Stem Cell Transplant Death Rate: Factors and Outcomes

Stem cell transplant carries real mortality risk, but that risk has dropped substantially over the past three decades. In the 1990s, roughly three-year overall survival after an allogeneic (donor-based) transplant hovered around 41%; by the 2020s it had climbed to about 58%, driven largely by fewer treatment-related deaths rather than better control of the underlying disease. The factors that push mortality up or down are surprisingly varied, spanning the type of transplant, the patient’s fitness beforehand, how well the donor matches, the intensity of pre-transplant chemotherapy, and even the hospital where the procedure takes place.

How Much Has Survival Improved Over Time

The single biggest driver of better outcomes has been a decline in what transplant doctors call non-relapse mortality, meaning deaths caused by the transplant itself rather than by the cancer coming back. A 35-year Australian cohort study found that non-relapse mortality after allogeneic transplant fell from 55% in the 1990s to 25% in the 2020s, even though patients being transplanted today are older and sicker on average than those treated decades ago.¹ A large U.S. registry analysis tells a similar story: the frequency of early death (within about 200 days) unrelated to relapse fell from roughly 30% in the early 1990s to about 11% in the 2013–2017 era.²

These gains come from better supportive care, improved infection prevention, safer conditioning regimens, and more precise donor matching. After autologous transplants, where patients receive their own stem cells back, mortality from infections, organ toxicity, and graft complications also declined between an earlier era (1980–2001) and a more recent one (2002–2015).³ Still, despite these improvements, the risk of dying from a stem cell transplant remains considerable, and understanding where that risk comes from helps patients and families make informed decisions.

Autologous Versus Allogeneic Transplant

The most fundamental distinction in transplant risk is whether the stem cells come from the patient (autologous) or from a donor (allogeneic). Autologous transplants carry far lower treatment-related mortality because there is no risk of the graft attacking the patient’s body. In a meta-analysis of patients with relapsed B-cell lymphoma, treatment-related mortality was dramatically lower after autologous transplant, while allogeneic recipients had much higher rates of transplant-related death but also a lower rate of their cancer returning.⁴

The trade-off shows up clearly in specific diseases. For peripheral T-cell lymphoma, the three-year treatment-related mortality rate was about 32% for allogeneic transplant versus roughly 7% for autologous, yet overall survival at five years ended up comparable between the two approaches.⁵ A similar pattern appears in plasma cell leukemia: allogeneic transplant carried about 27% non-relapse mortality at three years versus roughly 7% for autologous, but allogeneic recipients relapsed less often, so by five years the two groups converged.⁶ In other words, allogeneic transplant is riskier upfront, but the donor immune system can keep the disease in check long-term. Choosing between the two is ultimately a bet: more upfront danger in exchange for a stronger anti-cancer effect, or a safer procedure with a higher chance the disease comes back.

Patient Fitness and Pre-Transplant Health

A patient’s overall health before transplant is one of the strongest predictors of whether they survive it. Transplant teams use a scoring tool that tallies pre-existing conditions like lung disease, liver problems, heart issues, and diabetes. In a large study of patients transplanted for non-cancerous blood disorders, those with the lowest comorbidity scores had about 83% two-year survival, while those with the highest scores survived just 56% of the time at two years.⁷

Age alone matters less than many people assume. What truly counts is a patient’s functional status, meaning how well they can carry out daily activities. A German registry study of transplant recipients aged 60 to 79 found that low functional status before transplant nearly doubled the risk of death, while just being older had a more modest effect.⁸ A Japanese validation study found that a combination of age over 50, poor functional status, and an unfavorable donor type together predicted worse outcomes, whereas comorbidity scores alone sometimes fell short in predicting who would die from the transplant itself.⁹ The practical takeaway: a fit 65-year-old with no major organ damage can be a better transplant candidate than a 40-year-old with significant heart or lung problems.

