Severe Aplastic Anemia: The Donor and the Decade Decide the Answer
Two independent patients, the same severe aplastic anemia. The disagreement isn't about the disease — it's about whether a matched donor and three fewer decades of life change which therapy actually serves each of them best.
Tomas M. finished his first full year teaching kindergarten this spring, and had just signed a lease on his own apartment two blocks from the elementary school when a routine physical — required for a summer camp counseling job he'd taken on the side — turned up a white count his primary care doctor didn't like the look of. He mentioned gum bleeding when he flossed and a faint bruise on his shin he couldn't place, both of which he'd dismissed until the labs made them worth asking about directly.
Two weeks and a bone marrow biopsy later, he has severe acquired aplastic anemia: a hypocellular marrow at roughly ten percent cellularity, an absolute neutrophil count under 200, platelets under 10,000, and a reticulocyte count that confirms his marrow simply isn't making cells, not that something is destroying them faster than they're produced. Cytogenetics and PNH clone testing came back negative, and there's no dysplasia on the marrow smear to suggest an early myelodysplastic process masquerading as aplasia.
He has an older brother, healthy, willing, and HLA-matched on typing — identified within days, since their parents had both boys typed years earlier during an unrelated family medical scare. That single fact changes everything about how his case is being framed. At 24, with no organ dysfunction and no transfusion burden yet accumulated, he sits squarely inside the population where allogeneic transplant has historically outperformed immunosuppression, not because immunosuppression fails often, but because a young marrow given a genuinely new, matched graft avoids the clonal evolution risk — myelodysplastic syndrome, later leukemia — that shadows even a successful immunosuppressive response over the following decade.
Case A: A matched donor, no reason to wait
At 24, with a matched sibling ready and no organ damage yet to complicate conditioning, this is close to the ideal transplant candidate. Long-term data comparing matched-sibling transplant against immunosuppression in younger patients consistently favor transplant on overall survival once you follow patients out far enough to capture the clonal evolution risk — MDS, and later AML — that shows up in a meaningful fraction of immunosuppression responders a decade out.
Immunosuppression can work well in the short term, and I'm not disputing that. The argument for transplant here is specifically about the fifteen-year horizon, not next month.
I'd take the same position for a different reason worth stating explicitly: transplant upfront also avoids the alloimmunization risk that comes with any transfusion exposure before transplant, and he hasn't been transfused yet. Every unit of blood he receives before a graft raises graft-rejection risk modestly; starting transplant workup now, before he needs transfusion support to get through the wait, protects the graft he's about to receive.
Immunosuppressive therapy with ATG and cyclosporine, with eltrombopag added per RACE, has narrowed some of the older survival gap between the two approaches — but RACE was run in a broader population than just young matched-sibling candidates, and doesn't erase the case for transplant specifically here.
Agreed: proceed directly to matched-sibling allogeneic transplant workup, holding transfusion support to the minimum needed to get him safely to conditioning, specifically to protect graft outcomes.
Loretta S. retired from the postal service eighteen months ago after thirty years sorting mail at the same regional facility, and had just started volunteering twice a week at her granddaughter's school library when unexplained bruising sent her to urgent care. She'd noticed the bruises appearing without any memory of an injury, and a nosebleed the week before that took unusually long to stop — both, in retrospect, the same picture Tomas's bleeding gums had already shown his own team.
Her workup mirrors his almost exactly — severe aplastic anemia, ANC under 200, platelets under 10,000, a hypocellular marrow without dysplasia, negative cytogenetics and PNH clone testing. What doesn't mirror his case is everything around the diagnosis: she is 58, has no siblings, and an unrelated-donor registry search turned up no close match within the timeframe her counts can tolerate waiting. That timeframe is not open-ended: an absolute neutrophil count under 200 is what makes her severe rather than merely cytopenic, and it is the number that converts a donor search from a background process into a clock. The best available option was a single-antigen mismatch, not the full match Tomas's brother provided instantly.
Age alone would not rule out transplant — older patients are transplanted successfully — but the combination of her age, the absence of a matched donor, and mild baseline hypertension shifts the actual risk-benefit calculation in a way Tomas's case never had to confront. A mismatched unrelated donor transplant in a 58-year-old carries meaningfully higher graft-versus-host disease and transplant-related mortality risk than a matched sibling graft in a 24-year-old, which is precisely the comparison the team has to make honestly rather than treat her case as a delayed version of his.
Case B: The same disease, no matched donor
Without a matched donor, this isn't the same decision Tomas faced with a different number attached — it's a genuinely different balance. A mismatched unrelated transplant at 58 carries real, well-documented increases in graft-versus-host disease and treatment-related mortality compared to a matched sibling graft in someone thirty years younger. Immunosuppressive therapy with ATG, cyclosporine, and eltrombopag added per RACE is where I'd start.
RACE randomized eltrombopag added to standard IST from the outset. Its primary endpoint was complete response at three months, met by twenty-two percent against ten percent on IST alone; by six months the overall response rate was sixty-eight percent against forty-one percent, and median time to first response fell from nearly nine months to three — a real improvement in exactly the population where transplant's downside risk is highest, and speed matters here.
I agree IST is the right starting point given the donor situation, and I'm not pushing for a mismatched unrelated graft as a first move. But I'd want the search for a matched unrelated donor kept open in parallel, not closed — if IST fails to produce a response by three to six months, a donor search started only then costs her real time she may not have.
Her hypertension is mild and well-controlled and shouldn't itself be read as disqualifying her from transplant later if a better-matched donor is eventually found; it's the mismatch, not her comorbidity burden, doing the real work in this recommendation.
Agreed: ATG, cyclosporine, and eltrombopag started together as induction, with an unrelated donor search kept open in parallel rather than deferred.
Not agreed: how long a response window to allow before escalating to a mismatched unrelated graft if a better match doesn't surface — the transplant physician favored a firmer three-month checkpoint, the hematologist preferred waiting the full six months RACE used before calling the induction a failure.