Stem Cell Transplants for Multiple Myeloma: What a 10-Year European Study Reveals About Combining Your Own Cells With a Donor's Cells vs. Using Your Own Cells Alone

Table of Contents

• Key Points
• Understanding the Research: Why This Study Matters
• How the Study Was Conducted
• Key Findings: What the Results Showed
• Patients With High-Risk Chromosome Changes
• Comparing Patients Who Actually Received Their Full Treatment Plan
• Graft-Versus-Host Disease: A Key Side Effect of Donor Transplants
• What This Means for Patients
• Study Limitations
• Questions to Discuss With Your Doctor
• Frequently Asked Questions
• Source Information

Key Points

• In a 357-patient European study, five-year progression-free survival was 35% with auto-allo transplant versus 18% with autologous transplant alone.
• At five years, relapse occurred in 49% of the auto-allo group versus 78% of the autologous-only group.
• Five-year overall survival was 65% with auto-allo versus 58% with autologous transplant alone.
• Nonrelapse mortality was higher with auto-allo: 12% versus 3% at two years.
• The study was not randomized and used older treatments, so current risks and benefits may differ.

Understanding the Research: Why This Study Matters

Multiple myeloma is a cancer of plasma cells, a type of white blood cell that normally helps fight infections. Despite significant advances in treatment, myeloma is still considered an incurable disease, although some patients do achieve long-lasting remissions, especially after intensive high-dose chemotherapy. For adults up to about age 65, high-dose therapy followed by an autologous stem-cell transplant (ASCT) — a procedure where a patient's own blood-forming stem cells are collected, stored, and returned after high-dose chemotherapy — has become part of the standard first-line treatment plan.

There is another type of transplant, called an allogeneic stem-cell transplant (alloSCT), where stem cells come from a donor. When this approach was first tried in the mid-1980s using very intensive (myeloablative) conditioning, it showed promise because of a unique "graft-versus-myeloma" effect, where the donor's immune cells attack the cancer. However, this older method was hampered by severe toxicity and a very high rate of treatment-related death, ranging from 30% to 50% in some studies. In fact, one case-control trial found that patients actually did better with their own stem cells than with a myeloablative donor transplant, despite the donor approach having a lower relapse rate.

The development of reduced-intensity conditioning (RIC) changed the picture. This gentler approach uses lower doses of radiation and chemotherapy, making the transplant safer. When high-dose chemotherapy with autologous transplant was combined with a subsequent reduced-intensity allogeneic transplant (called auto-allo), the treatment-related death rate dropped to 15% or less. The question remained, however: is this tandem auto-allo approach actually better than the "gold standard" autologous transplant alone? Earlier smaller studies had produced conflicting results.

This study was designed to answer that question in a large group of patients with newly diagnosed myeloma, with a long follow-up period to see whether the benefits — and risks — held up over time. The results were published in the Journal of Clinical Oncology, one of the most respected peer-reviewed journals in cancer medicine.

How the Study Was Conducted

Who Was Included?

From February 2001 through January 2005, a total of 357 patients with multiple myeloma, up to age 69 years, were enrolled at 23 European Bone Marrow Transplantation (EBMT) centers. To be eligible, patients had to have achieved at least stable disease on their first-line (initial) treatment. This included those in complete response (CR) — meaning no detectable cancer; partial remission (PR) — meaning significant reduction in cancer; or stable disease (SD) — meaning the cancer hadn't grown. All patients had undergone HLA typing, a blood test that determines tissue compatibility for transplantation.

A key point is how patients were assigned to treatment groups. This was not a "randomized" trial in the traditional sense, where patients are randomly assigned by a computer. Instead:

• 108 patients who had an HLA-identical sibling (a brother or sister with matching tissue type) were assigned to the auto-allo arm (autologous transplant followed by reduced-intensity donor transplant)
• 249 patients without a matched sibling were assigned to the auto arm (autologous transplant alone)

Interestingly, two patients in the auto-allo arm actually had a sibling donor with one HLA mismatch, not a perfect match, and were mistakenly treated on the auto-allo protocol. In keeping with standard statistical practice, they were still included in the auto-allo group in what's called an intention-to-treat (ITT) analysis — meaning patients are analyzed in the group they were assigned to, regardless of whether they received exactly the planned treatment.

