Background

Allogeneic hematopoietic cell transplantation (HCT) is one of the most curative therapeutic modalities for hematological malignancies. Recent advances in availability of stem cell sources, conditioning regimens, managements of graft-versus-host disease (GVHD) and supportive care have improved the outcomes of HCT. But even now, relapse is still the leading cause of treatment failure and some long-term survivors after HCT suffer late relapse of the original disease. To better understand the risk and risk factors of late relapse in HCT recipients, we conducted a nationwide retrospective study of the patients who received first HCT for hematological malignancies in Japan.

Patients and Methods

The study population included HCT recipients reported to the Japan Society for Hematopoietic Cell Transplantation. From this database, we extracted the data of patients with hematological malignancies who received first HCT between 1974 and 2011. There were 29,479 recipients selected according to this criterion. Then, we excluded 1,192 (4.0%) cases from the study because of missing key variables.

Results

A total of 28,287 recipients were evaluated in this study. Median age at transplant was 37 years old (range, 0-88), and 41.0% (n=11,605) were female. Underlying diseases were acute lymphoblastic leukemia (ALL) (n=7,300, 25.8%), acute myeloid leukemia (AML) (n=11,526, 40.7%), chronic myelogenous leukemia (CML) (n=3,038, 10.7%), myelodysplastic syndrome (MDS) (n=3,235, 11.4%), malignant lymphoma (ML) (n=2,956, 10.5%) and plasma cell disorder (PCD) (n=232, 0.8%). Type of transplant included bone marrow transplantation (BMT) from HLA-matched or 1 locus serological mismatched (“matched”) related donor (n=8,002, 28.3%), BMT from more than 1 locus-mismatched (“mismatched”) related donor (n=261, 0.9%), BMT from HLA-matched unrelated donor (n=9,074, 32.1%), BMT from HLA-mismatched unrelated donor (n=1,112, 3.9%), peripheral blood stem cell transplantation (PBSCT) from “matched” related donor (n=4,385, 15.5%), PBSCT from “mismatched” related donor (n=467, 1.7%), cord blood transplantation (CBT) from “matched” unrelated donor (n=2,326, 8.2%) and CBT from “mismatched” unrelated donor (n=2,556, 9.0%). Myeloablative conditioning regimens were applied to 77.0% (n=21,773) of recipients and reduced-intensity conditioning regimens to 22.1% (n=6,247). Median follow-up of survivors was 1,934 days (range, 18-11,123).

Event-free survival of all cases was 41.8% and cumulative incidence of relapse (RI) of them was 33.1% at 5 years. Disease-free survival (DFS) and RI at 7 years of recipients who had been in remission for 2 years since HCT (“DF2”, n=11,500) were 83.2% and 10.2%. DFS and RI at 10 years of recipients who had been disease-free for 5 years (“DF5”, n=7,093) were 91.5% and 2.9%. In multivariate analysis, disease, disease risk at transplant and patients’ age were powerful (p<1×10-6) significant variables for DFS both of DF2 and DF5. For risk of relapse, disease and disease risk at transplant were powerful (p<1×10-4) significant variables of both DF2 and DF5, but patients’ age and intensity of conditioning regimen were only of DF2.

Adjusted for patients’ age, disease risk at transplant, year of transplant, donor-recipient sex match, ABO disparity, type of transplant and intensity of conditioning regimen, acute and chronic GVHD were included as time-dependent covariates in Cox model for relapse. In DF2, grade II-IV acute GVHD was associated with a reduced risk of relapse for patients with ALL (Hazard Ratio 0.75, 95% confidence interval 0.57-0.99, p=0.043) and AML (0.74, 0.57-0.97, p=0.029), and chronic GVHD for those with CML (0.65, 0.47-0.89, p=0.0067) and ML (0.37, 0.20-0.60, p=0.00067). In DF5, grade II-IV acute GVHD significantly reduced risk of relapse for recipients with AML (0.40, 0.18-0.85, p=0.018), and chronic GVHD for those with CML (0.56, 0.34-0.92, p=0.021).

Conclusion

Recipients of HCT for hematological malignancies who remain in remission for more than 2 or 5 years have favorable long-term survival. Our study demonstrates there are relatively small but persistent risks of relapse and underlying disease and disease risk at transplant are the major risk factors of late relapse. We also show acute and chronic GVHD are associated with a reduced risk of late relapse but the strength of this association differs between underlying diseases.

Disclosures:

No relevant conflicts of interest to declare.

Sign in via your Institution