AI-driven platform improves cancer treatment and outcomes. They achieved remarkable results, including a two-year remission for a chemotherapy-intolerant patient at a fraction of CAR-T costs. 6- Clinical Trials: Predicting disease severity and patient stratification for #COVID-19 clinical trials. As CSO at BioSymetrics, Stacie outlined their platform's capabilities, promising phenomics-driven hit discovery in less than a year, with a timeline covering gene-disease drivers, in-vivo modeling, hit identification, and target identification. Stacie contributes to responsible and ethical AI in healthcare as a board The Alliance for Artificial Intelligence in Healthcare (AAIH), collaborating with regulators. AI is undoubtedly transforming drug development, and Stacie's insights shed light on its immense potential. Following her talk, the panel moderated by Anjali Pandey, SVP Sudo Biosciences, Frazier Life Sciences, engaged panelists on a similar topic. Nitin Kumar, CEO, Nuron.IO, Sachin Sontakke, Sr. Dir Gilead Sciences, Preetha Ram, CTO Pier 70 Ventures. The consensus was that AI has and will change healthcare and how patients are cared for.  Since 2019, AI drug discovery start-ups had 352 deals and raised $10B from 600 unique investors. 80% of this $10B was invested in the top 30 companies. The tech-first is the most appealing to investors. The investment gap remains in manufacturing. Big thanks Anurag Mairal, PhD (He/His), Ashutosh Shastry & Pushkar Hingwe

𝐀𝐈 𝐢𝐬 𝐜𝐨𝐥𝐥𝐚𝐩𝐬𝐢𝐧𝐠 𝐭𝐢𝐦𝐞𝐥𝐢𝐧𝐞𝐬 𝐢𝐧 𝐦𝐞𝐝𝐢𝐜𝐢𝐧𝐞, 𝐚𝐧𝐝 𝐫𝐞𝐰𝐫𝐢𝐭𝐢𝐧𝐠 𝐭𝐡𝐞 𝐞𝐜𝐨𝐧𝐨𝐦𝐢𝐜𝐬 𝐨𝐟 𝐜𝐮𝐫𝐞𝐬 A few years ago, predicting a protein’s structure took months or even years. Then came AlphaFold, DeepMind’s Nobel-winning breakthrough, unlocking the ability to understand proteins and disease mechanisms at scale and speed. This led to Alphabet’s spin-off, 𝐈𝐬𝐨𝐦𝐨𝐫𝐩𝐡𝐢𝐜 𝐋𝐚𝐛𝐬, now using AI to design therapies with $600M in funding and its first human trials underway for cancer and immune disorders. ▫️ The speed of progress: Every six months, AI advances like a full human year. What once took decades now unfolds in quarters. ▫️ The cost collapse: AI is driving the cost of drug discovery and testing so low that researchers can now explore thousands of drug candidates and disease targets at once, including treatments that would have been dismissed as too niche or unprofitable just a few years ago. ▫️ The scale of exploration: AI has already helped identify or repurpose over 3,000 drugs currently in clinical trials. We’re already seeing the results. Northwestern researchers, for example, used AI-enhanced screening to repurpose 𝐩𝐢𝐩𝐞𝐫𝐚𝐜𝐢𝐥𝐥𝐢𝐧, a decades-old FDA-approved antibiotic, for Lyme disease. In mouse studies, it cured infection at one-hundredth the dose of standard treatment without harming gut microbiota. That breakthrough emerged in days, not years, at a fraction of traditional costs. 🔺 𝐖𝐞 𝐚𝐫𝐞 𝐰𝐢𝐭𝐧𝐞𝐬𝐬𝐢𝐧𝐠 𝐚 𝐟𝐮𝐧𝐝𝐚𝐦𝐞𝐧𝐭𝐚𝐥 𝐬𝐡𝐢𝐟𝐭. AI isn’t just speeding up medicine, it’s enabling exploration and validation at scales and price points previously unthinkable. And every six months, the landscape tilts even further. This is what leaders need to see: the weight of the opportunity, and the urgency to align their thinking to the pace of this change. #AI #Healthcare #DrugDiscovery #GenAI #FutureOfWork #mindsetchange Forbes Technology Council Gartner Peer Experiences InsightJam.com PEX Network Theia Institute VOCAL Council IgniteGTM IA FORUM 𝗡𝗼𝘁𝗶𝗰𝗲: The views within any of my posts, or newsletters are not those of my employer or the employers of any contributing experts. 𝗟𝗶𝗸𝗲 👍 this? feel free to reshare, repost, and join the conversation!

