GoCART

25 clinical trials. 296 patients. No approved therapy. That is the current state of CAR-T in AML, summarized in a systematic review published in May 2025 in Current Oncology. The authors analyzed 25 CAR-T clinical trials involving 296 patients with relapsed or refractory AML, focusing on safety and efficacy outcomes. The most frequently targeted antigens were CD33, CD123, and CLL-1. Responses were heterogeneous and often short-lived when not consolidated with allogeneic stem cell transplantation. Three findings stand out. First, prolonged and severe myelosuppression was a frequent limiting toxicity, often requiring allogeneic stem cell transplantation to restore hematopoiesis. In other words, the therapy itself damaged the bone marrow to a degree that required a transplant to recover, because the targets it hit were not exclusive to leukemic cells. Second, disease progression remained the leading cause of death. Responses existed, but they did not last. Without a way to maintain selective pressure on the leukemic population without destroying healthy tissue, durable remission remains out of reach. Third, the antigens that dominated these trials, CD33, CD123, CLL-1 are the same antigens expressed on healthy hematopoietic stem cells. The toxicity was not incidental. It was structurally built into the targeting logic. This is the problem GoCART is designed to solve. By requiring two antigens to be present on the same cell before activation occurs, our AND-gate architecture can target AML cells while leaving healthy hematopoietic stem cells, which carry only one of the two signals - intact. The goal is not to avoid myelosuppression through dose reduction. It is to build selectivity into the receptor itself. 296 patients and no approved therapy is not a failure of effort. It is a signal that the current architecture has reached its limit. 🔗 https://lnkd.in/dhA-VZ-r #AML #CellTherapy #CART #Immunotherapy #Hematology #Oncology #Biotech #GoCART

AML is not a uniform mass of identical cells. It is a hierarchical disease. At the top of this hierarchy sits a rare, self-renewing population called leukemic stem cells - LSCs - which generate and continuously replenish the entire leukemic clone. This distinction has direct consequences for therapy. Most treatments for AML - chemotherapy, targeted agents, and current CAR-T approaches eliminate the bulk of leukemic blasts: the rapidly dividing cells that make up the visible disease. Patients achieve remission. But if LSCs survive, the disease is rebuilt from the top down. Relapse follows, often with resistance to the original treatment. LSCs are difficult to eliminate for two reasons. First, they divide slowly, which makes them resistant to therapies that work by targeting actively dividing cells. Second, their surface markers overlap significantly with healthy blood stem cells, the population responsible for producing all blood cells throughout life. A therapy aggressive enough to kill LSCs risks destroying the healthy bone marrow needed for recovery. This is why 5-year survival in AML remains below 20% in most adult patients. Remission is achievable. Long-term cure requires eliminating LSCs without damaging healthy hematopoiesis, and no approved therapy has reliably achieved both. For CAR-T design, this means targeting bulk disease is not enough. An effective therapy must reach LSCs, and must be precise enough to spare the healthy stem cells that share the same surface markers. This is the problem GoCART's AND-gate platform is built to address, using dual-antigen recognition to distinguish leukemic stem cells from their healthy counterparts. 🔗 https://lnkd.in/dVZQ3Zkx #AML #LeukemicStemCells #CellTherapy #CART #Immunotherapy #Hematology #Oncology #GoCART

