The prognostic significance of POD24 in peripheral T-cell lymphoma

ABSTRACT

Background:

Peripheral T-cell lymphomas (PTCL) are an aggressive group of mature T-cell neoplasms, often associated with poor outcomes, in part, due to frequent relapsed/refractory disease. The objective of this study was to assess the prognostic impact of disease progression within 24 months (POD24) on overall survival (OS) for patients diagnosed with PTCL.

Methods:

A retrospective analysis was conducted on a cohort of patients with newly diagnosed PTCL who underwent chemotherapy at the Affiliated Hospital of Xuzhou Medical University between January 2010 and September 2021. Prognostic assessment was limited to patients who were evaluable for POD24.

Results:

Records were reviewed for 106 patients with PTCL, of whom 66 patients experienced POD24 (referred to as the POD24 group) and 40 patients did not experience POD24 (referred to as the no POD24 group). Significant differences were observed between the POD24 group and the no POD24 group in regard to clinical stage, Eastern Cooperative Oncology Group (ECOG) performance status (PS), International Prognostic Index (IPI) score, lactate dehydrogenase (LDH) levels, β2-microglobulin (β2-MG) levels, prealbumin and albumin levels. Patients in the POD24 group had a significant shorter median OS compared to the no POD24 group (11.9 months vs not reached, respectively; P < 0.001). Non response (NR) to treatment and POD24 were identified as independent negative prognostic factors for survival in patients with PTCL.

Conclusion:

POD24 is a prognostic factor associated with unfavorable outcomes in patients with PTCL and can be used to identify high-risk patients and guide treatment decisions.

KEYWORDS:

• Peripheral T-cell lymphoma
• early disease progression
• POD24
• POD12
• prognostic analysis
• Non-Hodgkin lymphoma
• clinical practice
• overall survival

Introduction

Peripheral T-cell lymphoma (PTCL) is a hematologic malignancy affecting T lymphocytes and accounts for approximately 10–15% of all cases of non-Hodgkin lymphoma (NHL). PTCL is characterized by a substantial heterogeneity between subtypes, frequent progression, and a generally poor prognosis for most subtypes. Indeed, the 5-year OS rates vary across different pathological subtypes of PTCL, typically falling below 50% [1]. The most common subtypes of PTCL include PTCL not otherwise specified (PTCL-NOS), angioimmunoblastic T-cell lymphoma (AITL), extranodal NK/ T-cell lymphoma-nasal type (ENKTL), and anaplastic large-cell lymphoma (ALCL). The standard of care for frontline therapy is evolving; CHOP-like regimens are typically preferred, though modifications with etoposide and brentuximab vedotin are often considered. The international Prognostic Index (IPI) continues to serve as the primary prognostic indicator for assessing PTCL. Nevertheless, survival outcomes for patients with PTCL with identical IPI scores can vary significantly depending on the clinical context, necessitating the development of improved prognostic indices to inform clinical treatment decisions.

Progression of disease within 24 months (POD24) is a prognostic indicator that had been previously reported in follicular lymphoma [2]. Recently, the prognostic value of POD24 has also been observed in various other types of lymphoma, including marginal zone lymphoma (MZL), ENKTL, AITL, and mantle cell lymphoma (MCL) [3–6]. Currently, there is limited literature available on the prognostic significance of POD24 in PTCL [7]. Hence, we conducted an analysis of data from a cohort of patients at our institution to investigate the prognostic significance of POD24 in PTCL, and we further compared the clinical characteristics between patients who experienced POD24 and those who did not.

Materials and methods

Patients

We retrospectively reviewed clinical data from 117 patients with newly diagnosed PTCL admitted to the Affiliated Hospital of Xuzhou Medical University between January 2010 to September 2021. Written informed consent was obtained from each patient. The inclusion criteria were as follows: (1) patients had definitive histopathological diagnosis and newly diagnosed PTCL; (2) histopathologic subtypes included were PTCL-NOS, AITL, ENKTL, ALCL, hepatosplenic T-cell lymphoma (HSTCL), enteropathy associated T-cell lymphoma, or subcutaneous panniculitis-like T-cell lymphoma (SPTCL); (3) patients received at least two courses of frontline treatment at the Affiliated Hospital of Xuzhou Medical University. Patients were excluded if they were censored (i.e. lost to follow-up) or died within 24 months without POD. The final cohort consisted of 67 patients with POD24 (POD24 group) and 41 patients who did not experience POD24 (no POD24 group). The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of the Affiliated Hospital of Xuzhou Medical University (XYFY2023-KL005-01).

Treatment

All patients had indications for frontline treatment and received standard frontline care per institutional guidelines. Nineteen patients received chemotherapy combined with transplantation; 22 patients received chemotherapy combined with radiotherapy; 65 patients received chemotherapy alone. Treatment details are shown in Table 1.

Table 1. First-line therapy regimens for patients with PTCL, subset by subtype.