Donor Matching and Selection

For allogeneic transplants, how closely the donor’s immune markers match the patient’s is one of the biggest mortality levers. The human immune system relies on a set of surface proteins (HLA proteins) to distinguish self from non-self, and mismatches at these sites provoke graft-versus-host disease and graft rejection. In a study of Chinese patients receiving unrelated-donor transplants, mismatches at certain HLA sites more than doubled the risk of death.¹⁰

The penalty for mismatching grows steeper in older patients. A European study found that HLA mismatches are less well tolerated in patients over 55, making donor selection especially critical in that age group.¹¹ When cord blood is used instead of adult donor cells, higher-resolution matching at four key HLA sites made a meaningful difference: patients receiving a well-matched cord blood unit had essentially no treatment-related mortality in one study, while those with poor matches died at substantially higher rates.¹²

One reassuring finding is that donor age matters less than once feared. A study comparing grafts from donors over 60 to those from younger donors found no significant difference in rates of severe graft-versus-host disease, chronic graft-versus-host disease, or non-relapse mortality.¹³ Similarly, when comparing stem cell sources in adult unrelated-donor transplants, cord blood recipients had more early bacterial infections but no difference in infection-related mortality or overall survival at three years compared to recipients of bone marrow or peripheral blood stem cells.¹⁴

Conditioning Intensity

Before a transplant, patients receive chemotherapy (sometimes with radiation) to destroy their existing bone marrow and suppress the immune system enough to accept the graft. This conditioning can be full-intensity, called myeloablative, or gentler, called reduced-intensity. The choice creates a direct trade-off between transplant toxicity and relapse risk.

A landmark clinical trial following patients with acute myeloid leukemia and related cancers found that four-year treatment-related mortality was about 25% with full-intensity conditioning versus roughly 10% with reduced-intensity. But patients who received the gentler regimen relapsed at far higher rates, and overall survival was actually better in the full-intensity group.¹⁵ A separate propensity-matched analysis of patients with myeloid cancers found a similar pattern: no significant difference in overall survival or non-relapse mortality at two years, but reduced-intensity patients relapsed more often.¹⁶ For older or frailer patients who would not survive full-intensity treatment, reduced-intensity conditioning makes transplant possible at all, but the lower upfront death rate does not always translate into longer survival because the cancer is more likely to return.

Graft-Versus-Host Disease

Graft-versus-host disease (GVHD) is the complication most uniquely tied to allogeneic transplant. It occurs when the donor’s immune cells recognize the patient’s tissues as foreign and attack them. Acute GVHD, arising in the first weeks to months, can target the skin, gut, and liver. In one U.S. hospital analysis, patients who developed acute GVHD during their transplant hospitalization died at roughly three times the rate of those who did not (about 16% versus 5%).¹⁷

Severity matters enormously. A refined risk score for acute GVHD found that patients with high-risk disease involving multiple organs or severe organ damage had about 2.5 times the risk of transplant-related death compared to those with standard-risk GVHD.¹⁸ Chronic GVHD, which develops later and can linger for years, is a leading cause of late death after transplant. Paradoxically, chronic GVHD is associated with lower relapse risk because the donor immune system is actively policing the body for residual cancer cells. That anti-cancer benefit comes at a cost: more severe chronic GVHD increases non-relapse mortality and worsens quality of life.

One of the biggest recent advances in reducing GVHD has been the use of high-dose cyclophosphamide given shortly after transplant. In a comparison of haploidentical transplants (using half-matched family donors), a post-transplant cyclophosphamide regimen brought non-relapse mortality down to about 22% versus 30% with an older approach using anti-thymocyte globulin, with no increase in relapse.¹⁹ This strategy has expanded the donor pool dramatically, since nearly every patient has a half-matched family member available.

Infections and Organ Toxicity in the Early Period

The first few months after transplant are the most dangerous, and infection is a major killer. In an Australian registry report, among adults who died after allogeneic transplant, roughly half of reported deaths involved an infectious cause. Bacteria were the most common culprits, followed by fungi and viruses. In children, viruses were the leading infectious cause of death.²⁰

Another early threat is veno-occlusive disease (VOD), a condition where small blood vessels in the liver become blocked, sometimes leading to multi-organ failure. In a Spanish study of haploidentical transplants, about 6% of patients developed VOD, and among those, 43% died within 100 days of transplant. Nearly half the cases were classified as very severe.²¹ In children, VOD incidence varies by study (8–13%), and risk is highest in very young children and those receiving busulfan-based conditioning regimens.²²