Why does CAR-T work in B-cell lymphoma but not in AML? In B-cell malignancies, the math is straightforward. CD19 is expressed on essentially every malignant B-cell and critically, it is expendable. Eliminating all CD19-positive cells causes B-cell aplasia, which is clinically manageable. The therapy can be maximally aggressive because the collateral damage is tolerable. AML presents a fundamentally different problem on both counts. First, there is no expendable antigen. Take three of the most validated AML targets: CD33 is expressed on healthy myeloid progenitors throughout the bone marrow. CD38 is present on hematopoietic stem cells, NK cells, and plasma cells. CD56 is found on peripheral neurons, skeletal muscle, and NK cells. A therapy potent enough to eradicate the leukemia will also destroy the healthy tissue that shares these markers. This is not a dosing problem. It is a targeting logic problem. Second, AML is biologically heterogeneous in a way that B-cell lymphomas are not. Different leukemic subclones within the same patient can express different antigenic profiles and under therapeutic pressure, AML cells can downregulate or lose target antigen expression entirely, rendering single-antigen therapies blind. Solving AML requires a targeting logic that does not rely on finding a single exclusive antigen, but instead reads a combinatorial signature unique to the malignant cell, for example, requiring both CD33 and CD56 to be present on the same cell before activation occurs. Healthy tissue carries one signal. Tumor cells carry both. This is the problem GoCART is engineered to address. #AML #CellTherapy #CART #Immunotherapy #Hematology #Oncology #Biotech #GoCART

$5.9B → $16.1B by 2030. CAR-T is no longer a niche therapy. It's becoming a standard of care. The global CAR-T market is projected to expand from $5.9 billion in 2025 to $16.1 billion by 2030. But the more important story is what's driving that growth, and where the gaps still are. All currently approved CAR-T products target just two antigens - CD19 and BCMA. The indications they cover share one thing: a target antigen specific enough to hit safely with a single input. AML doesn't have that antigen. Not because the targets don't exist - CD33, CD123, CLL-1 are all well-validated, but because none of them are exclusive enough to tumor cells to use alone without damaging healthy bone marrow. The field's next step is already clear: move from single-antigen logic to combinatorial recognition. Require two signals instead of one. Make the therapy activate only when both markers appear on the same cell. This is the architecture GoCART is building, and the same direction the broader field is moving toward. #CellTherapy #CART #AML #Immunotherapy #Oncology #Biotech #GoCART

CD56 is one of the most potent cancer targets we have. It's also one of the most dangerous to use. Here's why. CD56 is highly expressed on tumor cells in Multiple Myeloma, AML, and NK/T-cell lymphomas. In many patients, it marks the malignant population with striking consistency. On paper, it looks like an ideal CAR-T target. In practice, it has been nearly impossible to use safely. The problem: CD56 is not tumor-exclusive. It is also expressed on natural killer (NK) cells, on peripheral neurons, and on skeletal muscle tissue. A conventional single-antigen CAR-T therapy directed at CD56 would not distinguish between a myeloma cell and a healthy neuron. The result, severe neurotoxicity and systemic NK cell depletion, has kept CD56 off-limits at therapeutic affinity levels for years. This is what "undruggable" actually means in cell therapy. Not that the target is unimportant, but that the collateral damage of hitting it exceeds any clinical benefit. The only way to unlock CD56 is to change the logic of activation. CAR-T cell requires CD56 and a second tumor-associated antigen to be present on the same cell before firing. This is the principle behind GoCART's AND-gate approach. GoCART pairs CD56 with CD33 in AML and CD38 in Multiple Myeloma. Both antigens must be present on the same cell for activation to occur. Healthy tissue carries one signal. Tumor cells carry both. A powerful target becomes a precise one, not by weakening the binder, but by requiring two keys instead of one. 💡 The antigen was never the problem. The logic was. #CellTherapy #CART #Immunotherapy #MultipleMyeloma #AML #Oncology #Biotech #GoCART

Ocaña-Cara et al. (Experimental Hematology & Oncology, Nov 2025) published one of the most comprehensive reviews of dual-antigen CAR-T strategies in AML to date. The core finding is sobering: every major antigen being tested in AML - CD33, CD123, CLL-1, TIM-3 is also expressed on healthy hematopoietic stem cells. Single-input therapies face an impossible trade-off: increase the dose to kill the tumor, and you risk destroying the bone marrow needed for recovery. The review draws a clear line between two logic strategies: → OR-gate: activates if either antigen is present. More potent, but higher toxicity risk, useful as a bridge to transplant, not as a standalone curative approach. → AND-gate: activates only when both antigens are present on the same cell. Lower collateral risk and, critically, the only architecture the authors identify as a realistic path to a transplant-independent therapy in AML. The catch? AND-gate success depends entirely on antigen pair selection, both targets must be broadly co-expressed on leukemic stem cells, not just blasts. This is exactly the problem GoCART is built to solve. Our ABC Trimer platform uses extracellular logic to enforce dual-antigen coincidence detection, while our in-silico pipeline is designed to identify the antigen pairs with the highest co-expression on LSCs, before a single experiment runs. The science is converging on the same conclusion: the next step in AML is not a better binder. It is smarter logic.

