Relationship between immune status after ATG treatment and PNH clone evolution in patients with severe aplastic anemia

Abstract Objectives To investigate the relationship between immune status and paroxysmal nocturnal hemoglobinuria (PNH) clonal evolution of severe aplastic anemia (SAA) patients who received anti‐human thymocyte globulin (ATG) treatment. Methods The clinical data of 102 SAA patients who received ATG were collected and retrospectively analyzed. The remission rate, remission time, response rate, hematopoietic, and immune status were compared. Malignant clones were also observed. Results The remission rate of the group with PNH clones appeared after treatment was significantly higher than the group without PNH clones. The response rate at 12 months of the groups with PNH clones was significantly higher than the group without PNH clones. The recovery of Hb and Ret % of patients with PNH clones was earlier than the patients without PNH clones. The reduction of percentage of CD8+HLA‐DR+/CD8+ and Th1/Th2 ratio of patients with PNH clones was both earlier than the patients without PNH clones. Six patients developed myelodysplastic syndromes (MDS). Conclusion In SAA patients with PNH clones, the cytotoxic T‐cell function and Th1 cell number recovered more quickly and had better response to IST. A small number of SAA patients with or without PNH clones developed MDS malignant clones.

functionalities. 7 So we retrospectively studied 102 SAA patients treated with anti-human thymocyte globulin (ATG) plus cyclosporine A (CsA), to investigate the relationship between immune status and PNH clonal evolution after treatment.

| Patients
We conducted a retrospective analysis in 102 patients who re- British Council for Standardization in hematology aplastic anemia treatment guidelines. 8 Treatment efficacy was determined based on the Camitta standard published in 1979. 9 The patients were all detected with complete blood count (CBC), PNH clone, liver and kidney functions, bone marrow smear, bone marrow (BM) biopsy, the immune parameters including Dendritic Cells (DC), T-cell, Th cell and activated CD8 + T-cell percentage at 0, 3, 6, 12, 24, and 36 months. PNH + was defined as the proportion of CD59 − on granulocytes ≥5%. Since the number of counting cells is 50,000, the accuracy of 0.1% cannot be achieved. It takes more than one million to reach the accuracy of 0.1%. The clinical characteristics of all the patients were showed in Table 1. The median follow-up time was 12 months (1 month-120 months).
No patient had evidence of clinical PNH at SAA diagnosis. So after diagnosis, all patients had received IST including ATG and CsA. The German rabbit ATG with a dose of 3.75 mg/kg body weight was used for a total of 5 days. The dose for CsA is 3-5 mg/ kg. All patients were also treated with hematopoietic growth factors (HGFs). The group with PNH clones appeared after treatment (n = 10)

| Statistical analysis
GraphPad Prism8 statistical software was used for statistical analysis. Results were expressed as mean ± standard deviations. The independent sample mean comparison had been done using the t test (for data with normal distribution) and nonparametric test (for data without normal distribution). Chi-square test was used to compare the rates between the groups. A value of p < 0.05 was considered statistically significant. They were divided into three groups: The group with PNH clones appeared before treatment, the group with PNH clones appeared after treatment, and the group without PNH clones. The three groups before treatment showed no significant statistical significance in white blood cells (WBC, ×10 9 /L), neutrophils (N%), hemoglobin (Hb, g/L), platelets (PLT, ×10 9 /L), Ret%, lactate dehydrogenase (LDH, U/L), and immune indexes (p > 0.05), as shown in Table 1.

| The clinical efficacy comparison of three groups
The remission rates of the three groups were statistically significant (68%, 90%, 56%, p = 0.0175), the remission rate of the group with PNH clones appeared after treatment was significantly higher than the group without PNH clones (p = 0.0031), but the remission time among these groups was not statistically significant (p = 0.1728). Analyzing response rates at different points, the response rate to IST at 12 months was statistically significant among these groups (p = 0.0349), the response rate of the groups with PNH clones appeared before and after treatment (68%, 70%) was significantly higher than the group without PNH clones (40%) (p = 0.0287, 0.0458). The result of response rate at 36 months was consistent with the remission rate. There was no statistical difference at 3, 6, and 24 months after ATG, which is shown in Table 2 and Figure 1.

