Abstract
Relapse is the major cause of failure after allogeneic hematopoietic cell transplantation (HCT) in myelodysplastic syndromes (MDS). We analyzed results in 258 MDS patients transplanted between June 2004 and December 2013 and who were evaluable for post-HCT minimal residual disease (MRD). Patient characteristics are summarized in Table 1. MRD markers were determined by multiparameter flow cytometry (MFC) and cytogenetics/Fluorescence In Situ Hybridization (FISH) on marrow aspirates at day-28 post-HCT. Thirty-two patients (12.4%) were MRD positive by either MFC or cytogenetics, and 20 patients (7.8%) had early relapse (more than 5% blasts on marrow aspirate). The 2-year overall survival (OS) from day-28 post-HCT of MRD positive patients and early relapsing patients was 21% and 5%, respectively, p=0.001, log-rank test. Despite a statistically significant difference, the outcome of both the subgroups was very poor. Conversely, the 206 MRD negative patients 2-year OS was 70%, Figure 1.
We compared the probability of being MRD positive or relapsing early between patients who received low- vs. high-intensity conditioning using stepwise logistic regression. Candidate variables included gender, age at transplant, secondary MDS, pre-HCT treatments (induction chemotherapy given as first line treatment or as salvage, hypomethylating agents only, no therapy, or other), revised international prognostic scoring system (IPSS-R) score at diagnosis, pre-HCT evolution to acute myeloid leukemia (AML). The odds of day-28 early relapse/MRD positivity was higher in patients treated with low-intensity conditioning compared to high intensity [Odds Ratio (OR) 2.30, 95%CI 1.10 - 4.80, p=0.03].
Patients having 5% or more marrow blasts by morphology pre-HCT were defined as NO-CR. Among patients with less than 5% marrow blasts, regardless of treatment, we defined two groups, cytogenetics positive and cytogenetics negative. The criterion for cytogenetics positivity was the presence of an identifiable abnormality by cytogenetics or FISH (excluding isolated deletion Y). Among patients classified as NO-CR or cytogenetics positive at pre-HCT evaluation, the estimated probability of early relapse/MRD positivity was 57% and 18% among low- and high-intensity conditioned patients, respectively (OR 3.68, 95%CI 1.51 - 9.01, p<0.01). This was a suggestively stronger (interaction p=0.14) effect than among patients classified as cytogenetics negative (OR 1.95, 95%CI 0.49 - 7.77, p=0.34).
In conclusion, the outcome of patients with day-28 post-HCT early relapse or MRD-positive MDS or AML evolving from MDS is very poor. Those patients should be considered for clinical trials or experimental treatments. Patients not in CR or cytogenetics positive at pre-HCT evaluation have a significantly higher risk of early relapse/MRD positivity at day-28 when treated with low-intensity conditioning.
patient characteristics
| Characteristic | |
|---|---|
| Patients, no. | 258 |
| Male / female, no. (%) | 150 (58) / 108 (42) |
| Median age at diagnosis, years (range) | 56 (5.4 - 74.7) |
| Median age at HCT, years (range) | 56.7 (6.1 - 76.3) |
| Etiology | |
| Primary, no. (%) | 184 (71) |
| Secondary, no. (%) | 74 (29) |
| Disease duration | |
| Median time between diagnosis and HCT, months (range) | 7 (0.5 - 82.4) |
| IPSS-R risk at diagnosis | |
| Very low, no. (%) | 10 (4) |
| Low, no. (%) | 46 (18) |
| Intermediate, no. (%) | 67 (26) |
| High, no (%) | 70 (27) |
| Very high, no (%) | 61 (24) |
| Not evaluable, no (%) | 4 (2) |
| Donors | |
| HLA-identical family members, no. (%) | 104* (40) |
| Haploidentical, no. (%) | 2 (1) |
| Matched unrelated, no. (%) | 106 (41) |
| Mismatched unrelated, no. (%) | 32 (12) |
| Cord, no. (%) | 14 (5) |
| Source of stem cells | |
| Peripheral blood, no. (%) | 219 (85) |
| Marrow, no. (%) | 25 (10) |
| Cord, no. (%) | 14 (5) |
| AML evolution before HCT | |
| Yes, no. (%) | 59 (23) |
| No, no. (%) | 199 (77) |
| Characteristic | |
|---|---|
| Patients, no. | 258 |
| Male / female, no. (%) | 150 (58) / 108 (42) |
| Median age at diagnosis, years (range) | 56 (5.4 - 74.7) |
| Median age at HCT, years (range) | 56.7 (6.1 - 76.3) |
| Etiology | |
| Primary, no. (%) | 184 (71) |
| Secondary, no. (%) | 74 (29) |
| Disease duration | |
| Median time between diagnosis and HCT, months (range) | 7 (0.5 - 82.4) |
| IPSS-R risk at diagnosis | |
| Very low, no. (%) | 10 (4) |
| Low, no. (%) | 46 (18) |
| Intermediate, no. (%) | 67 (26) |
| High, no (%) | 70 (27) |
| Very high, no (%) | 61 (24) |
| Not evaluable, no (%) | 4 (2) |
| Donors | |
| HLA-identical family members, no. (%) | 104* (40) |
| Haploidentical, no. (%) | 2 (1) |
| Matched unrelated, no. (%) | 106 (41) |
| Mismatched unrelated, no. (%) | 32 (12) |
| Cord, no. (%) | 14 (5) |
| Source of stem cells | |
| Peripheral blood, no. (%) | 219 (85) |
| Marrow, no. (%) | 25 (10) |
| Cord, no. (%) | 14 (5) |
| AML evolution before HCT | |
| Yes, no. (%) | 59 (23) |
| No, no. (%) | 199 (77) |
*2 HLA identical parents.
Abbreviations: HCT = allogeneic hematopoietic cell transplantation; IPSS-R = Revised International Prognostic Scoring System; HLA = human leukocyte antigen; AML = acute myeloid leukemia.
Overall survival curve from day-28 post-HCT
Abbreviations: MRD = minimal residual disease; OS = overall survival.
Overall survival curve from day-28 post-HCT
Abbreviations: MRD = minimal residual disease; OS = overall survival.
Sandmaier:Gilliad: Honoraria; ArevaMed: Honoraria; Jazz Pharmaceutical: Honoraria; Seattle Genetics: Honoraria; Abmit: Research Funding; Bellicum: Research Funding. Deeg:Medac: Research Funding; Medac: Consultancy. Fang:Affymetrix: Research Funding.