The time point for enrolling in the trial was at the time of the first autologous transplant, after completion of induction treatment. Patients could have either a single autologous transplant or a tandem (double) autologous transplant — this was left to each study center's discretion. Patients with substantial kidney failure (glomerular filtration rate below 50 mL/min), significant liver impairment (bilirubin more than 2 times the upper limit of normal), severe heart failure (left ventricular ejection fraction below 40%), or major dysfunction of other organ systems were excluded from the study.

Baseline characteristics were evenly balanced between the two groups, with one notable exception: age at diagnosis. The auto group was slightly older, with a median age of 57 years, compared with 54 years in the auto-allo group. The median follow-up time from the first transplant was 61 months (approximately 5 years), with a range of 21 to 91 months, for patients alive at their last follow-up visit.

What Treatments Did Patients Receive?

All patients received induction chemotherapy before the transplant. The most common regimen was VAD, which stands for vincristine, doxorubicin, and dexamethasone. VAD was used in 73% of patients in the auto-allo arm and 67% in the auto arm. The remaining patients received a variety of other regimens, mostly based on cyclophosphamide or dexamethasone. Importantly, no patients in this study were treated with newer "novel" drugs such as thalidomide, lenalidomide, or bortezomib, which have since become standard in myeloma treatment — this reflects the era when the study was conducted (2001–2005).

All 357 patients received high-dose chemotherapy with melphalan at a dose of 200 mg/m², followed by an infusion of their own collected stem cells. Supportive care (medications to prevent infections, growth factors to boost blood counts, and similar measures) was given according to each center's usual routines.

Of the 108 patients assigned to the auto-allo arm, 91 actually received the reduced-intensity allogeneic transplant according to the protocol. The remaining 17 did not receive their planned donor transplant for these specific reasons:

• 7 patients had disease progression before the second transplant
• 4 patients declined the transplant
• 1 patient died before the allogeneic transplant could be done
• 1 patient developed kidney failure
• 1 patient had failure to collect donor stem cells
• 3 patients had a donor who became ill or unavailable; in one of these cases where the donor declined, the patient received a transplant from a matched unrelated donor instead

All 17 of these patients were still included in the auto-allo group for the intention-to-treat analysis. The median time between the autologous transplant and the allogeneic (donor) transplant was 4.2 months, with a range of 1.3 to 22.2 months.

The reduced-intensity conditioning (gentle prep) regimen for the donor transplant consisted of fludarabine 30 mg/m² per day for 3 days plus total-body irradiation (TBI) at a low dose of 2 Gy, given in one session. To prevent graft-versus-host disease (GvHD) — a condition where the donor's immune cells attack the patient's body — patients received cyclosporine (6.5 mg/kg taken by mouth twice a day, or 1.5 mg/kg by vein twice a day, starting from day −1 until it could be taken orally) plus mycophenolate mofetil (15 mg/kg by mouth twice a day, from day 0 through day 24).

Patients in the auto-only group who had no matched sibling donor received either no further treatment (145 patients) or, at their center's discretion, underwent a second autologous transplant as part of a tandem transplantation program (104 patients). The conditioning for that second autograft was the same as the first: melphalan 200 mg/m². After any allogeneic transplant, further treatment was optional.

How Were Outcomes Measured?

The primary end point (main outcome being measured) was progression-free survival (PFS) — the length of time a patient lives without the cancer growing or returning, measured from the date of enrollment in the study (the date of the first autologous transplant). Secondary end points included overall survival (OS), relapse rate, complete remission (CR) rate, and the incidence of nonrelapse mortality (NRM) — that is, death from causes other than the myeloma itself, such as complications of the transplant.

