Primary BMT CTN 1702 results support use of search prognosis to guide donor search and selection

January 2026

Allogeneic hematopoietic cell transplant (alloHCT) is the standard of care for patients with high-risk or advanced hematological malignancies. Although the use of alloHCT has grown over the past decades, it can be challenging to get all eligible patients to transplant at the optimal time with an HLA-matched donor. About two-thirds of patients who need a transplant do not have a fully HLA-matched donor available and must rely on alternative donor sources, including haploidentical donors, mismatched unrelated donors (MMUD) or umbilical cord blood (UCB). In the era of improved graft-versus-host disease (GVHD) prophylaxis and alternative donor HCT outcomes, delaying transplant to find a fully matched donor for patients unlikely to have one may compromise outcomes.

Search prognosis refers to the estimated likelihood that a patient has an 8/8 HLA-matched unrelated donor (MUD) on the registry. Researchers evaluated if using a search prognosis score—based on patient HLA typing and ancestry—to guide the initial search strategy could improve transplant timing and outcomes. Under this approach, patients unlikely to find a suitable MUD would immediately pursue an alternative donor rather than prolonging their donor search.

BMT CTN 1702 was a multi-center, U.S.-based biologic assignment study analyzed on an intention-to-treat basis. Eligible patients did not have an available matched sibling/related donor and were suitable candidates for alloHCT. Participating centers were required to conform to the recommended search algorithm and to perform alloHCT within the next 6 months if a suitable donor was identified.

The analysis included 1,751 evaluable patients with hematological malignancies or other blood disease (severe aplastic anemia or sickle cell disease) requiring alloHCT. Patients were assigned to three groups based on search prognosis:

• Very Likely (experimental arm): 958 patients (54.7%)
• Very Unlikely (experimental arm): 276 patients (15.8%)
• Less Likely (observational arm following usual treatment practice): 517 patients (29.5%)

1,179 patients were transplanted (67.3%), stratified by search prognosis:

• Very Likely: 668 patients
• Less Likely: 340 patients
• Very Unlikely: 170 patients

The distribution of donor types differed between the Very Likely and Very Unlikely groups:

• Very Likely: MUD (94%), haploidentical related (3.9%), UCB (1%), MMUD (0.7%) or 1-locus mismatched related donors (0.3%)
• Very Unlikely: haploidentical related (59.4%), MMUD (22.4%), MUD (9.4%) UCB (7.6%) or 1-locus mismatched related donors (1.2%)

The primary endpoint for this study was 2-year overall survival (OS) from evaluability, compared between the Very Likely and Very Unlikely patient groups. Secondary endpoints included treatment-related mortality (TRM), disease-free survival (DFS), acute GVHD (aGVHD) and chronic GVHD (cGVHD). 

There was no significant difference in 2-year OS between the Very Likely and Very Unlikely groups:

• Univariate analysis: HR=1.00 [95%CI, 0.82-1.21], p=0.98
• Multivariate analysis: HR=1.07 [95%CI, 0.86-1.33], p=0.56

Multivariate analysis showed no differences in relapse, TRM, DFS, aGVHD and cGVHD among the Very Likely, Very Unlikely and Less Likely groups.

Using a donor search prognosis strategy to prioritize an alternative donor for patients who were very unlikely to find an MUD resulted in survival and transplant outcomes that were statistically comparable to those who were very likely to find an MUD. Results support that patients very unlikely to have an 8/8 MUD should be directed to alternative donors without a prolonged search for an MUD. These findings support expanded access to alloHCT for patients who are racially or ethnically diverse and may not have an MUD on the registry.

To further understand how considering search prognosis early in a patient’s transplant journey can inform search strategy, view a summary of updated donor selection guidelines. For personalized support to identify and incorporate search prognosis into your workflow, reach out to your NMDPSM Clinical HLA Services partner or contact search-strategies@nmdp.org.

Lee, et al., published in the Journal of Clinical Oncology