Servier Symphogen’s Post

Over the past year, artificial intelligence has moved from a promising buzzword to a daily clinical enabler in #oncology trials. #AI models are now redefining how imaging-based endpoints are set, allowing sponsors and CROs to assess real-time patient response, stratify participants more safely, and ultimately lower trial operational costs, all while maintaining rigorous scientific standards. This transformation was on full display at #ASCO2025, where leading studies demonstrated how AI systems are being deployed for complex endpoints, such as HER2 assessment and therapy response prediction, particularly in breast and gastrointestinal cancers. These systems excel not just in accuracy, but in speed and scalability, helping global trials adapt to evolving protocols. What once lived in innovation labs is now routinely used to support adaptive trial designs, in silico screening, and earlier go/no-go decisions for new drugs. The practical impact? Sponsors are seeing improved trial efficiency, better protocol adherence, and higher data quality. Importantly, these AI-driven approaches are clearing regulatory hurdles, setting new benchmarks for harmonized imaging data and integrated analyses. Their adoption ensures that #imaging endpoints aren’t just surrogate markers, but predictive, actionable measures that support robust, patient-centric oncology research. The present and the future of oncology trials are being shaped by AI-powered imaging, moving us decisively past hype to transformative real-world results. Collaboration is key to successful #clinicaltrials. Partner with IAG to leverage our imaging expertise and optimize your clinical operations for better patient outcomes. Explore our services at www.ia-grp.com #ClinicalPartnerships #ImagingSolutions Source: https://lnkd.in/esU9jJPc

Bummed that I couldn't go to #ESMO2025... Other than a trip to Europe, our Picture Health team has presented ground breaking work with our QVT radiomics biomarker the past two years. Measuring cancer tumor blood vasculature is a game changer! 🚀 Avastin (bevacizumab), the first VEGF anti-angiogenic, was approved nearly 20 years ago to decrease 🩸blood flow feeding tumors. Why did it work in only some patients? Because only some patients have complex vasculature but until now, we couldn't measure that. With QVT, we can! Non-invasively, no new lab equipment, not shipping samples to an outside lab, not destroying biopsy tissue, not waiting weeks for turnaround time, just a standard radiology scan with results in minutes. 🎗️Imagine how that can change how an oncologist selects the best treatment for each patient. 💉Imagine how much it could help accelerate clinical trials. 👀 Follow Picture Health for more over the next few weeks https://lnkd.in/gD9qquqZ

📢 𝐖𝐡𝐚𝐭 𝐢𝐟 𝐬𝐭𝐚𝐧𝐝𝐚𝐫𝐝 𝐢𝐦𝐚𝐠𝐢𝐧𝐠 𝐜𝐨𝐮𝐥𝐝 𝐩𝐫𝐞𝐝𝐢𝐜𝐭 𝐭𝐫𝐞𝐚𝐭𝐦𝐞𝐧𝐭 𝐨𝐮𝐭𝐜𝐨𝐦𝐞𝐬 𝐰𝐢𝐭𝐡𝐢𝐧 8 𝐰𝐞𝐞𝐤𝐬? At Picture Health, we can. At #ESMO2025, we’re presenting a novel AI imaging biomarker, validated in the 𝐏𝐡𝐚𝐬𝐞 𝐈𝐈𝐈 𝐅𝐑𝐄𝐒𝐂𝐎-2 𝐭𝐫𝐢𝐚𝐥 in mCRC. Built with 𝐐𝐕𝐓™ and advanced radiomic features, this signature: ✨ Captures mechanistic changes linked to VEGF inhibition ⏱   Signals survival benefit as early as the first on-treatment scan 🎯  Reflects true treatment-specific benefit - observed only in the treatment arm, not placebo 📩 𝐃𝐢𝐬𝐜𝐨𝐯𝐞𝐫 𝐡𝐨𝐰 𝐐𝐕𝐓 𝐢𝐬 𝐭𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐢𝐧𝐠 𝐜𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐝𝐞𝐜𝐢𝐬𝐢𝐨𝐧-𝐦𝐚𝐤𝐢𝐧𝐠. 𝐂𝐨𝐧𝐧𝐞𝐜𝐭 𝐰𝐢𝐭𝐡 𝐮𝐬 𝐚𝐟𝐭𝐞𝐫 𝐭𝐡𝐞 𝐜𝐨𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞 𝐭𝐨 𝐥𝐞𝐚𝐫𝐧 𝐦𝐨𝐫𝐞 𝐚𝐛𝐨𝐮𝐭 𝐰𝐡𝐚𝐭 𝐐𝐕𝐓 𝐜𝐚𝐧 𝐝𝐨 𝐟𝐨𝐫 𝐲𝐨𝐮.  💡Where AI meets oncology innovation — 𝐟𝐨𝐥𝐥𝐨𝐰 𝐮𝐬 𝐟𝐨𝐫 𝐰𝐡𝐚𝐭’𝐬 𝐧𝐞𝐱𝐭. Huge thanks to our collaborators on this groundbreaking work. Sara Lonardi, Ozlem Yardibi, N. Arvind Dasari., Haojia Li, Varsha S., Lucy C., Kai Zhang, Nathaniel Braman, Matthew McMahon, Chien-Lin (Caroline) Yeh, , Jayant Narang, MD.