The researchers used sophisticated statistical methods. Because the risk patterns changed over time (the curves "crossed" early on, meaning one group could have worse outcomes initially but better outcomes later), standard survival analysis had to be adjusted. They used models with time-varying effects, adjusted for age, and performed landmark log-rank tests to assess differences in the long term (after 2 years for most outcomes, and after 3 years for overall survival). The main analysis followed the intention-to-treat principle. Additionally, an exploratory ITT analysis was done in two subgroups based on whether a specific chromosomal abnormality was present, and outcomes were also compared in a per-protocol analysis that included only patients who actually received the planned second transplant type (91 auto-allo patients vs. 104 tandem auto patients).

Key Findings: What the Results Showed

Progression-Free Survival: A Clear Advantage for the Tandem Approach

At 60 months (5 years) after the first autologous transplant, progression-free survival was significantly better in the auto-allo group: 35% compared with just 18% in the auto-only group (P = .001). In plain terms, this means patients who received the tandem auto-allo approach were roughly twice as likely to be alive without their myeloma returning at the 5-year mark. The benefit for the auto-allo group began to emerge after about 2 years of follow-up, corresponding to a significantly lower risk of relapse or disease progression (P = .003).

Relapse Rates: Dramatic Difference

At 60 months, the incidence of relapse or disease progression was 49% in the auto-allo group versus 78% in the auto group — a difference of nearly 30 percentage points. This is one of the most striking findings of the study and directly illustrates the power of the "graft-versus-myeloma" effect, where the donor's immune system actively works to eliminate the cancer cells.

Overall Survival: Better Long-Term Results

Long-term overall survival was also significantly superior in the auto-allo group. The auto-allo group showed a significant reduction in the risk of death over time (P = .006), with a significantly lower risk of death after the 3-year mark (P = .047). At 60 months, overall survival was 65% in the auto-allo group compared with 58% in the auto group. While these numbers may look somewhat close, the statistical analysis confirms the difference was meaningful, particularly as time went on.

Complete Remission Rates: Higher With Auto-Allo

Complete remission — the disappearance of all detectable signs of cancer — was achieved more often with the tandem approach. The CR rate within 60 months was 51% in the auto-allo group versus 41% in the auto group (P = .020). For patients who did not achieve CR, the best response status differed as well:

• Partial response: 43% (auto-allo) vs. 50% (auto)
• No response: 3% (auto-allo) vs. 5% (auto)
• Progressive disease: 3% (auto-allo) vs. 4% (auto)

Nonrelapse Mortality: The Price of the Tandem Approach

Nonrelapse mortality — death caused by the transplant itself rather than by the myeloma — was higher in the auto-allo group, as might be expected with a more intensive treatment. The cumulative NRM at 24 months was 12% after auto-allo versus 3% in the auto group (P < .001). At 60 months, the figures were 16% and 4%, respectively. In other words, while the donor transplant approach led to fewer deaths from myeloma, it caused more deaths from transplant complications. The "balancing act" between these competing risks is a central theme in interpreting this study.

Patients With High-Risk Chromosome Changes

A particularly important aspect of this study was the analysis of patients with a specific chromosomal abnormality known as deletion of chromosome 13 — del(13q14), which has historically been associated with a poorer prognosis in myeloma. Cytogenetic analysis was performed in 214 patients using fluorescent in situ hybridization (FISH), a technique that allows scientists to see specific genetic changes within cells.

Of the 214 patients tested:

• 92 patients had the del(13) abnormality (29 in the auto-allo group, 63 in the auto group)
• 122 patients were negative for del(13) (34 in the auto-allo group, 88 in the auto group)

Findings in Patients With del(13)

For patients with this high-risk chromosome deletion, the benefit of the tandem approach was especially pronounced:

• Progression-free survival at 60 months: 31% with auto-allo vs. 11% with auto (P = .002)
• Overall survival at 60 months: 69% with auto-allo vs. 55% with auto (P = .003) — a striking 14-percentage-point improvement
• Relapse/progression risk after 2 years: significantly lower in the auto-allo group (P = .004); at 60 months, the relapse rate was 55% vs. 86%

These numbers suggest that patients with the del(13) abnormality — a group often considered to have harder-to-treat disease — derived substantial benefit from the donor immune cells' anti-myeloma effect.

