The number of neutrophils present in graft-versus-host disease (GvHD) lesions strongly correlates with the severity of intestinal GvHD,¹ but it is still unclear whether a difference in the speed of neutrophil recovery, after allogeneic hematopoietic cell transplantation (allo-HCT), has some influence on posttransplant complications and prognosis.

In a retrospective study, recently published in Transplant International, Takashi Nagayama and colleagues evaluated the influence of the speed of neutrophil recovery on the incidence of acute GvHD (aGvHD), in adult patients who underwent a first unrelated bone marrow transplantation.²

We hereby present a summary of their findings.

Study design

Patients included in the study were ˃ 17 years old and underwent a first unrelated bone marrow transplantation, with granulocyte colony-stimulating factor (G-CSF) support, between January 2009 and December 2018.

Treatment

Conditioning regimens:

• Myeloablative conditioning (MAC) regimens included: total body irradiation (TBI) > 8 Gy, melphalan > 140 mg/m², or oral busulfan (BU) > 8 mg/kg (> 6.4 mg/kg IV)
• Reduced-intensity conditioning (RIC) regimens consisted of fludarabine-based regimens, such as fludarabine combined with BU or melphalan

GvHD prophylaxis:

• Continuous infusion of tacrolimus (TAC) combined with short-term methotrexate (MTX)
  • MTX: 10–15 mg/m² on Day 1, 7–10 mg/m² on Days 3 and 6, and an optional dose on Day 11
• In some cases of human leukocyte antigen (HLA) 1-locus mismatched transplantation, antithymocyte globulin (ATG) was administered at 1.25 mg/kg per day on Days 4 and 3

G-CSF was started on 7 days after allo-HCT in all patients, except for two patients who started G-CSFs on Day 1, for accelerating bone marrow recovery.

Defining the cutoff for neutrophil recovery slope:

• The median neutrophil slope (N slope) for all patients was 205.5/µl/day (range, 26.4–7574). The cutoff value of N slope for grade II–IV aGvHD, defined by the area under the receiver operating characteristic (ROC) curve, was 207.5/µl/day. Based on these values, a cutoff value of 200/µl/day was determined and patients were then classified into low and high N slope groups according to this cutoff.

Patient characteristics

120 Patients were evaluated in this analysis and baseline characteristics are reported in Table 1.

Table 1. Patient characteristics²

Characteristic	Total N = 120	Low N slope n = 59	High N slope n = 61	p value
AA, aplastic anemia; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; ATG, antithymocyte globulin; CML, chronic myeloid leukemia; HLA, human leukocyte antigen; MAC, myeloablative conditioning; MDS, myelodysplastic syndrome; ML, malignant lymphoma; MTX, methotrexate; RIC, reduced-intensity conditioning; TBI, total body irradiation. *Standard-risk diseases included acute leukemia in first or second complete remission, CML in the first or second chronic phase, lymphoma in complete remission, MDS, AA, myelofibrosis, chronic myelomonocytic leukemia, and multiple myeloma. Other diseases were classified as high-risk. Values indicated in bold are statistically significant.
Age, years, median (range)	50 (18–69)	45 (19–68)	52 (18–69)	0.03
Disease, n (%)
AML	48 (40)	24 (40.7)	24 (39.3)	0.69
ALL	27 (22.5)	16 (27.1)	11 (18)
MDS	20 (16.7)	9 (15.3)	11 (18)
CML	4 (3.3)	2 (3.4)	2 (3.3)
ML	9 (7.5)	4 (6.8)	5 (8.2)
AA	3 (2.5)	0 (0)	3 (4.9)
Other	9 (7.5)	4 (6.8)	5 (8.2)
Disease risk*, n (%)
High	18 (15)	8 (13.6)	10 (16.4)	0.80
Standard	102 (85)	51 (86.4)	51 (83.6)
HLA matching, n (%)
Match	84 (70)	45 (76.3)	39 (63.9)	0.17
Mismatch	36 (30)	14 (23.7)	22 (36.1)
Donor age, n (%)
< 40 years	65 (54.2)	33 (55.9)	32 (52.5)	0.72
≥ 40 years	55 (45.8)	26 (44.1)	29 (47.5)
Day 11 MTX, n (%)
Yes	65 (54.2)	40 (67.8)	25 (41)	0.004
No	55 (45.8)	19 (32.2)	36 (59)
ATG, n (%)
Yes	16 (13.3)	7 (11.9)	9 (14.8)	0.79
No	104 (86.7)	52 (88.1)	52 (85.2)
Conditioning, n (%)
MAC	100 (83.3)	54 (91.5)	46 (75.4)	0.03
RIC	20 (16.7)	5 (8.5)	15 (24.6)
TBI regimen, n (%)
Yes	78 (65)	35 (59.3)	43 (70.5)	0.25
No	42 (35)	24 (40.7)	18 (29.5)
Infuse CD34+ cells, n (%)
< 1.5 × 106/kg	50 (44.6)	30 (55.6)	20 (34.5)	0.04
≥ 1.5 × 106/kg	62 (55.4)	24 (44.4)	38 (65.5)
Transplantation year, n (%)
2009–2013	45 (37.5)	16 (27.1)	29 (47.5)	0.03
2014–2018	75 (62.5)	43 (72.9)	32 (52.5)

