RIC allo-SCT vs HMA or best supportive care in older patients with high-risk MDS

To date, allogeneic stem cell transplantation (allo-SCT) remains the only curative treatment for patients with myelodysplastic syndrome (MDS) and provides better outcomes in patients with high-risk MDS when compared with hypomethylating agents (HMAs).¹ Although, early allo-SCT is widely offered to younger patients, its potential benefit versus non-transplant therapy has not been fully addressed in older patients with high-risk MDS.¹ To address this, the Blood and Marrow Transplant Clinical Trials Network (BMT CTN) designed the phase II randomized trial 1102 (NCT02016781), to assess the benefits of reduced intensity conditioning (RIC) allo-SCT vs non-transplant therapies (HMAs or best supportive therapy) in older patients with high-risk MDS.¹

The first results from this trial were presented at the 62nd American Society of Hematology (ASH) Annual Meeting and Exposition and are summarized below.

Study design

• Randomized, open-label, multicenter, biological assignment, phase II trial in patients aged 50–75 years with International Prognostic Scoring System (IPSS) high- or intermediate-2-risk de novo MDS
• Enrolled patients (N = 384) were assigned to arms based on the availability of a fully human leukocyte antigen (HLA)-matched related (6/6) or unrelated (8/8) donor, into:
  • Donor arm (n = 260): patients expected to undergo RIC allo-HCT within 6 months of enrollment. The specific RIC regimen used was at the discretion of each treating physician, or;
  • No donor arm (n = 124): patients receive either HMA or best supportive care at the discretion of the treating physician
• Patients with prior HMA therapy were allowed but those with mismatched or haploidentical donors or umbilical cord blood were excluded
• Primary endpoint was adjusted 3-year overall survival (OS)
• Secondary outcomes include leukemia-free survival (LFS), number of relapsed patients in the donor arm, cost-effectiveness, and quality of life (QoL) assessments
• The arms were well-balanced with regards to baseline patient characteristics (Table 1)

Table 1. Patient baseline characteristics from the BMT CTN 1102 trial¹

Baseline characteristic	Donor arm (n = 260)	No donor arm (n = 124)	Total (N = 384)
BMT CTN, Blood and Marrow Transplant Clinical Trials Network; CR, complete response; HCT-CI, Hematopoietic Cell Transplantation Comorbidity Index; HMAs, hypomethylating agents; IPSS, International Prognostic Scoring System; KPS, Karnofsky Performance Score; MDS, myelodysplastic syndrome; PR, partial response. *MDS duration from diagnosis to enrolment.
Age   Median, years   ≥ 65, n (%)	66.3 155 (59.6)	67.3 80 (64.5)	66.7 235 (61.2)
Males, n (%)	165 (63.5)	76 (61.3)	241 (62.8)
KPS, n (%)   90–100   < 90	99 (55.0) 81 (45.0)	35 (41.7) 49 (58.3)	134 (50.8) 130 (49.2)
Median MDS duration,* months (range)	2.5 (0.2–182.3)	2.2 (0.3–211.6)	2.3 (0.2–211.6)
IPSS score, n (%)   Intermediate-2   High-risk	173 (66.5) 87 (33.5)	81 (65.3) 43 (34.7)	254 (66.1) 130 (33.9)
Response to HMAs, n (%)   CR   PR   No response   No prior HMAs   Unknown	10 (3.8) 46 (17.7) 79 (30.4) 88 (33.8) 37 (14.2)	7 (5.6) 23 (18.5) 42 (33.9) 33 (26.6) 19 (15.3)	17 (4.4) 69 (18.0) 121 (31.5) 121 (31.5) 56 (14.6)
Donor type, n (%)   Matched unrelated	180 (69.2)	—	—
HCT-CI, n (%)   0   1   2   3+   Missing	41 (15.8) 31 (11.9) 35 (13.5) 98 (37.7) 55 (21.2)	—	—