Findings in Patients Without del(13)

For patients negative for del(13), the results also favored the auto-allo approach, though the differences were somewhat smaller:

• Progression-free survival at 60 months: 44% vs. 20% (P = .017)
• Overall survival at 60 months: 70% vs. 61% (P = .363) — not statistically significant in this subgroup
• Relapse/progression rate: 39% vs. 76% (P = .005 for the hazard after 2 years)

The researchers noted that a tendency toward better outcome was found in both the del(13) and non-del(13) patient groups, which supports and corroborates the findings seen in the overall study population. In other words, the benefit of the tandem approach was broad-based, not limited to one genetic subgroup.

Comparing Patients Who Actually Received Their Full Treatment Plan

To get a cleaner picture of the true effect of the two treatment strategies, the researchers performed a "per-protocol" analysis, comparing the 91 patients who actually received their planned reduced-intensity donor transplant with the 104 patients who received a second autologous transplant as part of a planned tandem program. This analysis measured outcomes from the time of the second transplant:

• Progression-free survival at 60 months: 39% (auto-allo) vs. 19% (tandem auto) (P = .004)
• Overall survival at 60 months: 63% vs. 60% (P = .753) — but with a highly significant trend of reduction in risk over time (P < .001)
• Relapse/progression rate: 43% vs. 78% (P = .001 for the hazard after 2 years)
• Complete remission rate within 60 months: 56% vs. 44% (P = .007)
• Nonrelapse mortality at 60 months: 18% vs. 3% (P < .001)

For patients who did not achieve complete remission in this per-protocol analysis, the best response status was partial response in 35% (auto-allo) vs. 51% (auto), no response in 6% vs. 3%, and progressive disease in 3% vs. 2%.

This analysis reinforces the main findings: the auto-allo approach produced roughly double the progression-free survival and drastically lower relapse rates, at the cost of a higher nonrelapse mortality rate. The overall survival difference was less dramatic early on, but the trend over time strongly favored the donor transplant group.

Graft-Versus-Host Disease: A Key Side Effect of Donor Transplants

Among the 91 patients who actually received the reduced-intensity donor transplant, graft-versus-host disease was a significant concern. GvHD occurs when the donor's immune cells (the graft) recognize the patient's body (the host) as foreign and attack healthy tissues. It can affect the skin, liver, gastrointestinal tract, and other organs, and it can range from mild to life-threatening.

Acute GvHD (occurring early, usually within the first 100 days)

• Grade 1: occurred in 10 patients (11%)
• Grade 2: occurred in 8 patients (9%)
• Grade 3: occurred in 8 patients (9%)
• Grade 4: occurred in 2 patients (2%)
• No acute GvHD at all: 60 patients (67%)

In total, the incidence of grade 2 to 4 acute GvHD was 20% — meaning one in five patients experienced a moderate to severe early immune reaction. This aligns with what's expected for reduced-intensity conditioning transplants in myeloma patients.

Chronic GvHD (occurring later, often after day 100)

A total of 49 patients (54%) developed chronic GvHD. This was classified as:

• Limited chronic GvHD: 28 patients (31%) — typically affecting only the skin or a single organ
• Extensive chronic GvHD: 21 patients (23%) — involving multiple organs or more severe manifestations

Chronic GvHD can significantly affect quality of life, requiring ongoing immunosuppressive treatment. However, it's worth noting that the presence of GvHD is also often associated with a stronger graft-versus-myeloma effect, which may partially explain the lower relapse rates seen in the auto-allo group. This is sometimes called the "graft-versus-leukemia/lymphoma/graft-versus-tumor" trade-off: some GvHD may be a marker that the donor immune cells are actively working, but too much can be harmful.

What This Means for Patients

This study provides important, long-term evidence about treatment strategies for multiple myeloma. Here's how to make sense of the results:

The tandem auto-allo approach offers dramatically better disease control. The relapse rate at 5 years was 49% with auto-allo compared with 78% with auto alone — a massive difference. Progression-free survival was nearly doubled (35% vs. 18%). For patients who are candidates for this approach, the chance of staying in remission longer without the myeloma coming back is substantially higher.

The survival benefit grows with time. While overall survival at 5 years was 65% vs. 58%, the statistical analysis showed that the survival advantage of auto-allo strengthened over time, particularly after the 3-year mark. This suggests the benefit is durable and not just an early effect.

