MagBodies: Combining the Best of Surgery and Drugs

Cancer treatment has always faced a fundamental tension: surgery can precisely target a region of the body, but it can't chase down every last cell; drugs can reach the whole body, but they can't distinguish tumor from healthy tissue well enough to avoid collateral damage.

The history of blood cancers illustrates just how much that distinction matters. For decades, blood cancers were among the hardest to treat — there was no tumor to cut out. Then researchers discovered that you could wipe out an entire population of B cells and the body would simply regenerate them. Rituximab, a monoclonal antibody targeting CD20 on B cells, was approved by the FDA in 1997 and became instrumental in treating B-cell lymphomas and leukemias.¹ That permissiveness — the fact that B cells are expendable and renewable — turned blood cancers from a worst-case scenario into one of oncology's bigger success stories, with CAR-T cell therapies and bispecific antibodies continuing to improve outcomes.² Solid tumors haven't been so lucky. Finding a molecular target that is unique to tumor cells, consistently expressed, present across enough patients, and otherwise well-behaved has proven extraordinarily difficult. If tumor cells looked obviously different from healthy tissue, our own immune systems would have caught them first.

Surgery, meanwhile, remains the frontline treatment for the solid tumors that make up roughly 90% of adult cancers.³ For all the sophistication of modern drug development, the first-line approach to most solid tumors is still: find the region where the cancer lives and cut it out. About 60% of cancer patients undergo some form of surgery during treatment.⁴ Drugs typically play a supporting role — mopping up residual cells before, during, or after surgery.⁵ When drugs fail, they tend to fail in one of two ways: they lack clinical efficacy (accounting for 40–50% of clinical trial failures) or they cause unmanageable toxicity (roughly 30% of failures).⁶ And when surgery fails, it's usually because the disease has spread to more places than a surgeon can safely reach — metastatic disease is responsible for at least two-thirds of deaths from solid tumors.⁷

This is the problem MagBodies are designed to solve. A MagBody is, in essence, a logic gate — an antibody that can be switched between a potent and an inert state by the placement of external magnets. This combines the regional specificity of surgery with the molecular specificity of a drug. Place a magnet over the tumor site and the antibody activates locally; remove the magnet and it goes quiet. Unlike photodynamic therapy, which is limited by the penetration depth of light — typically only a few millimeters in tissue⁸ — or focused ultrasound, which struggles to transmit through bone structures like the skull and offers only transient, moderate localization⁹, magnetic fields can reach any depth in the body safely and continuously — even at home.

MagBodies also introduce a capability that neither surgery nor conventional drugs can offer: local masking. Rather than activating a drug at the tumor, you can suppress it at the body's most sensitive organs while letting it fight metastatic disease everywhere else. And because magnet placement and timing can be adjusted after the drug is already in a patient, MagBodies are effectively retunable in the clinic — an almost unheard-of property for a therapeutic. The most punishing cost in drug development is failure in clinical trials: the process takes over a decade and costs on the order of $1–2 billion per approved drug, with approximately 90% of candidates failing after entering clinical studies.⁶ In oncology specifically, the attrition rate exceeds 95%.¹⁰ As Ruxandra Teslo has argued, clinical trials are not merely a yes/no gate but the critical feedback loop of drug development — and when a trial fails, the informational value of that failure is largely wasted because redesigning the molecule means starting the entire regulatory and manufacturing cycle over.¹¹ Companies often simply die after a first failed trial. A drug whose performance parameters can be refined through external hardware, much like a dosing regimen, compresses that feedback loop dramatically: instead of scrapping the molecule and recapitalizing a new entity, you adjust the magnet protocol and try again.

Finally, in an era where antibody patents are increasingly hard to defend against copycat biologics — accelerated by manufacturing scale in China and emerging AI-driven protein design — genuine mechanistic innovations like MagBodies offer something commodity antibodies cannot: defensibility rooted in a fundamentally new approach, not just a new sequence.

The core insight is simple. Decades of oncology research have been spent searching for the perfect molecular target — one that is unique to cancer and absent from healthy tissue. MagBodies suggest a different path: use good-enough targets and control where they activate. It's the difference between finding a better bullet and learning to aim.