The effect of granulocyte colony‐stimulating factor dose and administration interval after allogeneic hematopoietic cell transplantation on early engraftment of neutrophil and platelet

Abstract Background Hematopoietic stem cell transplantation (HSCT) is one of the treatments for hematologic malignancies. Numerous factors affect the HSCT outcome. The purpose of this study was to investigate the effect of post‐HSCT administration of granulocyte colony‐stimulating factor (post‐G‐CSF) on early neutrophil and platelet engraftment in allogeneic HSCT (allo‐HSCT). Material & methods The study was performed on 76 patients diagnosed with AML and ALL. All patients underwent allo‐HSCT at Taleghani stem cell transplantation center, Tehran, Iran, from February 2016 to December 2018. Chemotherapy regimens based on patients' conditions were selected between myeloablative and reduced‐intensity regimens. Results Statistical analysis revealed that the number of administered G‐CSF units after HSCT was a time‐dependent variable. Statistical analysis before day +11 reported that patients who received G‐CSF <14 units had three times better early neutrophil engraftment than those with G‐CSF ≥14 (CI 95%, AHR = 3.03, p:0.002). CD3+ cells count <318.5 × 106/kg was associated with fast platelet engraftment (CI 95%, AHR 2.28, p:0.01). Conclusion In this study, post‐G‐CSF stimulation was associated with early engraftment in a time‐ and dose‐dependent manner. Administration of G‐CSF beyond 14 units resulted in adverse effects on neutrophil early engraftment. It also appeared that with a reduction in CD3+ cell counts, the likelihood of GVHD decreases, and platelet engraftment occurs earlier. Further investigations in the future are required to determine the factors affecting the process of early engraftment.


Allogeneic hematopoietic stem cell transplantation (allo-HSCT)
is a promising treatment in patients with blood disorders. 1  plasminogen activator (uPA)/uPA-receptor, neutrophil-derived proteases, and complement system components. 13,14 Neutrophil engraftment is defined as the first day of three consecutive days in which the absolute neutrophil count (ANC) is ≥0.5 × 10 9 /L, whereas platelet engraftment is defined as at least seven consecutive days without platelet transfusion in which the platelet count is >20 × 10 9 /L. [15][16][17] To the best of our knowledge, previous literature in this field shows discrepancies about the best time to initiate G-CSF administration after HSCT. Besides, there is no solid evidence suggesting the optimum number of G-CSF units that can accelerate and improve neutrophil and platelet engraftment. Therefore, in this study, we evaluated the effect of post-HSCT G-CSF administration on neutrophil and platelet early engraftment in allo-HSCT patients.

| Chemotherapy regimens
Generally, our patients received conditioning regimens in four categories. Myeloablative conditioning (MAC)-1 regimen consisted of busulfan and cyclophosphamide (Bu/Cy) as 0.8 mg/kg intravenous busulfan (Bu) every 6 h for 4 days and 60 mg/kg cyclophosphamide (Cy) for 2 days. In the MAC-2 (Bu/Flu) regimen, fludarabine (Flu) 30 mg/m 2 once a day for 5 days was prescribed instead of cyclophosphamide. In the MAC-3 (Bu/Flu/ ATG) regimen, 60 mg/kg of anti-thymocyte globulin (ATG) was added to the MAC-2 regimen, but the total dose of busulfan was

| Serum glucose level
Both fasting and non-fasting whole-blood glucose for each patient was measured daily. The patients were categorized to three groups based on their mean glucose level: a) <100 mg/dl, b) 100-124 mg/ dl, c) ≥125 mg/dl.

| Engraftment evaluation
Hematopoietic recovery was defined based on daily complete blood cell counts. Myeloid engraftment was considered to occur on the first day of three consecutive days in which absolute neutrophil count (ANC) was ≥0.5 × 10 9 /L (15). Platelet engraftment was defined as at least seven consecutive days without platelet transfusion in which the platelet count was >20 × 10 9 /L. 16,17

| Statistical analysis
In the descriptive analysis, the categorical variables with the frequencies and percentages, the normally distributed continuous

| Univariate analysis
The medians of infused CD34+, 'CD3+, and MNC dosage were selected as the cut-off point, and the patients were divided into two groups based on these thresholds for the analysis. In univariate anal-   (Table 2). Also, early platelet engraftment in recipient's blood groups (p-value = 0.14), recipient's BMI (p-value = 0.12), number of injected GCSF after HSCT groups (p-value = 0.06), and chemotherapy type (p-value = 0.03) were statistically significant (Table 3).

| Multiple analysis
For early neutrophil engraftment with backward selection method, the diagnosed disease and the number of G-CSF injected after trans-

| Association of CBC parameters of donors with the infused CD3, CD34, and MNC dosage
As shown in Table 4, the association between donor hemoglobin level on the day of peripheral blood stem cell harvesting and MNC count in apheresis product (known as baseline-PBSC) was nearly The CD34+ cell count and MNC count were increased by increasing donor hemoglobin level in pre-PBSC (Figure 1).