But there's a real trade-off. The nonrelapse mortality was 12% at 2 years in the auto-allo group vs. 3% in the auto group. This means that some patients in the donor transplant group died from transplant complications, including GvHD and infections. For every 100 patients treated with the auto-allo approach, roughly 9 to 12 additional deaths from transplant-related causes occurred compared with the auto-only approach. However, these early risks were counterbalanced by many fewer deaths from myeloma over the long run.

High-risk patients may benefit most. Patients with the del(13) chromosomal abnormality — often considered a poor-prognosis marker — showed especially large benefits from the tandem approach: overall survival at 5 years was 69% vs. 55%. This suggests that the donor immune effect may be particularly valuable in patients whose disease is otherwise harder to control.

This study reflects an older treatment era. None of the patients received novel agents like thalidomide, lenalidomide, or bortezomib, which are now standard in first-line myeloma therapy. Modern regimens incorporating these drugs may alter the risk-benefit calculus. Patients today might have better outcomes with either approach, and the comparison could differ with newer treatments. Additionally, the study enrolled patients from 2001 to 2005, so supportive care and transplant techniques have improved since then.

Study Limitations

Understanding the limitations of this study is crucial for interpreting the results appropriately:

• Not a randomized trial. Patients were assigned to treatment groups based on whether they had a matched sibling donor. This creates the possibility of "selection bias" — the two groups might differ in ways beyond just the treatment received. The authors note that age was slightly different (median 54 vs. 57 years), but other factors — including general health, fitness for transplantation, and unmeasured variables — could also have differed. Having a sibling donor can be associated with socioeconomic and family factors that could influence outcomes. The researchers adjusted for age in their statistical models, but as with any nonrandomized study, residual confounding is possible.
• Limited to a specific era of treatment. No patients received modern agents like thalidomide, lenalidomide, or bortezomib, and the conditioning regimens and supportive care reflect 2001–2005 practices. Results may not fully apply to patients treated with today's standard protocols.
• Relatively small subgroups. The cytogenetic subgroup analyses were exploratory, and the confidence intervals around the estimates were wide. For instance, in the non-del(13) subgroup, the overall survival difference was not statistically significant (P = .363).
• Donor availability as a selection criterion. The fact that the auto-allo group all had siblings who were HLA-identical could introduce bias — these patients were "selected" by a biological availability factor that could correlate with other characteristics.
• Treatment modifications. Seventeen patients in the auto-allo arm (16%) didn't receive their planned allogeneic transplant. While the intention-to-treat analysis handles this appropriately for statistical purposes, it also means the "real-world" application of the auto-allo plan can be disrupted by disease progression or other complications before the second transplant.
• Genetic testing only for del(13). The study didn't incorporate other important high-risk markers now known to affect myeloma prognosis, such as chromosome 17p deletions, translocations like t(4;14) or t(14;16), or the more recent gene expression profiling that's often used in modern practice.

Questions to Discuss With Your Doctor

Based on this research and its implications, here are some important questions patients with multiple myeloma (or their family members) may want to discuss with their oncology team:

1. Am I a candidate for a tandem auto-allo transplant? Factors like your age, overall health, organ function, and availability of a matched donor (sibling or unrelated) all matter. The study specifically included patients up to age 69 with adequate organ function.
2. What is my cytogenetic (chromosome) risk profile? The results suggest that patients with high-risk features like del(13) may derive particularly large benefits from the donor transplant approach. Ask what testing has been done and how it affects your treatment recommendations.
3. What are my personal risks of transplant-related complications? Nonrelapse mortality at 2 years was about 12% in the auto-allo group vs. 3% in the auto group. Your doctor can help assess how your individual health status (heart, kidney, liver function, age, fitness) might shift these risks.
4. How do modern drugs change the equation? This study was conducted before thalidomide, lenalidomide, bortezomib, and carfilzomib became routine. Ask how your planned treatment, including any novel agents, might interact with or modify the transplant strategy.
5. What would my quality of life look like? Chronic GvHD occurred in over half of the auto-allo patients (54%), which can require long-term immunosuppression and can impact daily life. Discuss what supportive care measures would be in place and how GvHD would be managed.
6. What's the timing? In this study, the median interval between the autologous and allogeneic transplant was about 4 months. Ask about the recommended timeline for your own treatment plan.