Last month, a team of scientists and physicians achieved something extraordinary: they developed and delivered the first-ever personalized #CRISPR therapy to treat an infant with a life-threatening #raredisease — in just six months. A one-letter change in the baby’s DNA was corrected using a custom-built gene editor. The child, who was once facing the prospect of a liver transplant, is now steadily improving. It’s a powerful example of what’s becoming possible at the intersection of #science and #technology, urgency and purposeful ambition. And this isn’t an isolated win. Across labs, clinics, and companies, CRISPR is being used as a therapeutic modality to correct inherited disorders, engineer immune cells, disable viral DNA, and even edit entire chromosomes. New gene-editing systems—like TIGR-Tas, unveiled earlier this year—are expanding what’s possible in tissues or conditions where current tools fall short. Clinical results are emerging fast—and the pace of #innovation is only picking up. At Recursion, we’re also applying #geneediting tools like CRISPR beyond therapeutics—using the technology as a tool to better understand #biology at scale. By systematically “knocking out” thousands of individual genes and measuring how those changes affect cell behavior, we’re generating large, structured datasets that feed directly into #AI models. This is helping us uncover new biological relationships and power #drugdiscovery in ways that were previously unimaginable. What ties all of this together is a commitment to applying game-changing #innovation in service of real, urgent human needs. It signals a much-needed mindset shift in #healthcare and #biopharma: to move faster, think bigger, and tackle challenges once considered out of reach—and to truly deliver on the promise of #precisionmedicine. And we’re seeing this ambition in many other areas as well – just last week, for example, GRAIL announced more promising than ever performance stats for its #Galleri blood test for the early detection of 50+ types of #cancer. There’s still work ahead to ensure breakthroughs translate into broad, equitable impact. But this moment – this momentum – is worth pausing to recognize. We’re no longer just imagining a future where science works smarter and faster for patients. We’re building it.

AI is transforming drug development - accelerating discovery, refining biologics, and boosting efficiency across the pipeline. A recent Forbes article by Dr. Sai Balasubramanian, M.D., J.D., captured that momentum perfectly. But speed alone doesn’t define progress. Direction does. That direction comes from real-world evidence (RWE): decades of clinical research and patient journeys that reveal how diseases evolve and therapies perform in practice. When AI is trained on RWE, it stops guessing and starts understanding - turning predictions into precision. Very often, genomics powers that transformation. Curated variants, gene-disease associations, and functional annotations aren’t just background data - they’re the fuel and engine of precision medicine. But even with the best engine, progress demands a skilled driver. That’s where human expertise comes in. AI can process data at scale - but only humans can ask the right questions, shape the structure of evidence, and draw out what matters most for clinical impact. To truly advance medicine, we need all three: AI as the engine, RWE as the fuel, and human expertise at the wheel. Because in healthcare, progress isn’t about how fast you go - it’s how safely you get there. #AIinHealthcare #DrugDevelopment #PrecisionMedicine #Genomics #RealWorldEvidence #ClinicalResearch