	Total	AITL	PTCL-NOS	ENKTL	ALCL	Others
Number	106	29	24	33	11	9
First-line treatment
CHOP	46 (43.4%)	17 (58.6%)	9 (37.5%)	5(15.2%)	8 (72.7%)	5 (55.6%)
CHOPE	19 (17.9%)	6 (20.7%)	8 (33.3%)	2 (6.1%)	2 (18.2%)	1(11.1%)
CHOP-like	7 (6.6%)	3 (10.3%)	1 (4.2%)	3 (9.1%)
COP	10 (9.4%)	2 (6.9%)	4 (16.7%)	2 (6.1%)	1 (9.1%)	1(11.1%)
CTOP	6 (5.6%)	1 (3.4%)	2 (8.3%)	1 (3.0%)		2(22.2%)
GDP	3 (2.8%)			3 (9.1%)
DDGP	15 (14.1%)			15 (45.5%)
GemOx	1 (0.9%)			1 (3.0%)
SMILE	1 (0.9%)			1 (3.0%)
Combined RT	22 (20.8%)		3 (12.5%)	18 (54.5%)	1 (9.1%)
Combined HSCT	19 (17.9%)	3 (10.3%)	5 (20.8%)	8 (24.2%)	1 (9.1%)	2(22.2%)

CHOP, cyclophosphamide doxorubicin vincristine prednisolone; CHOPE, CHOP plus etoposide; COP, cyclophosphamide vincristine prednisolone; CTOP, cyclophosphamide pirarubicin vincristine prednisolone; GDP, gemcitabine dexamethasone cisplatin; DDGP, gemcitabine dexamethasone cisplatin L-asparaginase; GemOx, gemcitabine oxaliplatin; SMILE, dexamethasone methotrexate ifosfamide L-asparaginase etoposide; RT, radiotherapy; HSCT, hematopoietic stem cell transplantation.

Clinical characteristics of patients

Clinical information such as age, sex, lactate dehydrogenase (LDH) levels, clinical staging (Ann-Arbor classification), B symptoms, Eastern Cooperative Tumor group (ECOG) performance status (PS), number of extranodal lesions (≥2), and frontline therapy were collected at the time of initial diagnosis. Prior to treatment, hemoglobin (HB) levels, platelet count (PLT), albumin (ALB) levels, prealbumin (PA) levels, β2-microglobulin, and other laboratory indicators were recorded.

The cutoff for follow-up in the study was August 31, 2022, with a median follow-up time of 47.7 (range, 1.4–154.5) months. Overall survival (OS) was defined as the time initial diagnosis to death from any cause. Progression-free survival (PFS) was defined as the interval from clinical diagnosis to the date of disease progression or death. For OS and PFS, patients were censored if alive at last follow up. POD24 was defined as disease progression or relapse within 24 months after diagnosis, and no POD24 was defined as patients without POD within 24 months after diagnosis. POD24 could not be evaluated if the patient was lost to follow-up within 24 months, or if the patient had died in the absence of POD. Similarly, POD12 was defined as progression or relapse within 12 months after diagnosis, and no POD12 was defined as those without POD within 12 months after diagnosis.

Statistical analysis

SPSS 26.0 software (IBM Corp., Armonk, NY, USA) was used for statistical analysis. χ² test or Fisher exact test was used to compare incidence or constituent ratios between groups. Continuous variables were expressed as median (range). PFS and OS were assessed using the Kaplan–Meier method, and the groups were compared using the log-rank test. Cox regression analysis was performed to assess the impact of various baseline prognostic factors on OS. The hazard ratio (HR) was calculated in conjunction with the 95% confidence interval (CI). Variables with a p-value of P ≤ 0.05 in univariate analyses were subjected to multivariate analysis. A p-value of P ≤ 0.05 was considered statistically significant.

Results

Patient characteristics

Among the 117 enrolled patients, a total of 9 individuals were lost to follow-up or died within a 24-months period due to complications such as infections, and 2 individuals were younger than 18 years old (Figure 1). Thus, 106 patients were included in our study, with a median age of 52 years (range, 19–82) at the time of diagnosis. The distribution of different histological subtypes was as follows: PTCL-NOS (n = 24,22.6%); AITL (n = 29,27.4%); ENKTL (n = 33,31.1%); ALCL (n = 1,10.4%). Additionally, EATL, HSTCL, and SPTCL together accounted for 9 cases (8.4%). Of these patients, 66 individuals (62%) experienced POD24 (POD24 group), and 40 patients (38%) did not experience POD24 (no POD24 group). Fifty-two patients (49.1%) experienced POD12, and 54 patients (50.9%) did not. A total of 85 patients responded following initial treatment, resulting in an overall response rate (ORR) of 80.2%. This included 43 patients who achieved complete response (CR) and 42 patients who achieved partial response (PR). The remaining 21 patients (19.8%) were assessed as having stable disease (SD).

Figure 1. Consort diagram for screening PTCL patients.

Patients with POD24 were more likely to have a higher stage, worse ECOG PS, higher IPI score, elevated LDH, low prealbumin levels, low albumin levels, low HB levels, and elevated β2-MG levels compared to the no POD24 group. No significant differences were found in age, platelet (PLT) level, white blood cell count (WBC), neutrophil count, and lymphocyte count between the two groups (Table 2). Moreover, the POD12 group and the no POD12 group also had significant differences in various clinical variables with the exception of age, PLT, WBC, and lymphocyte count. (Table S1).