The decision between an autologous transplant alone versus a tandem auto-allo approach is deeply personal and depends on a careful weighing of risks and benefits. For some patients — particularly those with high-risk disease, a good performance status, and a suitable donor — the substantially lower relapse rate and better long-term survival may clearly justify the higher early transplant-related risks. For others, particularly those with significant coexisting health conditions, the auto-only approach may represent the safer, more appropriate choice.

What makes this study valuable is its size (357 patients), its long follow-up (median 61 months), and its comprehensive reporting of both benefits (improved PFS, OS, relapse rates) and harms (increased NRM and GvHD). It provides patients and doctors with realistic, evidence-based information to make shared treatment decisions.

Frequently Asked Questions

What are the two transplant strategies for multiple myeloma compared here?

One is high-dose chemotherapy followed by an autologous transplant using your own collected stem cells. The other, called auto-allo, adds a later reduced-intensity donor transplant. In a study of 357 patients, the auto-allo approach roughly doubled the chance of being alive without the myeloma returning at five years, but carried higher early risks.

Who could participate in this transplant study?

Patients were up to age 69 with newly diagnosed multiple myeloma and had to have at least stable disease after first-line chemotherapy. All had adequate organ function and HLA typing. Those with a matched sibling were assigned to the auto-allo approach; those without received an autologous transplant alone. This was not a randomized trial.

How much better was progression-free survival with the tandem auto-allo approach?

At five years, progression-free survival was 35% with auto-allo versus 18% with autologous transplant alone. That means patients in the donor transplant group were roughly twice as likely to be alive without myeloma recurrence at that point, based on 357 patients followed for a median of 61 months.

What were the main risks of the auto-allo transplant approach?

The donor transplant caused more treatment-related deaths and graft-versus-host disease. At two years, nonrelapse mortality was 12% with auto-allo versus 3% with autologous transplant alone. Chronic GvHD affected 54% of patients who actually received the donor transplant. However, fewer patients in the auto-allo group died from myeloma over the long term.

Are these results still relevant with modern myeloma drugs?

This study treated patients from 2001 to 2005, before drugs like thalidomide, lenalidomide, and bortezomib were standard. Modern treatments may improve outcomes with either approach. Also, the study was not randomized, so results may not fully apply to today's protocols. Discuss with your oncology team how newer therapies change the balance.

Should I get a second opinion on choosing a tandem auto-allo stem cell transplant versus an autologous transplant alone for multiple myeloma?

A second opinion is worth considering if your treatment team recommends one strategy but you want a deeper look at the trade-offs. In a 357-patient European study with a median 61-month follow-up, the tandem auto-allo approach roughly doubled 5-year progression-free survival (35% vs. 18%) and cut relapse rates (49% vs. 78%), but raised nonrelapse mortality at 2 years (12% vs. 3%). Benefit appeared largest in patients with the del(13) high-risk change. Because assignment was based on donor availability rather than randomization, another expert can help you weigh these risks and benefits. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original Article Title: Tandem Autologous Reduced-Intensity Conditioning Allogeneic Stem-Cell Transplantation Versus Autologous Transplantation in Myeloma Long-Term Follow-Up Journal of Clinical Oncology

Authors: Bo Björkstrand, Simona Iacobelli, Ute Hegenbart, Astrid Gruber, Hildegard Greinix, Liisa Volin, Franco Narni, and Gösta Gahrton (with additional contributors)

Journal: Journal of Clinical Oncology, July 05, 2011, Volume 29, Issue 22, pages 3016–3022

DOI: 10.1200/JCO.2010.32.7312

Study Registration: The trial was conducted across 23 European Bone Marrow Transplantation (EBMT) centers, enrolling patients from February 2001 through January 2005.

This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and is not a substitute for individualized medical advice from a qualified health care professional. Treatment decisions should always be made in consultation with your oncology team.