The drug therapeutic landscape expands almost daily. Pairing them with multi-omics is a no brainer! That isn't rocket science. But what IS rocket science is developing the tests that make sure those therapies work, and continue to work, when administered to patients! We do have some pretty good experience with this sort of thing already though. Technologies such as pharmacogenetic testing have emerged to help predict how patients will respond to certain classes of drugs. This is done by looking at genetic markers that indicate how quickly someone might metabolize, absorb, or eliminate a drug! But we also have experience developing ‘companion diagnostics.’ These are diagnostic tests that are used to place patients on a specific therapy. The first of these was used in 1998! HercepTest was introduced to identify patients who would respond best to Herceptin, an early antibody treatment for breast cancer. This test was necessary because Herceptin only worked in patients if their tumors overexpressed the HER2 receptor! And because of this integration with a biomarker test, Herceptin is often referred to as the poster child for precision medicine! But we’ve come a long way since Herceptin, and there are some really cool new precision therapeutics on the horizon: PROteolysis TArgeting Chimeras (PROTACs) - Small molecule drugs that have one end that binds to a target protein and another that binds to E3 Ubiquitin Ligase (a protein that marks other proteins for destruction!) Antibody Drug Conjugates (ADCs) - These are antibodies that are physically bound to drugs to make their delivery more targeted. The biggest successes here have been in targeting chemotherapy drugs to tumors! Translation Activating RNAs (taRNAs) - RNA molecules designed to bind to a target RNA to supercharge its translation. This is done by adding a sequence called an Internal Ribosome Entry Site (IRES). These boost ribosome binding on the target RNA and increases production of the target protein. mRNA Vaccines - We're all familiar with these, but what you might not know is that they can also be quickly programmed to create personalized cancer treatments. Chimeric Antigen Receptor - T cells and Macrophages (CAR-T/M) - Are immune cells that have been programmed or personalized to seek out and destroy tumor cells. These are all very exciting, but their development requires a lot of testing to tailor each treatment to an individual. But what excites me the most in this space is getting the opportunity to move beyond the single biomarker tests of old! Because seeing the full picture of a patient's response to a therapy through expanded proteomic and metabolomic screening could: 1) Show us how well a drug is working 2) Signal when someone will relapse 3) Mitigate side effects before they're felt 4) Indicate when to change therapies This is the version of precision medicine that we were promised and I’m hopeful we see these applied more broadly in the clinic soon!

📈Emerging New Drug Modality💡 ⚗️Peptide-drug conjugates (PDCs)🧪 Improve cancer treatment options by offering targeted therapy that enhances efficacy while minimizing systemic toxicity. Key benefits include: 1. **Tumor-Specific Targeting**: PDCs use homing peptides to selectively bind to overexpressed receptors or tumor-specific antigens, ensuring precise drug delivery to cancer cells while sparing healthy tissues. 2. **Reduced Side Effects**: By directly targeting cancer cells, PDCs minimize off-target toxicity, addressing the severe side effects of conventional chemotherapy. 3. **Improved Tumor Penetration**: Peptides are smaller (2–20 kDa) than antibodies, allowing better tissue diffusion and penetration into tumors. 4. **Versatility in Targeting Mechanisms**: PDCs can employ receptor-dependent or receptor-independent strategies, enabling treatment of tumors with heterogeneous or low receptor expression. 5. **Enhanced Payload Delivery**: PDCs can deliver highly potent cytotoxic agents that are otherwise too toxic for standalone use, transforming undruggable compounds into precision therapeutics. 6. **Modular Design**: The combination of homing peptides, linkers, and payloads allows customization for specific cancer types, improving therapeutic outcomes. 7. **Reduced Clearance and Improved Stability**: Innovations in linker chemistry ensure payload stability during circulation and efficient release at the tumor site, overcoming challenges like premature cleavage and rapid renal clearance. 8. **Potential for Resistant Cancers**: PDCs can address unmet needs in oncology by targeting resistant or refractory cancers, as demonstrated by candidates like CBX-12. These advancements position PDCs as a promising frontier in precision oncology, complementing or surpassing current therapies like antibody-drug conjugates (ADCs). https://lnkd.in/g_4Rm47G

