"Not a Perfect Match" Is No Longer a Dead End

If you have read anything about bone marrow donation and race, you have met this number: a white patient searching the United States registry for a fully matched unrelated donor finds one roughly 75 to 80 percent of the time, and a Black patient finds one around 30 percent of the time or less.

That number is real, and it is still the strongest argument for growing the registry. But for years it carried an implication that has quietly stopped being true — that a patient who does not find a perfect match is out of options.

What "a perfect match" actually means

Marrow transplants are matched on HLA markers, proteins that tell your immune system which cells belong to you. Transplant teams typically look at eight of them. A donor who matches on all eight is described as 8/8, and for a long time 8/8 was close to a requirement, because mismatched cells trigger graft-versus-host disease — the transplanted immune system attacking the patient's body.

HLA markers are inherited in linked sets, which is why ancestry matters so much. It is also why the registry's composition, not the patient's biology, drives the gap.

What changed

The change is a drug called cyclophosphamide, given after the transplant rather than before. Timed correctly, post-transplant cyclophosphamide appears to remove the donor immune cells most likely to attack the patient while sparing the ones that rebuild the blood system.

The approach was developed for half-matched family donors. The open question was whether it would work for mismatched unrelated donors — the ones a registry can actually supply.

The ACCESS results

The NMDP-sponsored ACCESS trial tested exactly that. Results were published in the peer-reviewed *Journal of Clinical Oncology* (JCO-25-00856).

The trial enrolled 145 adult patients across 21 United States transplant centers, using unrelated donors matched at 4/8 to 7/8 — every one of them a mismatch. Notably, 59 percent of participants self-identified as members of underrepresented racial and ethnic groups, which is close to the reverse of most transplant trial populations.

One-year overall survival was 83.8 percent for patients who received myeloablative conditioning and 78.6 percent for reduced-intensity or non-myeloablative conditioning. The historical benchmark for fully matched 8/8 unrelated donors is about 75 percent.

Severe graft-versus-host disease, the thing the whole approach was designed to prevent, stayed low: grade III-IV acute GVHD at six months was 8 percent in the myeloablative group and 10 percent in the reduced-intensity group. Severe chronic GVHD at twelve months was 3 and 4 percent.

At the 2026 Tandem Meetings in February, a larger prospective analysis of **268 patients** reported one-year overall survival above 80 percent across every match level — 86 percent for patients with donors matched at less than 7/8, and 79 percent at 7/8. Patient-reported quality of life and physical function had returned to baseline at one year, with no clinically meaningful difference between the 7/8 group and the 4/8-to-6/8 group. Rachel Cusatis, PhD, of the Medical College of Wisconsin put it plainly: patients are not only surviving after a mismatched unrelated donor transplant, they feel they are returning to baseline.

Monzr M. Al Malki, MD, of City of Hope, an ACCESS study co-chair, has gone further, stating that virtually every patient searching international registries now has a greater than 99 percent likelihood of identifying a suitable donor.

What this does not mean

That last figure is a physician's characterization presented at a conference, not a published survival statistic, and the word doing the work in it is suitable — not matched. It is worth holding onto the difference.

Several other things are worth saying directly:

• These are results from clinical trials at experienced transplant centers with protocol support. They are not a promise about any individual patient's outcome.
• Mismatched transplant is not risk-free. In the 268-patient analysis, 68 percent of patients had at least one grade 2 or higher infection within the first year, and chronic GVHD was the single biggest factor affecting quality of life.
• Finding a donor is not the same as reaching transplant. Insurance, distance, caregiver support, and time all decide whether an available donor becomes an actual transplant.
• None of this closes the registry gap. It widens the door for patients who were previously turned away at it.

So why keep growing the registry

Because a wider door still needs people standing behind it.

Every one of those trial patients had a donor to receive cells from. Somebody had joined a registry years earlier, for no particular reason, and stayed reachable. Mismatched transplant expands who counts as a usable donor; it does not create donors.

And better odds at 7/8 do not remove the advantage of 8/8. The goal has not changed — it has just stopped being the only acceptable outcome.

If you have ever been told that your ancestry makes you unlikely to help anyone, or that your family search failed and that was the end of it, the honest 2026 answer is different from the 2015 answer. It is worth asking again.