MRD and survival outcomes in patients with Down syndrome and acute lymphoblastic leukemia

Patients with both down syndrome (DS) and acute lymphoblastic leukemia (ALL) are at an increased risk of relapse and treatment related mortality when compared with patients who have ALL without DS. Whether the increased risk associated with DS is a result of high-risk genetic features in these patients or DS itself causes the poor outcomes has yet to be thoroughly investigated. Michels, et al. analyzed the data of patients with DS and ALL from eight trials and compared the measurable residual disease (MRD) levels along with long-term survival outcomes.¹

Study design

Patients with DS and ALL and non-DS patients with ALL (matched controls) aged 1–23 years were included in this study. MRD level was the primary endpoint examined and this was measured at the end of induction in all trials included. In two of the trials included in this study MRD was also measured at the end of consolidation. Absolute MRD levels were categorized into two groups, low (<0.0001) and high (≥0.0001).

Secondary endpoints included:

• event-free survival (EFS)
• overall survival (OS)
• relapse
• treatment related mortality (TRM)

Results

Patient characteristics

The baseline characteristics for patients included in this study are shown in Table 1. In total for the MRD analysis 543 patients were included, 136 patients with ALL and DS and 407 patients with ALL only. For the long-term outcome only 450 patients were added to this analysis including 119 patients with both ALL and DS and 331 with ALL. Patient characteristics were well matched between groups including with respect to cytogenetic features.

Table 1. Baseline patient characteristics*

MRD

The percentage of patients with high MRD levels was very similar for those with DS and ALL and the matched controls, both at the end of induction (38% vs 39%, respectively; p = 0.88; Figure 1A) and the end of consolidation (6% vs 7%, respectively; p = 0.76; Figure 1B).

Figure 1. Association between MRD category (low/high) and risk factors for patients with ALL and DS, versus matched controls. MRD categories are shown for patients at A end of induction, B end of consolidation, C end of induction in the IKZF1 deleted and IKZF1 wild-type group, and D end of consolidation in the IKZF1 deleted and IKZF1 wild-type group* 

ALL, acute lymphoblastic leukemia; DS, down syndrome; MRD, measurable residual disease.
*Adapted from Michels, et al.¹ 

The impact of an IKZF1 deletion on MRD status was examined in patients and matched controls. The proportion of patients with high MRD levels was found to be similar between the two groups, in both the patients harboring the deletion and wild type IKFZ1. This was true at the end of induction and end of consolidation (Figure 1C and D).

Secondary outcomes

The median follow-up time was 7.2 years. Patients with DS and ALL demonstrated reduced EFS compared with the matched controls (hazard ratio [HR], 2.5; 95% confidence interval [CI], 1.6–3.9; p < 0.0001). OS was also worse in the group of patients with DS than for the matched controls (HR, 3.8 [2.2–6.3]; p < 0.0001).

5-year EFS was:

• 75% (95% CI, 67–83) in patients with DS and ALL
• 88% (95% CI, 84–92) in the matched controls

5-year OS was:

• 77% (95% CI, 70–85) in patients with DS and ALL
• 94% (95% CI, 91–97) in the matched controls

Patients with DS and ALL were more likely to experience TRM than the matched controls with a 5-year post-induction TRM of 12.2% (95% CI, 7.0–18.9) compared with 2.7% (95% CI, 1.3−4.9), respectively. When looking at the 5-year cumulative incidence of relapse, it was slightly higher in the patients with DS and ALL at 11.7% (95% CI, 6.6–18.5) vs 9.1% (95% CI, 6.2–12.7) for the matched controls.

When IKZF1 status was also considered, analysis showed that patients with DS and ALL that had the IKZF1 deletion were more at risk of relapse than the matched controls (HR, 4.3 [1.6–11.0]; p = 0.0028). However, for patients without this deletion there was no difference in risk of relapse between the patients with DS and ALL and the matched controls (HR, 1.0 [0.5–2.1]; p = 0.99).

5-year cumulative incidence of relapse was:

• 37.1% (95% CI, 17.1–57.2) in patients with DS and ALL with IKZF1 deletion (n = 22)
• 13.2% (95% CI, 6.1–23.1) in the matched controls with IKZF1 deletion (n = 65)
• 5.8% (95% CI, 2.1–12.2) in patients with DS and ALL with wild-type IKZF1 (n = 95)
• 8.1% (95% CI, 5.1–12.0) in the matched controls with wild type IKZF1 (n = 265)

For patients with DS and ALL with the IKZF1 deletion the 5-year incidence of relapse was higher in the high MRD group (56.4% [95% CI, 23.1–80.0]; n = 13) compared with the low MRD group (11.1% [95% CI, 0.5–40.6]; n = 9). Results were similar for the IKZF1 wild-type groups.

Conclusion

Patients with DS and ALL were confirmed to have a poorer prognosis than the matched patient controls. This study also demonstrated that there was an increased risk of relapse for patients with DS and ALL which was not exclusively due to an increased prevalence of adverse cytogenetics but may be associated with DS itself. However, as this subset of patients was small, further data is required to confirm this finding. Patients with DS showed increased TRM and demonstrated that there is a need for specialized treatment strategies for patients with DS and ALL.