| DISCUSS ION
Allo-HSCT has been confirmed as a treatment for many hematologic disorders. 20 Stem cell engraftment depends on various factors such as MNC and CD34+ cell count, HLA-matching, ABO compatibility, age, sex, and BMI of donor and patient, disease status before HSCT, CMV reactivation, conditioning regimens, GVHD prophylaxis regimens, and various unknown factors. In this study, we aimed to determine the effects of these factors especially post-G-CSF administration on neutrophil and platelet early engraftment.
Pre-transplant G-CSF is used as a part of regimens to increase stem cells in peripheral blood for harvesting. The most common problem that affects patients after HSCT is infection due to prolonging the neutropenic phase. 21 G-CSF administration after HSCT could reduce the neutropenic phase duration, decrease the risk of life-threatening infections, and accelerate the granulocyte recovery. 22 We prescribed G-CSF as intravenous (IV) infusion with a dose of 5-10 µg/kg/day on day one after HSCT until the ANC became ≥0.5 × 10 9 /L for three consecutive days. Our results showed that the median day of neutrophil and platelet engraftment was 10 and in the G-CSF group relative to that of the placebo group (13 days and 15.5 days). 23 Another report showed that starting 5 µg/kg G-CSF administration on day +1 could reduce neutrophil engraftment time from 16 to 10.5 days in the G-CSF group compared to the control group. Furthermore, the median time of hospitalization (18 vs. 24 days) and days on broad-spectrum antibiotics (11 vs. 15 days) were significantly reduced with the administration of G-CSF after HSCT. 24 In a similar study, lenograstim (another recombinant human G-CSF) decreased the length of myeloid recovery from 12.5 to 9 days after HSCT. 25 However, documents were controversial about the time of platelet engraftment. We reported that platelet engraftment occurred 1 day later than neutrophils. In Bishop's study, filgrastim, used as a granulocyte growth factor, ameliorated the platelet engraftment for 2.5 days (13 vs. 15.5 days). 23  SDF-1 (CXCL12) is an important member of the chemokine family, which plays a vital role in stem cell homing. 28 In the HSCT process, conditioning regimen and chemotherapy before transplantation increase the secretion of SDF-1 from osteoblast in the endosteal niche. The reciprocal link between SDF-1 and CXCR4 begins HSC homing and engraftment after transplantation. 28,29 Our study revealed that injection of more than 14 G-CSF units in 11 days after HSCT could be associated with a reduced occurrence of early neutrophil engraftment. Thompson et al. 30 found that in patients undergoing auto-HSCT, starting G-CSF on the same day of HSCT caused faster recovery of hematological parameters. In addition, in patients undergoing HSCT, entering severe pancytopenia phase before transplantation increases opportunistic infection risk. 31 Hence, it is recommended that G-CSF should be used on the first day of cell transplantation. Accordingly, we started the G-CSF at day +1 to prevent the possibility of infection.
However, in other study, it was noted that there was no significant difference between the time of starting G-CSF at day 0, +5, +10  Moreover, our study indicated that AML patients had a lower risk of early neutrophil engraftment than ALL group. To the best of our knowledge, patients with advanced stages of the disease are more likely to receive multi-stage chemotherapy regimens, so they may have minimal residual leukemic cells before the time of HSCT that is called partial remission (PR). These patients will probably relapse after a while, and these conditions are more common in AML patients with a poorer prognosis than ALL patients. 34 It was mentioned that several cycles of chemotherapy and also the leukemic cells that survived after chemotherapy caused the delay in HSC engraftment in AML patients. 35,36 Interestingly, the time of platelet engraftment in our study was independent of G-CSF doses before day +11. However, Shimoda

TA B L E 3 (Continued)
et al. 37 reported that G-CSF injection triggered platelet aggregation and led to transient thrombocytopenia in healthy donors before the apheresis process.
We could not conclude that CD34+ cell count had an association with the early neutrophil and platelet engraftment. Regardless, a high dose of CD34+ was nearly associated with rapid engraftment of neutrophils and platelets. 38,39 It has been reported that the CD34+ cell dose between 2 and 4 × 10 6 /kg of recipient weight was necessary for rapid engraftment. 39 previous reports showed the critical role of CD3+ T cells in sustaining the engraftments. 40,41 The role of CD3+ cells in the occurrence of GVHD has already been proven. 42 High numbers of CD3+ cells, HSCs, and monocytes in the graft limit the rate of graft rejection. 43 In a study with 256 patients, Urbano-Ispizua et al. 41 concluded that the patients with less than 0.2 × 10 6 / kg CD3+ cells in the graft were at risk of graft failure. Rauofi et al. 44 stated that the rate of cGVHD was decreased in patients with CD3+ <365 × 10 6 /kg compared to patients with CD3+ >365 × 10 6 /kg.
Contrarily, Chang et al. 45 declared that higher counts of CD3+ cells (median, 1.64 × 10 8 /kg) might be a guarantee to sustain the engraftment. Likewise, we demonstrated that patients who received <318.5 × 10 6 /kg CD3+ cells had early platelet engraftment, probably due to the low incidence of GVHD. Abbreviation: Pre-PBSC, preliminary peripheral blood cell count before stem cell collection. r:correlation coefficient, it shows how one variable affects another; p: P value, this item indicates the significance of the data relationship.

TA B L E 4
Association characteristics of donors with CD3, CD34, and MNC in apheresis product

F I G U R E 1 Association pre-PBSC hemoglobin with CD34 and MNC
Our results showed that the amount of donor hemoglobin before stem cell isolation was significantly related to the count of CD34+ cells and MNCs in the graft product. Since the apheresis device is adjusted based on the physiological conditions of the donor before the apheresis procedure, including age, blood pressure, hemoglobin level, etc., it is possible that in donor with higher hemoglobin, the volume of blood entering the device set higher and therefore, there might be more amount of isolated CD34+ cells and MNCs per circulation.
The findings of this study showed that administration of G-CSF in the first days after HSCT could accelerate the time of neutrophil engraftment. In fact, we suppose that higher doses of G-CSF before the 11 days post-HSCT result in faster hematologic reconstitution.