Lung toxicity is another feared complication. Idiopathic pneumonia syndrome, a non-infectious inflammatory lung injury, affected about 4% of allogeneic transplant recipients in one large study. Among those who developed it, nearly half died within the first 120 days, compared to roughly 12% of patients who did not develop the syndrome.²³ In pediatric transplant recipients, overall transplant-related mortality was about 64% among children who developed this lung complication versus 17% among those who did not.²⁴

Relapse Remains the Biggest Threat

Despite all the dangers the transplant itself poses, disease relapse is the leading cause of death after stem cell transplant for leukemia and other blood cancers.²⁵ Once a cancer returns after transplant, treatment options narrow considerably. While some patients respond to additional therapy and achieve sustained remissions, many face high early mortality from aggressive relapse.²⁶

One approach to preventing relapse is donor lymphocyte infusion, where additional immune cells from the original donor are given after transplant to boost the anti-cancer immune response. A large European study found that when these infusions were given preemptively, in response to early signs of residual disease, about 70% of patients showed a response, and long-term survival among responders reached 55–76%.²⁷ A more recent comparison found that giving donor lymphocyte infusions early and routinely, rather than waiting for signs of disease, reduced two-year relapse rates from about 49% to 23% in patients with refractory or relapsed acute myeloid leukemia.²⁸

Secondary Cancers After Transplant

Surviving the transplant and staying in remission does not eliminate cancer risk. The conditioning chemotherapy and radiation, immune suppression, chronic GVHD, and disrupted immune surveillance all contribute to an elevated risk of developing an entirely new cancer years later. A Taiwanese population study found substantially higher rates of several cancers in transplant survivors, with bone cancer and cancers of the larynx and kidney among the most elevated.²⁹

A single-center study of allogeneic transplant survivors found that about one in five developed a secondary cancer, with the earliest appearing less than three months after transplant and no sign of risk leveling off at ten years. Among patients who developed a secondary cancer other than skin cancer, nearly half of all deaths were directly caused by that new malignancy.³⁰ A large European registry study of over 220,000 transplant recipients who developed solid secondary cancers found that five-year survival after the new cancer diagnosis varied widely: pancreatic, lung, and brain cancers had grim median survival of under a year, while thyroid, breast, and prostate cancers had median survival of ten or more years.³¹ These findings underscore why long-term cancer screening is not optional for transplant survivors.

Where You Get Transplanted Matters

An underappreciated factor in transplant mortality is the hospital itself. A U.S. study comparing transplant centers found that one-year survival was about 62% at high-volume centers versus 56% at low-volume centers, a gap of six percentage points. After adjusting for patient characteristics, transplant at a high-volume center was associated with roughly 30% better odds of being alive at one year. Centers that had dedicated survivorship programs for transplant recipients also showed independently better outcomes.³² A German registry analysis confirmed the pattern: non-relapse mortality at one year was modestly but significantly lower at centers performing 40 or more transplants per year compared to smaller programs.³³

The reasons are intuitive: high-volume centers develop deeper institutional expertise in managing rare complications, maintain more specialized nursing and pharmacy support, and often have quicker access to intensive care resources tailored to transplant patients. For patients with a choice, seeking out an experienced center is one of the few modifiable factors that can meaningfully shift their odds.

Socioeconomic and Cardiovascular Risks After Transplant

Social and economic factors also shape outcomes in ways that have nothing to do with biology. A study of unrelated-donor transplant recipients in the U.S. found that patients in the lowest income quartile had about 15% higher overall mortality and roughly 21% higher treatment-related mortality than wealthier patients, and this held true across all racial groups.³⁴ The mechanisms likely include disparities in access to follow-up care, caregiver support, housing near the transplant center, and the ability to comply with the demanding post-transplant medication and monitoring regimen.

For those who survive the transplant and avoid relapse, cardiovascular disease is an emerging long-term concern. The conditioning chemotherapy, radiation, and chronic GVHD treatments can damage the heart and blood vessels over time. Transplant survivors face elevated risk of heart failure, heart attacks, irregular heart rhythms, and fluid around the heart, with the risk profile depending on whether the transplant was autologous or allogeneic and what specific drugs were used.³⁵ As transplant survival improves and more patients live years or decades beyond their procedure, cardiovascular monitoring is becoming a standard part of survivorship care.