Acute Myeloid Leukemia remains one of the most difficult cancers for targeted therapy, not because of a lack of potential targets, but because no single antigen appears sufficient to achieve both efficacy and safety. 📌 Most antigens currently under investigation in AML, including CD33, CD123, FLT3, and CLL-1, are not exclusively expressed on leukemic cells. They are also present, to varying degrees, on healthy hematopoietic tissues. As a result, therapies directed against a single antigen may eliminate malignant cells, but also risk damaging the normal bone marrow required for hematopoietic recovery. This challenge is further compounded by the marked biological heterogeneity of AML. Different patients, and even distinct leukemic subclones within the same patient, may express different antigenic profiles. Under therapeutic pressure, AML cells may also downregulate or lose expression of a targeted antigen, creating an additional route to resistance. The future of AML treatment may therefore depend less on identifying a single ideal antigen and more on developing therapeutic strategies capable of recognizing combinations of signals that are uniquely associated with leukemic cells. In this context, further progress in AML is likely to require not simply more potent therapies, but more selective and biologically informed approaches capable of distinguishing malignant from healthy tissue with greater precision. 👀 This principle underlies the strategy pursued by GoCART, which is developing multi-antigen recognition systems designed to activate only in the presence of signal combinations associated with leukemic cells. #AML #Biotech #CART #CancerImmunotherapy #CellTherapy #Immunotherapy #Biotech #Hematology #Oncology #Leukemia #GoCART https://lnkd.in/diQ9pBrd

I spent the first week of March in San Diego at the Festival of Biologics. The energy at the conference was a reminder that while the biotech market has been difficult, the optimism for new modalities is very much alive. The consensus among the teams I spoke with is that we are reaching a ceiling with current cell therapies. We have proven we can kill cancer cells, but we are still struggling to do it without damaging healthy tissue. Most of the technical conversations kept returning to the same problem: precision. At GoCART, our approach is to move beyond simple binding and toward actual biological logic. Our platform uses a physical AND-gate to ensure therapy only activates when two specific markers are present on a tumor. This spatial coincidence sensing is what allows us to target indications like Multiple Myeloma and AML that were previously considered too risky for this type of treatment. Leaving San Diego, it is clear that the future of this field belongs to those who can solve the toxicity bottleneck. We are focused on building the operating system that makes that precision possible. I am looking forward to the next few months as we move our current validation work toward the clinic.

Breaking the Toxicity Ceiling in Solid Tumors The recent review by Rosa et al. (2026) in Med clarifies why CAR-T therapies often struggle outside of blood cancers. The primary culprit is "on-target, off-tumor" (OTOT) toxicity. When a target antigen is expressed—even at low levels—on vital organs like the lungs or liver, the treatment can be as dangerous as the disease. Most current solutions try to "tune down" the sensitivity of the cell, but this often leads to the tumor escaping the therapy entirely. At GoCART Therapeutics, we are building a different solution. Our physical AND-gate ensures that the therapy only activates when two specific markers are present on the same tumor cell. This spatial coincidence sensing allows us to target indications that were previously off-limits, providing a "safety-by-design" bridge to the clinic. Precision isn't just a goal; it's the operational standard required to move these therapies into solid tumors. https://lnkd.in/g6vqp3ic #Biotech #Immunotherapy #Oncology #DeepTech #GoCART

Check out BuildWithShreyas podcast with our CEO and Co-Founder Henry Erdlei! Have a great weekend you all! #Biofounders #CarTcells #Immunotherapy