What If You Could Kill Cancer Before It Even Starts?
A tiny trial at Dana-Farber just showed that a single CAR-T infusion wiped out detectable disease in every patient with high-risk smoldering myeloma, and none have progressed to cancer. It's only 20 patients, but the implications are massive.
Imagine going to your doctor for a blood test. The results come back weird, not cancer exactly, but something that could become cancer. Your doctor says you have two options: watch and wait, or try a single treatment that might stop it from ever becoming a problem.
That's the choice facing patients with high-risk smoldering multiple myeloma (SMM), a precursor condition that sits in an uncomfortable gray zone between "you're fine" and "you have cancer." And new data from a small but striking trial suggest that a one-time infusion of CAR-T cells could wipe the slate clean.
The Ticking Clock Nobody Wants
Smoldering myeloma is like a check engine light that won't turn off. It's not full-blown cancer yet. But for patients classified as high-risk, the odds of progressing to active multiple myeloma are brutal: roughly 50% within two years and 65–70% within five years.
For decades, the standard approach has been "active surveillance." Translation: come back every few months, get your blood drawn, and hope things don't get worse. It's a bit like watching a slow leak in your ceiling and deciding to just keep checking on it instead of fixing the roof.
Some doctors have started offering early treatment with drugs like lenalidomide. Europe recently approved daratumumab for high-risk SMM, and the FDA is moving that direction too. But none of these approaches have produced the kind of jaw-dropping responses that just came out of Dana-Farber Cancer Institute.
A Trial Called CAR-PRISM
Dr. Omar Nadeem presented the results at AACR 2026, and the numbers are hard to ignore.
The trial, called CAR-PRISM, enrolled 20 patients with high-risk smoldering myeloma. Each received a single infusion of cilta-cel (sold as Carvykti), a CAR-T therapy that targets a protein called BCMA on the surface of plasma cells. CAR-T, for the uninitiated, works by extracting a patient's own immune cells, genetically reprogramming them to hunt cancer, and infusing them back in. Think of it as giving your immune system a software update with very specific instructions.
Get tomorrow's biotech intelligence before your competitors.
Join thousands of biotech professionals who start their day with our free, daily briefing.
The trial design was deliberately aggressive in its simplicity. No induction chemotherapy beforehand. No bridging therapy. Just lymphodepletion (a short course of chemo to make room for the new cells), then a single CAR-T infusion on Day 1. One shot.
The investigators also offered stem cell collection for transplant-eligible patients, just in case myeloma showed up later. A safety net, essentially.
The Results That Turned Heads
Every single patient responded. All 20 hit MRD negativity at the deepest level tested (10⁻⁶ by next-generation sequencing). MRD stands for minimal residual disease; it's the most sensitive way to check whether cancer cells are still lurking. At that sensitivity level, you're looking for one cancer cell among a million normal ones.
Patients reached MRD-negative status fast, too. Median time: about one month. By month two, all patients were MRD-negative at the deepest threshold.
The clinical responses were equally impressive. Ninety percent of patients achieved a complete or stringent complete response, the highest bar for remission. Among those followed for at least six months, that number hit 100%.
At a median follow-up of about 15.3 months, zero patients had progressed to active myeloma. No one had died. Six patients followed for over 18 months remained in complete response with sustained MRD negativity.
For a disease where half the patients would typically progress within two years under observation, that's a striking contrast.
Safety: Better Than Expected, But Not Without Wrinkles
Safety was the primary endpoint, which makes sense for a trial that's essentially asking: is it responsible to use this powerful therapy in people who don't technically have cancer yet?
The short answer: toxicity was manageable and milder than what doctors typically see when CAR-T is used in heavily pretreated, late-stage patients. All cases of cytokine release syndrome (CRS), the inflammatory reaction that can accompany CAR-T expansion, were low-grade. No patient experienced grade 3 or higher CRS.
Neurotoxicity told a more nuanced story. Seven patients developed non-ICANS neurologic events: four had facial nerve palsy that resolved completely, and three had mild motor symptoms with residual but improving deficits. No high-grade ICANS (the specific neurotoxicity syndrome associated with CAR-T) was reported. Transient blood count drops, including grade 3–4 neutropenia, were common but expected.
No dose-limiting toxicities were observed. But with only 20 patients and roughly 15 months of follow-up, the long-term safety picture (late relapses, infections, secondary cancers) remains an open question.
The Elephant in the Waiting Room
This is where enthusiasm needs to meet reality. A few important caveats:
The sample size is tiny. Twenty patients at a single academic center is proof-of-concept territory, not practice-changing evidence. These were carefully selected high-risk smoldering patients at Dana-Farber; whether the results generalize to broader populations is unknown.
Follow-up is short. Fifteen months sounds good, but myeloma is a disease measured in decades for younger patients. We don't know if these responses will hold at three, five, or ten years. Late relapses are a real concern.
Cost is staggering. CAR-T therapy runs $500,000 to $1 million per patient when you factor in the drug, hospitalization, and complication management. The cell product alone costs around $400,000. Using this in asymptomatic patients who might never progress (even high-risk SMM patients have roughly a 50% chance of not developing myeloma at five years) raises thorny cost-effectiveness questions.
The FDA boxed warning looms. In January 2024, regulators added warnings about the risk of secondary T-cell cancers to all CAR-T products. Giving a therapy with that label to someone who doesn't have cancer yet is a tough sell without rock-solid long-term data.
Why This Still Matters Enormously
Despite the caveats, CAR-PRISM represents something genuinely new. This is the first trial to deploy CAR-T in a precursor cancer setting, and the results are unlike anything seen before in smoldering myeloma. Universal MRD negativity at this depth, with no progressions, from a single infusion with no prior therapy? Nothing else has come close.
The conceptual leap is the real story. CAR-T was born as a last resort, the thing you tried when everything else failed. Then it moved to second-line therapy. Now researchers are asking whether it belongs at the very beginning, before cancer even fully arrives. It's like using a missile defense system to shoot down threats while they're still on the launchpad.
A Bigger Pipeline Is Building
CAR-PRISM isn't happening in isolation. An entire wave of trials is pushing CAR-T earlier in the myeloma timeline:
CARTITUDE-6 is a phase 3 trial directly comparing cilta-cel against autologous stem cell transplant in newly diagnosed, transplant-eligible myeloma patients. If cilta-cel wins, it could replace transplant as the standard consolidation approach.
CARTITUDE-5 is testing cilta-cel versus standard maintenance therapy in newly diagnosed patients who aren't candidates for transplant.
KarMMa-4 is exploring ide-cel (the other approved BCMA CAR-T, sold as Abecma) in high-risk newly diagnosed myeloma after just a few cycles of induction therapy.
The pattern is clear: the field is systematically testing whether CAR-T works better when disease burden is lower and the immune system is less battered by prior treatments.
The Bottom Line
CAR-PRISM hasn't proven that CAR-T can prevent myeloma. What it has shown is that a single infusion can erase detectable disease in every patient tested, with a safety profile that looks manageable. That's a remarkable starting point.
The road from 20 patients at one hospital to a new standard of care is long and full of potholes. We need randomized trials, years of follow-up, solutions to the cost problem, and honest conversations about treating asymptomatic people with powerful therapies.
But for the first time, the question isn't "can we treat myeloma better?" It's "can we stop it from ever showing up?" And the early answer, cautiously but unmistakably, is maybe.
5 min read
Genentech's $490 Million Shopping Spree Has a Dark Side
Genentech cut 103 research scientists the same week it signed a $490 million breast cancer deal with Astex Pharmaceuticals. It's the starkest example yet of big pharma's new playbook: shrink internally, shop externally.