Results

• At data cutoff:
  • Donor arm (n = 260): 62 patients completed the 3-year follow-up, 71 are currently alive and on study, 125 died
  • No donor arm (n = 124): 14 patients completed the 3-year follow-up, 23 are alive and on study, 86 died
• Primary outcomes are shown in Table 2. Overall, the intention-to-treat (ITT) analysis (N = 380) showed an absolute improvement in OS of 21.3% in the donor arm vs the no donor arm (p = 0.0001). In the sensitivity analysis, where patients who died or withdrew within the 90 days donor search period were excluded, the 3-year adjusted OS was similarly higher in the donor arm (p = 0.0004)
• Secondary outcomes are also shown in Table 2. From the ITT analysis, a statistically significant absolute improvement in 3-year LFS of 15.2% was observed in the donor arm vs the no donor arm (p = 0.003). This statistical significance between the arms was also maintained in the sensitivity analysis (p = 0.0074)
• The non-compliance trial rate was 26%:
  • Donor arm:
    • 44 patients did not undergo allo-SCT due to disease progression, subject preference, progressive comorbidity, or donor/insurance issues
    • 26 patients underwent myeloablative conditioning allo-SCT
  • No donor arm:
    • 31 patients underwent transplantation, including nine who found a donor after the 90-day search window
• Taking the noncompliance cases into consideration, the as-treated analysis, compares RIC allo-SCT with best supportive care in patients who actually underwent allo-SCT. The results of this analysis are shown in Table 2. Overall, this analysis indicated an even greater benefit of RIC allo-SCT in terms of OS and LFS in patients with MDS

Table 2. Primary and secondary outcomes from the BMT CNT 1102 trial¹

Outcome	Donor arm	No donor arm	p value
BMT CTN, Blood and Marrow Transplant Clinical Trials Network; CI, confidence interval; ITT, intention-to-treat; LFS, leukemia-free survival; OS, overall survival. *In this analysis, patients assigned to the no donor arm who died or withdrew within the first 90 days of the study (donor search window) were excluded †In this analysis, RIC allo-SCT versus best supportive care is compared, since patients in the donor arm (n = 44) did not undergo allo-SCT due to disease progression, subject preference, progressive comorbidity, or donor/insurance issues, and 26 patients underwent myeloablative allo-SCT. In the no donor arm, 31 patients underwent transplantation, including nine patients who found a donor after the 90-day search window.
ITT analysis	(n = 260)	(n = 124)
3-year adjusted OS estimate, % (95% CI)	47.9 (41.3-54.1)	26.6 (18.4-35.6)	Absolute improvement: 21.3%; p = 0.0001
3-year LFS estimate, % (95% CI)	35.8 (29.8-41.8)	20.6 (13.3-29.1)	Absolute improvement: 15.2%; p = 0.003
Sensitivity analysis*
3-year adjusted OS estimate, %	48.0	28.1	0.0004
3-year LFS estimate, %	35.9	21.8	0.0074
As-treated analysis†	(n = 190)	(n = 85)
3-year OS estimate (%, 95% CI)	47.4 (40.1–54.4)	16.0 (8.4–25.9)	Absolute improvement: 31.4%; p < 0.0001
3-year LFS estimate (%, 95% CI)	39.3 (32.2–46.4)	10.9 (4.4–21.0)	Absolute improvement: 28.4%; p < 0.0001

• Further subgroup analysis, showed that the following had no effect on OS or LFS:
  • Response to prior HMAs (either no or any response)
  • Prior HMA treatment
  • Age (≤ 65 or > 65 years)
  • MDS duration (< 3 or ≥ 3 months)
  • IPSS intermediate-2 or high-risk
  • Revised IPSS score very low, low, intermediate, high, or very high
• No clinically meaningful differences in QoL scores were identified between the arms
• The cost-effectiveness analysis is anticipated

Conclusions

The results of this phase II trial indicate that the early use of RIC allo-SCT in older patients (50–75 years) with advanced MDS results in improved outcomes when compared with HMAs or best supportive care. Irrespective of age, RIC allo-SCT led to significantly improved 3-year OS and LFS rates without any associated posttransplant QoL detriments. According to the trial investigators, early allo-SCT referral should be considered as a standard of care not only in younger patients but also in those with advanced age and high-risk de novo MDS.