AI Just Made Drug Discovery 1,000× Faster! || #ai #scifi #podcast #science . . Imagine curing diseases before they even spread — not in decades, but in days. Ray Kurzweil believes AI will revolutionize drug discovery by simulating billions of molecules, designing potential cures in record time, and testing them on millions of digital patients in virtual clinical trials. That means safer, faster, and possibly more accurate results — 1,000× faster than today’s methods. The question is, which disease deserves to be conquered first — cancer, Alzheimer’s, or something even deadlier?

Tempus AI ($TEM) is reshaping how healthcare decisions get made. After joining the #HTEC and #THNQ indexes in Q3 2024, its stock has climbed ~60% as investors back its AI-driven approach to precision medicine. In Q2 2025, the company nearly doubled revenue, driven by growth in cancer screening (Ambry Genetics), digital pathology (Paige AI), and a $200M data partnership with AstraZeneca—each move sharpening its AI models and competitive edge. 🔗 Rafael Silva Serrano explains how Tempus is turning diagnostics into a durable business and what it means for healthcare investors: https://lnkd.in/ggHgsWfW #AIInvesting #HealthTech #ETFs #ROBO

What if we could design new medicines in a fraction of the time? For decades, a major bottleneck was "seeing" the shape of proteins our drugs need to target. G-protein-coupled receptors (GPCRs) are one such family of proteins. They act as the "gatekeepers" for our cells and are the target for an estimated 40% of all modern drugs. The problem? Their structures are incredibly difficult to map. This is where AI changes the game. DeepMind's AlphaFold can now predict a protein's 3D shape from its genetic sequence with astonishing accuracy—solving a 50-year-old grand challenge in biology. This breakthrough is supercharging drug discovery, allowing scientists to virtually screen millions of potential new medicines and understand "orphan" receptors we previously knew nothing about. I wrote an article breaking down how this revolutionary AI works, why GPCRs are so important, and how this new partnership between AI and biology is set to transform medicine. You can read it here: https://lnkd.in/dHukg8mV

🧵 CRISPR, CAR-T & Radiopharma are converging — and this changes what young founders can build. CRISPR let us rewrite genes — first in rare diseases, now increasingly in cancer. CAR-T showed that immune cells can be reprogrammed like software. Radiopharma is making cancer treatment targeted instead of destructive, delivering radiation only to tumor cells. These used to be separate revolutions. Now? They’re blending into programmable, targeted medicine: CRISPR = Rewrite instructions CAR-T = Reprogram the soldiers Radiopharma = Aim the scope This is the shift from shotgun therapy → sniper therapy. And here’s the door young founders can walk through: You don’t need to invent CRISPR 3.0. You can build the bridges: Data + sequencing interpretation tools Cell-therapy QC + scheduling workflows Radiopharma cold-chain + traceability Patient recruitment + clinical navigation Delivery systems + logistics layers As a student-founder, this clicked for me: Innovation isn’t only at the molecule — it’s at the delivery layer. Takeaway: Curiosity + speed > credentials. The opportunity is in making precision medicine usable at scale. ➡️ Young founders: Which layer are you claiming — Gene Editing / Cell Programming / Precision Targeting / Clinical Ops? Drop your play → let’s compare notes. #Healthtech #PharmaStartups #Biotech #Innovation #StudentFounder #FutureOfHealth #AIinHealthcare #HealthTech #MedTech #FutureOfMedicine #PharmaInnovation #Biotech #CRISPR #CellTherapy #Radiopharma #PharmaStartups #StudentFounder #FutureOfHealth

FDA Expands Libtayo, FDA AI Review Tool Takes Shape The FDA approved an expanded adjuvant use for Libtayo (cemiplimab) in high-risk skin cancer patients, reducing recurrence by 68% Reuters+1. Meanwhile, the FDA is rolling out “Elsa,” a generative AI tool designed to speed scientific reviews and protocol assessment Reuters. These moves reflect dual trends: first, therapeutic expansion beyond oncology into immuno-oncology niches; second, adoption of AI in regulatory operations themselves. For African innovators, there’s a lesson: if regulators begin to internalize AI capacities, we must meet them in that space — by delivering adaptive protocols, structured data, and explainable AI models. At Camair AI and Camair-AI CRO | Point-of-Care & Medical Device Trials we are already deploying AI and digital endpoints. The FDA’s official embrace of generative review systems reinforces our choice: interoperable, transparent, high-fidelity data pipelines will be a differentiator in global trial access. Let’s align our trials with the next era of AI-augmented regulation #FDA #AIinRegulation #ClinicalTrials #HealthTech #CamairAI

