Appropriate Timing of Allogeneic Bone Marrow Transplant in Patients with Myelodysplastic Syndrome

Introduction:

The is no clear consensus on the appropriate timing of allogeneic bone marrow transplant (Allo-SCT) for patients with Myelodysplastic syndrome (MDS). A study by Cutler et al showed improvement in life expectancy with early Allo-SCT for patients with intermediate-2 and high-risk IPSS scores but without clear guidance on the appropriate timing for Allo-SCT. Herein, we looked at the risk of disease recurrence (RFS) and overall survival (OS) among MDS patients who had Allo-SCT within 6 months Group (A), 6-12 months (Group B), and more than 12 months (Group C) from the time of diagnosis among different IPSS-R categories.

Methods: We used the publicly available dataset from CIBMTR associated with the following publication, “Alternative donor transplantation for myelodysplastic syndromes: haploidentical relative and matched unrelated donors” by Mehta et, to evaluate the difference in outcomes among patients who had Allo-SCT early after diagnosis (within six months) (Group A) VS 6-12 months (Group B) and more than 12 months (Group C). The three groups were matched on age categories (40-50,50-60 and more than 60), sex, donor type (MUD VS Haploidentical), type of MDS (therapy-related VS non-therapy-related), Karnofscky score, HCT-CI score, conditioning regimen, race, cytogenetics, IPSS-R score, CMV status (both donor and recipient) and year of diagnosis using inverse probability of treatment weights. The difference in RFS and OS was calculated using a doubly robust Cox model. All statistical analysis was done using R.

Results:

A total of 604 patients were identified in the dataset. 66% were males. 383 (64%) were diagnosed from 2015-2017 and 220 (36%) from 2012-2014. 72.2% under 50 years old had an Allo-SCT within a year from diagnosis compared to 55.5% older than 60 (P-value= 0.039). 169 (28%), 198 (33%) and 236 (39%) patients had Allo-SCT within 6 months (Group A), 6-12 months (Group B) and more than 12 months (Group C). Appropriate balance was achieved after matching on the previously mentioned variables. An effective sample size of 108 (64%) in Group A, 155 (79%) in Group B and 158 (67%) in Group C were matched. In the matched dataset, 33.3% VS 33.2% VS 33.4% of Group A, B and C were older than 60 (P value=1). One-third of each group in the matched dataset had Allo-SCT from MUD (P value=1). One third of Group A, B and C after matching received HMA compared to 22% VS 27% VS 27% respectively before matching. The median time to relapse was not reached in each of the groups with a P value of 0.21. However, among high and very high IPSS-R group, the median time to relapse was 10 months for Group C (0.95 CI 5.59-NR) compared to NR (0.95 CI 7.6-NR) for Group B and NR for group A (0.95 CI 10.4-NR) with a P-value of 0.097.

In the double-robust Cox model. Patients with high and very high IPSS scores and transplanted more than one year from the diagnosis (Group C) had a 2.75 risk of recurrence compared to those transplanted in less than one year (Group A and B) (P-value of 0.046 (HR=2.75, 0.95 CI 1.014-7.43)). Patients with intermediate and good risk groups had a similar risk of relapse regardless of the timing of Allo-SCT. This was translated into higher mortality among patients with high IPSS scores who were transplanted more than 12 months from the time of diagnosis (HR=2.66, 0.95 CI 1.01-6.98). Hypomethylating agents (HMA) before the transplant were associated with a 45% higher risk of relapse (P-value<0.01). However, there was no interaction between the HMA and the Groups with median time to recurrence NR for Group C , 24 months and 12 months for Groups B and A respectively with a P value=0.16.

Conclusion:

Among high and very high-risk MDS patients who have Allo-SCT with RIC. Delaying the transplant to more than a year from the diagnosis is associated with a higher risk of recurrence. This could be due to selection of resistant clone from multiple prior treatment.