CAR-T Therapy's First Decade: What Long-Term Follow-Up in Pediatric Leukemia Shows

What This Article Covers

In 2012, a six-year-old girl named Emily Whitehead became the first pediatric patient ever treated with a CAR-T cell therapy, after standard treatment and two relapses had left her with no other options. Five years later, that experimental therapy — developed at Children's Hospital of Philadelphia (CHOP) and the University of Pennsylvania and commercialized as tisagenlecleucel (Kymriah) — became the first CAR-T product ever approved by the FDA. Nearly a decade and a half on, the field finally has what it needed most: real long-term follow-up data, tracking the original trial patients for five, six, and now approaching ten years. This article walks through what that follow-up has actually shown about durable remission, late relapse, and long-term safety — and what it means for how oncologists now talk about CAR-T as a potentially curative treatment for children and young adults with relapsed or refractory B-cell acute lymphoblastic leukemia (ALL).

A Therapy Born From a Single Child's Case

Tisagenlecleucel's story begins earlier than its 2017 approval. Emily Whitehead's CD19-directed CAR-T infusion in April 2012, developed by Stephan Grupp, Carl June, and colleagues at CHOP and Penn under the name CTL019, was a last resort for a child whose leukemia had relapsed twice. She achieved remission — and, remarkably, has stayed cancer-free ever since. In a 2022 CHOP retrospective marking her ten-year mark, Grupp put it plainly: "Now we're 10 years out... I believe they are [cured]." That single case seeded the multicenter ELIANA trial, whose results led the FDA to approve tisagenlecleucel in August 2017 as the first CAR-T therapy — and the first gene therapy of any kind — approved in the United States, for patients up to age 25 with B-cell ALL that is refractory or has relapsed at least twice.

What the Original Trial Showed

The pivotal ELIANA trial enrolled 75 pediatric and young adult patients with relapsed or refractory B-cell ALL, most of whom had already failed multiple prior lines of therapy, including in many cases a stem cell transplant. Within three months, 81% achieved remission (60% complete remission, 21% with incomplete blood count recovery) — a striking result in a population where standard salvage chemotherapy typically produces poor and short-lived responses. At one year, 76% of patients were still alive and 50% remained relapse-free, according to the trial's original 2018 New England Journal of Medicine publication. Cytokine release syndrome (CRS) occurred in 77% of patients, severe in 45%, and neurologic side effects occurred in 40% — real, serious risks that required intensive supportive care, including ICU admission in nearly half of patients. These numbers were encouraging, but they only covered the first year. The real question — the one that matters most to a family deciding whether this is a cure or a reprieve — was what would happen five, six, ten years out.

The Long View: What Years of Follow-Up Have Now Shown

That question now has real answers. A 2025 long-term analysis published in the Journal of Clinical Oncology followed the ELIANA cohort for a median of roughly 79 months — about six and a half years, with some patients followed past seven and a half years. Among the 70 patients who achieved a complete response, five-year relapse-free survival landed in the range of 47% to 51%, depending on how later therapies were accounted for, and five-year overall survival was 55% to 62%. Median overall survival was not reached at all — meaning more than half of the original responders were still alive well beyond the five-year mark, a milestone that essentially never happened for this heavily pretreated population before CAR-T existed. Earlier interim analyses told a consistent, encouraging story building toward this: by the trial's roughly five-year checkpoint, five-year relapse-free survival among responders was reported around 44%, with overall survival around 55%. In other words, the curves didn't collapse the way relapse curves so often do after other salvage therapies in this disease — they flattened, and stayed flat. For oncologists who have spent careers watching relapsed pediatric ALL come back within a year or two, a plateau that holds past five years is the signature of something durable, not just a longer pause before the same outcome.

Late Relapses Do Happen — And They Follow a Pattern

None of this is a claim that CAR-T guarantees a cure for everyone who receives it, and the long-term data are honest about that. Relapse remains the leading cause of treatment failure in this population; among patients who recovered normal B cells (a signal that the CAR-T cells had stopped patrolling for leukemia), the cumulative incidence of relapse reached roughly 40%. Most relapses cluster in the first one to two years, which is also when loss of CAR-T cell persistence — measurable as the return of normal B cells — tends to occur. Later relapses are less common but not unheard of, which is exactly why oncologists now monitor B-cell counts as an early warning sign and, in some cases, use it to guide decisions about consolidative stem cell transplant or re-treatment. In the 2025 long-term analysis, the deaths recorded after the one-year mark were attributed to relapse rather than to treatment toxicity — a pattern that reinforces where the residual risk actually sits.

Long-Term Safety: Reassuring, With One Real Caveat

On safety, the extended follow-up brought good news: no new or unexpected late toxicities emerged as patients were tracked further out, and the adverse events reported in year two and beyond were dominated by ordinary infections rather than anything specific to the CAR-T cells themselves. That said, the field as a whole has had to reckon with one serious, class-wide safety signal. In April 2024, the FDA required a boxed warning — its strongest label warning — on all six approved CD19- and BCMA-directed CAR-T products, tisagenlecleucel included, after reports of secondary T-cell malignancies, some fatal, emerging as early as weeks after infusion across the product class. The agency called for lifelong monitoring of all CAR-T recipients as a result. It's important to represent this accurately rather than either minimizing or overstating it: this is a real, FDA-mandated risk that deserves lifelong vigilance, but in the tisagenlecleucel pediatric ALL long-term follow-up data specifically, only one secondary malignancy was documented among the ELIANA cohort, and it was not one of the CAR-positive T-cell malignancies driving the class warning. Independent reviews have also noted that background secondary-malignancy rates in heavily pretreated leukemia survivors — a population that has already received intensive chemotherapy, and often prior transplant — are not trivial to begin with, which complicates attributing any single case to the CAR-T cells versus prior treatment. The honest summary: the risk is real and the monitoring requirement is appropriate, but it has not, in this specific dataset, translated into a wave of secondary cancers overtaking the survival benefit.

What This Changes About How the Field Talks About CAR-T

A decade ago, "durable remission" in relapsed pediatric ALL meant a year, maybe two. The accumulated ELIANA follow-up — now stretching past six years of median observation, with individual patients like Emily Whitehead now over a decade out — has allowed oncologists to use a word they rarely used lightly in this disease before: cure. Not for every patient; roughly half of the original responders in the long-term data have experienced relapse or later therapy, and that is a real, significant limitation that shouldn't be softened. But for the other half, the survival curves have behaved the way cure curves behave — flat, not declining — for long enough that specialists are increasingly comfortable describing durable CAR-T remission as a genuine plateau rather than a delay. That shift has downstream effects: it is changing how families weigh CAR-T against consolidative transplant, how survivorship clinics plan monitoring, and how the next generation of pediatric CAR-T trials are designed and measured.

Bottom Line

Nearly a decade after tisagenlecleucel's FDA approval, long-term follow-up of the original pediatric and young adult ALL trial population shows something the field could only hope for in 2017: durable remissions that hold for more than half of patients past the five-year mark, with a safety profile that has not worsened with time, aside from a serious but so-far rare class-wide secondary malignancy risk that now requires lifelong monitoring. Relapse — often in the first one to two years — remains the primary way this therapy fails, and roughly half of treated patients do not reach that durable plateau. But for the children and young adults who do, the data increasingly support what Emily Whitehead's own physician was willing to say out loud after ten years: for a meaningful share of patients who had exhausted every other option, this looks like a cure.