Background: Our center has reported that the infused viable CD34+ cell dose of the dominant unit determines the speed & success of neutrophil recovery after dCBT (Purtill et al, Blood 2014). However, how unit characteristics affect lymphocyte recovery has not been determined, & the effect of both delayed neutrophil & lymphocyte recoveries on transplant-related mortality (TRM) requires investigation.

Methods: The association between 1) dominant unit characteristics (TNC, CD34+, CD3+ cell doses & 8-allele HLA-match) & lymphocyte recovery, & 2) the effect of both neutrophil & lymphocyte recoveries on TRM were analyzed in 98 adult myeloablative dCBT recipients (median age 42 years, range 16-69) transplanted for hematologic malignancies. The analysis was restricted to adults given the significant differences in patients (pts), grafts, & neutrophil recovery in children.

Results: The cumulative incidence of day 45 neutrophil engraftment ≥ 0.5 x 109/L was 96% (95%CI:89-99) with a median recovery of 25 days (range 12-43). The day 45 cumulative incidence of lymphocyte engraftment ≥ 0.2 x 109/L was also 96% (95%CI:89-99, median 25 days, range 12-59). Dominant unit infused viable CD34+ cell dose determined both neutrophil (as previously reported) & also lymphocyte recovery. To analyze if the speed & success of neutrophil & lymphocyte recovery are important in determining subsequent TRM in engrafting pts, we performed a day 45 landmark analysis of the 93 patient subset who achieved neutrophil engraftment & were alive 45 days post-dCBT (5 pts with graft failure or early death were excluded). In univariate analysis, a trend toward increased day 45-180 TRM was observed among the 51 pts with delayed neutrophil engraftment > 25 days [HR 2.71 (95%CI 0.88-8.42), p = 0.084], but there was no association between the absolute neutrophil count on day 45 & subsequent TRM. Conversely, while the time to lymphocyte recovery ≥ 0.2 x 109/L was not significant, the day 45 lymphocyte count correlated significantly with day 45-180 TRM. A lymphocyte cut-point of 0.3 x 109/L was identified below which TRM risk was markedly increased: 17 dCBT recipients with a day 45 lymphocyte count < 0.2 had a TRM HR 12.04 (95%CI: 4.33-33.52), p < 0.001 (Figure). Age > 40 years & recipient CMV seropositivity were also associated with a trend toward increased day 45-180 TRM in univariate analysis whereas regimen intensity, disease status, & dominant unit HLA-match were not. Patient demographics (age, weight, disease stage, CMV serostatus, & conditioning intensity) & dominant unit characteristics (infused TNC, viable CD34+, CD3+ cell doses) were similar in pts with day 45 lymphocyte counts > or < 0.2 x 109/L. Low lymphocyte count < 0.2 x 109/L was not explained by differences in corticosteroid therapy. The causes of day 45-180 TRM in pts with low day 45 lymphocytes were organ failure (n = 5) & acute GVHD (n = 4) with one patient dying of infection. On multivariate analysis, day 45 lymphocyte count < 0.2 x 109/L was independently associated with a greater day 45-180 TRM risk (p < 0.001, Table) while a trend was observed for delayed neutrophil engraftment (p = 0.069). Recipient age > 40 years & CMV serostatus were not significant (p = 0.196 & p = 0.418, respectively). The relationship between lymphocyte recovery & TRM was maintained when the 20 pts on therapy for acute GVHD by day 45 were excluded.

Conclusion: Both early neutrophil & lymphocyte recovery are critically important factors in the success of adult myeloablative dCBT. While it is well appreciated that early failure of neutrophil recovery is associated with a very high mortality rate, this analysis of pts engrafting with neutrophils suggests a low lymphocyte count (<0.2) at day 45 was associated with a high risk of early TRM and was an independent and more powerful TRM risk factor than the speed of prior neutrophil recovery. The exact mechanisms that underlie this observation require further investigation but these findings emphasize the importance of considering both neutrophil and lymphocyte recovery in the assessment of the quality of early CB engraftment.

Table.
VariablenHR (95%CI)p
Speed of ANC recovery < 25 days 42 Reference  0.069 
≥ 25 days 51 2.94 (0.92-9.39) 
Lymphocyte count at day 45 > 0.2 x 109/L 76 Reference  < 0.001 
< 0.2 x 109/L 17 12.10 (4.21-34.81) 
VariablenHR (95%CI)p
Speed of ANC recovery < 25 days 42 Reference  0.069 
≥ 25 days 51 2.94 (0.92-9.39) 
Lymphocyte count at day 45 > 0.2 x 109/L 76 Reference  < 0.001 
< 0.2 x 109/L 17 12.10 (4.21-34.81) 

* Recipient age and CMV serostatus were included in the initial model but removed using a backward conditional method.

View Large

Figure 1
Figure 1.
View largeDownload slide
Disclosures

No relevant conflicts of interest to declare.

Topics:
lymphocytes, transplantation, umbilical cord blood transplantation, brachial plexus neuritis, graft-versus-host disease, acute, human leukocyte antigens, tissue transplants, glucocorticoids, hematologic neoplasms, infections

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

© 2014 by The American Society of Hematology
2014
Sign in via your Institution