| Recovery of hematopoietic function and immune state of three groups (3, 6, 12, 24, 36 months)
At 6 months after ATG, the neutrophil percentage (N%) in three groups were statistically significant (p = 0.0067), the N% of the group with PNH clones appeared after treatment and the group without PNH clones was higher than the group with PNH clones appeared before treatment (p = 0.0092, 0.0357). At 12 months after ATG, the Hb in three groups was statistically significant (p = 0.0484), and the Hb of the group with PNH clones appeared before treatment was obviously higher than the group without PNH clones (p = 0.0155). The Ret% in three groups were statistically significant (p = 0.0044), and the Ret% of the group with PNH clones appeared after treatment was obviously higher than the rest two groups (p = 0.0289, 0.0019).
At 6 months after ATG, the proportion of CD8 + HLA-DR + /CD8 + in three groups were statistically significant (p = 0.0027). The proportion of CD8 + HLA-DR + /CD8 + in group with PNH clones appeared after treatment was significantly lower than that in group without PNH clone (p = 0.0057). At 12 months after ATG, the ratio of Th1/ Th2 in three groups were statistically significant (p = 0.0266), the ratio of Th1/Th2 in group with PNH clones appeared before and after treatment were significantly lower than that in group without PNH clones (p = 0.0289, 0.0303). The proportion of CD8 + HLA-DR + /CD8 + in three groups were statistically significant (p = 0.0377), the proportion of CD8 + HLA-DR + /CD8 + in group with PNH clones appeared after treatment was obviously lower than that in group without PNH clones (p = 0.0024), which is shown in Tables 3-7.

| D ISCUSS I ON
Clonal hematopoiesis (CH) was prevalent in aplastic anemia, with CH detected in over two-thirds of AA patients. 10 CH is a non-neoplastic condition that can be associated with diverse genetic alterations, some of which improve cell fitness while others are neutral "passengers". 11 Somatic loss of PIGA is the most common manifestation of CH in AA. 12 AA patients with PNH clones are benign types of bone marrow (BM) failure with immune pathophysiology. 3,13 The mechanism by which the expansion of PNH cells occurs in AA remains unknown. Various experimental models demonstrated that PNH cells have no intrinsic growth advantage. 14 One hypothesis is that the PNH cells which can escape autoimmunity have a proliferative advantage over non-PNH cells by an immune mechanism of selection. 15,16 The potential mecha- Somatic mutations may also lead to enhanced growth by cooperating with PIGA loss. 20 Mortazavi Y et al suggested a process of hypermutation in the phosphatidylinositol glycan A gene (PIG-A) gene in AA stem cells. 21 In addition, their prognostic role remains controversial. There were studies found that an increase in the proportion of PNH clones cells was correlated with a good response to IST among patients with AA. [22][23][24] However, there were also reports found that there was no difference between AA patients presented with or without PNH clones, even hematological response of PNH+ group was lower than the PNH group. 25

F I G U R E 1
The response rate of three groups at different point times (3, 6, 12, 24, 36 months). The response rate to IST at 12 months and 36 months was statistically significant among three groups (p = 0.0349, p = 0.0175). The response rate of the groups with PNH clones appeared before and after treatment (68%, 70%) was significantly higher than the group without PNH clones (40%) (p = 0.0287, 0.0458) at 12 months. The response rate of the group with PNH clones appeared after treatment (90%) was significantly higher than the group without PNH clones (56%) (p = 0.0031). There were no statistical differences in response rate among the three groups at 3, 6, and 24 months after ATG TA B L E 3 The hematopoietic function and immune state of three groups (3 months)

Hematopoietic function and immune state
The group with PNH clones appeared before treatment (n = 18) The group with PNH clones appeared after treatment (n = 10)

TA B L E 5
The hematopoietic function and immune state of three groups (12 months)

Hematopoietic function and immune state
The group with PNH clones appeared before treatment (n = 18) The group with PNH clones appeared after treatment (n = 10) with ATG combined with cyclosporine can reach 60%-70%. 27 In our study, patients with PNH clones obtained a high remission rate, especially those with PNH clones appeared after treatment. The remission rate of the group with PNH clones appeared before treatment was 68%, higher than the group without PNH clones (56%), but there was no statistical significance, which was considered to be related to the large deviation of the sample size in this study, and it should be observed after expanding the sample size. At 12 months after ATG, we found that the recovery of Hb and Ret% of patients with PNH clones were earlier than the patients without PNH clones.
But the platelet recovery has no significant difference between the three groups, platelet recovered for a long time and very few patients were observed at the later stage maybe the major cause. In terms of immune status recovery, the percentage of activated CD8+ it is should used widely in these SAA patients for further study of the mechanism of PNH clones and malignant clones.

ACK N OWLED G EM ENTS
This work was supported by the National Natural Science

CO N FLI C T O F I NTE R E S T
All authors report no conflicts of interest.

AUTH O R S ' CO NTR I B UTI O N S
Rong Fu designed the research and revised the manuscript. Honglei

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.