Whilst 𝗖𝗵𝗮𝘁𝗚𝗣𝗧 decided to engage the 𝗹𝗼𝘄𝗲𝘀𝘁 𝗰𝗼𝗺𝗺𝗼𝗻 𝗱𝗲𝗻𝗼𝗺𝗶𝗻𝗮𝘁𝗼𝗿 of users, i.e. 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗲 𝗲𝗿𝗼𝘁𝗶𝗰 𝗰𝗼𝗻𝘁𝗲𝗻𝘁. 𝗚𝗼𝗼𝗴𝗹𝗲 𝗚𝗲𝗺𝗶𝗻𝗶 has been applied to definitely less click-bait, but important, 𝘁𝗵𝗲 𝗹𝗼𝗻𝗴𝗲𝘃𝗶𝘁𝘆 𝗼𝗳 𝗵𝘂𝗺𝗮𝗻 𝗹𝗶𝗳𝗲! - Alas, I suspect ChatGPT will illicit far more interest, unless you prove me wrong by liking this. Researchers from 𝗚𝗼𝗼𝗴𝗹𝗲 and 𝗬𝗮𝗹𝗲 𝗨𝗻𝗶𝘃𝗲𝗿𝘀𝗶𝘁𝘆 used a 𝟮𝟳-𝗯𝗶𝗹𝗹𝗶𝗼𝗻-𝗽𝗮𝗿𝗮𝗺𝗲𝘁𝗲𝗿 𝗚𝗲𝗺𝗺𝗮 𝗺𝗼𝗱𝗲𝗹 to 𝗱𝗶𝘀𝗰𝗼𝘃𝗲𝗿 𝗮 𝗻𝗼𝘃𝗲𝗹 𝗰𝗮𝗻𝗰𝗲𝗿 𝗺𝗲𝗰𝗵𝗮𝗻𝗶𝘀𝗺. This system, known as 𝗖𝟮𝗦-𝗦𝗰𝗮𝗹𝗲 𝟮𝟳𝗕, analysed single-cell data and predicted how 𝘀𝗽𝗲𝗰𝗶𝗳𝗶𝗰 𝗱𝗿𝘂𝗴𝘀 𝘄𝗼𝘂𝗹𝗱 𝗶𝗻𝗳𝗹𝘂𝗲𝗻𝗰𝗲 𝘁𝘂𝗺𝗼𝘂𝗿 𝗶𝗻𝘁𝗲𝗿𝗮𝗰𝘁𝗶𝗼𝗻𝘀 𝘄𝗶𝘁𝗵 𝘁𝗵𝗲 𝗶𝗺𝗺𝘂𝗻𝗲 𝘀𝘆𝘀𝘁𝗲𝗺. 𝗞𝗲𝘆 𝗳𝗶𝗻𝗱𝗶𝗻𝗴𝘀 𝗼𝗳 𝘁𝗵𝗲 𝘀𝘁𝘂𝗱𝘆 𝗶𝗻𝗰𝗹𝘂𝗱𝗲:  • 𝗨𝗻𝗲𝘅𝗽𝗲𝗰𝘁𝗲𝗱 𝗟𝗶𝗻𝗸: The model identified 𝘀𝗶𝗹𝗺𝗶𝘁𝗮𝘀𝗲𝗿𝘁𝗶𝗯, a well-known drug, as capable of aiding the immune system in tumour detection only when small amounts of interferon were already present.  • 𝗟𝗮𝗯 𝗖𝗼𝗻𝗳𝗶𝗿𝗺𝗮𝘁𝗶𝗼𝗻: Experiments conducted in living cells validated the prediction, demonstrating approximately a 𝟱𝟬% 𝗶𝗻𝗰𝗿𝗲𝗮𝘀𝗲 𝗶𝗻 𝗶𝗺𝗺𝘂𝗻𝗲 𝘃𝗶𝘀𝗶𝗯𝗶𝗹𝗶𝘁𝘆 when the drug combination was used.  • 𝗥𝗲𝗮𝗹 𝗕𝗶𝗼𝗹𝗼𝗴𝗶𝗰𝗮𝗹 𝗥𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴: Unlike simply repeating existing knowledge, the AI developed a novel testable hypothesis that smaller models had overlooked.  • 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 𝗩𝗮𝗹𝘂𝗲: Teams at Yale are now investigating how this discovery could potentially guide future immunotherapy treatments.

I've been exploring AI more and more recently and I have to say it's become a significant productivity tool for me. I would even say it's like having a PA who is very good at computer skills. AI isn't a new thing, I studied it at university 30 years ago. What's changed is the processing power that allows us to take vast amounts of data to teach AI about a specific subject. What I do find interesting is how we can use AI to improve cancer diagnosis and treatment. This is an absolute game changer and could benefit the human race significantly, possibly making cancer a thing of the past. Interesting article: https://lnkd.in/er4wWPXm People + Technology = better