The correlation between CD34+ cell dose and DFS in matched-related donor PBSC allo-SCT

The effect of CD34⁺ cell dose on patient outcome in allogeneic stem cell transplantation (allo-SCT) using peripheral blood stem cells (PBSCs) as the stem cell source has not yet been established. Whilst a low CD34⁺ cell dose has been reported to lead to engraftment failure, delayed engraftment and inferior outcome, conflicting reports have been published for transplantations using a high CD34⁺ cell dose. To evaluate the impact of infused CD34⁺ cell doses on PBSC allo-SCT matched-related donor outcomes, Yasuhisa Yokoyama, University of Tsukuba, Tsukuba, JP, and colleagues conducted a retrospective analysis using a large cohort of patients from a Japanese nationwide registry database.

Study design

• Patients (N = 851) who received PBSC transplantation from a human leukocyte antigen (HLA)-matched related donor were included
  • HLA-matched sibling transplant: n = 821
  • HLA-matched related donors (not siblings): n = 30
• Patients who underwent in vivo or ex vivo T-cell depletion were excluded
• Hematological malignancies included: acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), myelodysplastic syndromes (MDS) or chronic myeloid leukemia (CML)
  • Low-risk disease was defined as: AML in first complete remission (CR1), untreated MDS, MDS in CR1 and CML in chronic or accelerated phase at the time of transplant
  • High-risk disease was defined as all other disease statuses
• Patients were divided into two groups by CD34⁺ cell dose:
  • High-CD34⁺: ≥ 4.5 × 10⁶/kg (n = 265)
    • Further subdivided into > 8 × 10⁶/kg (n = 43) and 4.5–8.0 × 10⁶/kg (n = 222)
  • Low-CD34⁺: < 4.5 × 10⁶/kg (n = 586)
• Evaluated outcomes: disease-free survival (DFS), rates of graft-versus-host disease (GvHD), overall survival (OS), cumulative incidence of relapse (CIR) and non-relapse mortality (NRM)

Patient characteristics

Given as high-CD34⁺ vs low-CD34⁺

• Key patient characteristics with significant differences between CD34⁺ cell doses are shown in Table 1
• Significant differences between donor age and donor/recipient sex matching between high- CD34⁺ and low-CD34⁺ groups is likely due to a lower CD34⁺ cell harvest from female or older donors
• Median recipient age, years (range): 42 (16–74) vs 47 (16–73)
• Most patients had AML (59.2% vs 55.8%) or ALL (22.3% vs 23.9%), a performance status between 0 and 1 (89.1% vs 91.3%), received myeloablative conditioning (73.2% vs 69.7%) and received cyclosporine as GvHD prophylaxis (87.4% vs 88.9%)
• There was an even split of patients with low-risk and high-risk disease between CD34⁺ cell dosing groups (56.2% vs 57.0% and 43.8% vs 43.0%, respectively)
• Median dose of infused CD34⁺ cells: 3.5 × 10⁶/kg (range, 0.6–21.1)
• Median follow-up for survivors: 1,330 days (range, 40–2,773)

Table 1. Key patient characteristics with statistically significant differences between CD34⁺ cell doses

 Results

• Engraftment: Patients with high-CD34⁺ had significantly earlier recovery for both neutrophils and platelets
• DFS and OS: Overall, DFS and OS at three years was not different by CD34⁺ cell dose (Table 2), however patients with low-risk disease that received a high-CD34⁺ cell dose had an improved DFS, but without a difference in OS
  • In patients with high-risk disease, there were no differences by CD34⁺ cell dose
• CIR and NRM: In the total cohort, CIR and NRM were not significantly affected by CD34⁺ cell dose (Table 2). In patients with low-risk disease, high-CD34⁺was associated with a significantly lower CIR but without a difference in NRM. There were no significant differences were observed by CD34⁺ cell dose in patients with high-risk disease

Table 2. Analysis of outcomes by CD34⁺ cell dose

CIR, cumulative incidence of relapse; DFS, disease-free survival; NRM, non-relapse mortality; OS, overall survival
* Values indicated in bold are statistically significant

GvHD rates

The cumulative incidence of acute GvHD (aGvHD) at Day 100, Grade II–IV or III–IV, and the incidence of chronic GvHD (cGvHD) at three-years, overall or extensive, did not differ by CD34⁺ cell dose or by disease risk as shown in Table 3.

Table 3. GvHD risk by acute vs chronic and disease risk

GvHD, graft-versus-host disease; aGvHD, acute graft-versus-host disease; cGvHD, chronic graft-versus­-host disease

Multivariate analysis in patients with low-risk disease

Multivariate analysis found that high-CD34⁺ cell dose and development of Grade III–IV aGvHD were significantly associated with DFS whilst Grade III–IV aGvHD was also associated with OS. No other factors such as performance status, conditioning regimen, or GvHD prophylaxis were associated with DFS or OS.

• DFS:
  • High-CD34⁺vs low-CD34⁺: hazard ratio (HR) = 0.720 (95% CI, 0.520–0.998), p = 0.048
  • Development of Grade III–IV aGvHD vs Grade 0–II: HR = 1.643 (95% CI, 1.089–2.480), p = 0.018
• OS:
  • Development of Grade III–IV aGvHD vs grade 0–II: HR = 1.782 (95% CI, 1.170–2.713), p = 0.007

Outcome analysis for transplantation of excessive CD34⁺ cell count (> 8 × 10⁶/kg vs 4.5–8.0 × 10⁶/kg), found no significant associations in relation to three-year DFS (p = 0.713), OS (p = 0.791), overall cGvHD (p = 0.179) or extensive cGvHD (p = 0.489). However, excessive CD34⁺ cell dose (> 8.0 × 10⁶/kg) led to significantly higher levels of aGvHD:

• Grade II–IV aGvHD at Day 100: 47.6% vs 25.6%, p = 0.0360
• Grade III–IV aGvHD at Day 100: 28.6% vs 11.2%, p = 0.0361

Conclusion

While no correlation between outcome and CD34⁺ cell dose was found for high-risk patients, patients with low-risk disease receiving high-CD34⁺ cell dose had a reduced risk of relapse and improved DFS without a concomitant increase in GvHD incidence. However, an excessive dose of CD34⁺ was found to increase the first of Grade II–III and II–IV aGvHD and should be avoided. The importance of disease stratification is also shown in this analysis to ensure patients receive the appropriate dose of CD34⁺ cells.

Therefore, the authors of this retrospective study recommend a CD34⁺ cell dose of 4.5–8.0 × 10⁶/kg for patients with low-risk diseases receiving PBSC as donor source in HLA-matched related transplantation.