This paper from Arxiv titled Large Language Models in Drug Discovery and Development: From Disease Mechanisms to Clinical Trials discusses the integration of Large Language Models (LLMs) into drug discovery and development Key points include: 1. Drug Development Timeline and Cost: - Typically takes 10-15 years - Costs over $2 billion to bring a new drug to market 2. Two Main LLM Paradigms: - Specialized LLMs: Trained on specific scientific languages (e.g., SMILES for molecular structures, FASTA for protein/DNA/RNA sequences) - General-purpose LLMs: Trained on diverse textual information, including scientific papers and textbooks 3. LLM Applications in Drug Discovery and Development: - Understanding Disease Mechanisms - Drug Discovery - Clinical Trials 4. Understanding Disease Mechanisms: - Analyze genomic data, perform RNA analysis, and conduct pathway analysis -Help in clinical subtyping, target-disease linkage analysis, and target validation -DNA-BERT, Nucleotide Transformer, HyenaDNA 5. Drug Discovery: -LLMs can predict protein structures, design and optimize drug molecules, and predict drug-target interactions -Assist in automating chemistry experiments and retrosynthetic planning -Aid in drug-drug interactions and predicting ADMET properties 6. Clinical Trials: -LLMs can assist in patient-trial matching, trial design, and outcome prediction -Analyze electronic health records and clinical protocols -Aid in document writing and regulatory compliance 7. Maturity of LLM Applications: - Categorized as not applicable, nascent, advanced, or mature - Literature analysis are quite advanced, Automated experimentation in early stage 8. Notable LLM Achievements: - Med-PaLM: First LLM to reach human expert level in USMLE-styled questions - Geneformer: Pretrained on 30 million single-cell transcriptomes 9. Future Directions : - Developing more biologically-focused LLMs - Addressing ethical concerns and privacy issues - Overcoming technical limitations like hallucinations and context window limits - Improving model interpretability and scientific understanding 10. Challenges: - Need for better integration of domain knowledge - Handling of sensitive health data - Ensuring reliability of LLM outputs for critical medical decisions This paper discuss the two main paradigms of LLMs: specialized models trained on specific scientific languages, and general-purpose models trained on diverse textual information. The paper evaluates the current state of LLM applications in drug discovery, categorizing them based on their level of advancement and address ethical concerns, limit technical limitations, and improve model interpretability. #Largelanguagemodels #drugdiscovery #clinicaltrials #pdllms #diseasemechanisms #biolllm##proteinstructureprediction #retrosyntheticplanning #patientrialmatching #geneticvariant #targetdiseaselinkage #protocolanalysis #moleculeoptimization Source: www.arxiv.org Disclaimer: The opinion are mine and not of employer's

Google unveils AI-powered healthcare innovations spanning drug discovery, enhanced search, and integrated medical records: 💊In drug discovery, new open AI models (TxGemma) are designed to understand both text and molecular structures to help predict the safety and efficacy of potential therapies 💊An AI co-scientist tool built on Gemini 2.0 assists biomedical researchers by parsing scientific literature, generating novel hypotheses, and proposing experimental approaches 💊These tools will be available through the Health AI Developer Foundations program, aiming to streamline the early stages of drug development 🔎 In search, expanded health knowledge panels now cover thousands more topics and use AI to provide quick, credible answers to health-related queries 🔎 The "What People Suggest" feature aggregates user discussions from online platforms to offer personalized insights based on shared experiences with specific health conditions 🔎 These enhancements support multiple languages, including Spanish, Portuguese, and Japanese, and are initially rolling out on mobile devices in the U.S. 💿The global launch of Medical Records APIs for the Health Connect platform on Android enables apps to read and write standardized medical data, such as allergies, medications, immunizations, and lab results 💿The APIs support over 50 data types, integrating everyday health tracking with official medical records from healthcare providers 👇Links to source articles in comments #DigitalHealth #AI #Google