Results

aGvHD

The cumulative incidence of Grade II (n = 28) aGvHD was significantly higher in the high N slope group compared with the low N slope group (34.4% vs 11.9%; p = 0.002). Similarly, the incidence for Grade II–IV (n = 37) aGvHD was significantly higher in the high N slope group (44% vs 17%; p < 0.001). In contrast, no significant difference was observed in the cumulative incidence of Grade III–IV (n = 9) and Grade I (n = 36) aGvHD between the two groups.

The impact of neutrophil recovery on aGvHD target organs was analysed showing a significantly higher cumulative incidence of aGvHD for gut and liver in the high N slope group. The cumulative incidence of aGvHD in the high N slope group vs the low N slope group was:

• Gut aGvHD (Stage 1–4; n = 21), 26% vs 10% (p = 0.02)
• Liver aGvHD (Stage 1–4; n = 3), 7.5% vs 0% (p = 0.04)

Using univariate and multivariate analysis, the highest risk of developing Grade II–IV aGvHD was found in the high N slope group. Other significant risk factors were age > 40 years, receiving TBI, and having had allo-HCT between 2009 and 2013. Male sex was significant only in the univariate analysis, while major HLA mismatch was identified as a Grade II–IV aGvHD significant risk factor only in the multivariate analysis. Risk factors are reported in Table 2.

Table 2. Risk factors for Grade II–IV aGvHD in univariate and multivariate analyses²

	n	Cumulative incidence of Grade II–IV aGvHD at day +100, % (95% CI)	Univariate p	HR (95% CI)	Multivariate p
aGvHD, acute graft-versus-host disease; CI, confidence interval; HR, hazard ratio; N, neutrophil; TBI, total body irradiation. Values indicated in bold are statistically significant.
Donor age
< 40	65	18.5 (10.1–28.8)	0.002	1	Reference
≥ 40	55	45.5 (31.9–58.1)		3.63 (1.57–8.40)	0.003
Sex
Male	77	23.4 (14.6–33.3)	0.02	1	Reference
Female	43	44.2 (28.9–58.4)		0.56 (0.27–1.18)	0.13
ABO mismatch
Major mismatch	27	44.4 (25.1–62.2)	0.06	3.05 (1.47–6.31)	0.003
Other	93	26.9 (18.3–36.2)		1	Reference
TBI
Yes	78	38.5 (27.7–49.1)	0.01	2.83 (1.20–6.67)	0.02
No	42	16.7 (7.2–29.4)		1	Reference
N slope
< 200	59	16.9 (8.7–27.6)	< 0.001	1	Reference
≥ 200	61	44.3 (31.5–56.3)		3.73 (1.88–7.42)	< 0.001
Transplantation year
2009–2013	45	46.7 (31.5–60.5)	0.003	3.28 (1.40–7.68)	0.006
2014–2018	75	21.3 (12.9–31.2)		1	Reference

Survival

After a median follow-up of 29 months (range, 1─116), 40 patients died, 16 of them from non-relapse complications. The N slope had no impact on the cumulative incidences of non-relapse mortality (p = 0.32), relapse (p = 0.76), and overall survival (p = 0.65).

Conclusion

In this retrospective study, the authors found an association between a rapid rate of neutrophil recovery following allo-HCT and the increased risk of developing Grade II–IV aGvHD.

However, besides being retrospective, this study presents other limitations such as:

• the inclusion of patients with heterogeneous characteristics and diagnoses
• the cutoff value that could be not appropriate for events not directly related to aGvHD
• the significant increase of aGvHD between 2009 and 2013 (probably due to underuse of ATG in HLA 1-locus mismatched transplantation)

Additional studies are required to demonstrate a clear association between neutrophil recovery and aGvHD.