Table 2. Clinical characteristics of 106 patients with PTCL, all patients and subset by POD24 status.

	All (N = 106)	POD24 (N = 66)	No POD24 (N = 40)	p-value
Age (years)	19–82	22–77	19–82
Median age	52	54	50
Sex Male/Female	63/43	39/27	24/16
Age > 60 years	29 (27.4%)	19 (28.8%)	10 (25.0%)	0.672
Stage III or IV	65(61.3%)	47(71.2%)	18(45.0%)	0.007
ECOG PS 2–4	30(28.3%)	26(39.4%)	4(10.0%)	0.001
LDH > ULN	58(57.4%)	42(66.7%)	16(42.1%)	0.016
IPI 3–5	37(34.9%)	32(48.5%)	5(12.5%)	<0.001
PA < LLN	51(48.1%)	41(62.1%)	10(25.0%)	<0.001
ALB < LLN	65(61.3%)	46(69.7%)	19(47.5%)	0.023
HB < 120 g/L	49(46.2%)	38(57.6%)	11(27.5%)	0.003
PLT < 100 × 10^9/L	14(13.2%)	11(16.7%)	3(7.5%)	0.177
β2-MG > ULN	36(52.2%)	30(69.8%)	6(23.1%)	<0.001
WBC < 3.5 × 10^9/L	14(13.2%)	10(15.2%)	4(10.0%)	0.448
NEC < 1.8 × 10^9/L	12(11.3%)	10(15.2%)	2(5.0%)	0.110
LYC < 1.0 × 10^9/L	31(29.2%)	19(28.8%)	12(30.0%)	0.894
bone marrow invasion	10(9.4%)	9(13.6%)	1(2.5%)	0.057
extranodal invasion	71(67.0%)	45(68.2%)	26(65.0%)	0.736

Notes: p < 0.05 is indicated in bold.

ECOG PS, Eastern Cooperative Tumor group performance status; LDH, lactate dehydrogenase; ULN, upper limit of normal; LLN, lower limit of normal; PA, prealbumin; AL-B, albumin; HB, hemoglobin; PLT, platelet; β2-MG, β2-microglobulin; WBC, White blood cell; NEC, neutrophil count; LYC, lymphocyte count; POD24, progression of disease within 24 months.

PFS and OS

The median OS for all patients was 60.7 months, with the POD24 group exhibiting a median OS of 11.9 months, while median OS was not reached for the no POD24 group (P < 0.001). The POD12 group had a median OS of 8.7 months, and median OS was again not reached for the POD12 group (P < 0.001) (Figure 2). The 3-year and 5-year OS rates for all patients were 51.3% and 48.4%, respectively. Additionally, the 3-year and 5-year PFS rates for all patients were 33.4% and 31.5%, respectively. In the POD24 group, the 3-year OS rate was 16.0% and 97.3% for the POD and no POD group, respectively. Similarly, the 5-year OS rate in the POD24 group was 16.0%, compared to 91.6% in the no POD24 group. The no POD24 group also had 3-year and 5-year PFS rates of 88.6% and 83.4%, respectively. In the POD12 group, the 3-year OS rate was 15.2% and 79.0% for the POD and no POD group, respectively. The 5-year OS rates in the POD12 and no POD12 groups were 15.2% and 74.3%, respectively. Moreover, the no POD12 group also had 3-year and 5-year PFS rates of 65.6% and 61.7%, respectively.

Figure 2. Overall survival (OS) and Progression-Free-Survival (PFS). (A) OS of all PTCL patients. (B) PFS of all PTCL patients. (C) OS in the POD24 group and no POD24 group. In the no POD24 group, OS was significantly superior to the POD24 group (P < 0.0001). The 3-year OS rate in the POD24 group was 16.0%. (D) OS in the POD12 group and no POD12 group. In the no POD12 group, OS was significantly superior to the POD12 group (P < 0.0001). The 3-year OS rate in the POD12 group was 15.2%.

We conducted an analysis of OS and PFS based on pathological subtype. ALCL exhibited the most favorable prognosis, with median PFS and OS not reached and a 5-year OS rate of 81.8%. Conversely, patients with PTCL-NOS had the worst prognosis, with a median PFS and OS of 7.3 and 21 months, respectively, and a 5-year OS rate of 37.7%. For AITL, the median OS, median PFS, and 5-year OS rate of were 31.5, 7.4 months and 37.4%, respectively. The median OS for ENKTL was not reached, while the median PFS and 5-year OS rate were 24.1 months and 60.5%, respectively (Figure 3) (Figure S1).

Figure 3. OS of PTCL subtypes. The 5-year OS rates in AITL, PTCL-NOS, ENKL, ALCL, and others was 37.4%, 37.7%, 60.5%, 81.8%, and 16.7%, respectively.

Prognostic factors for OS in all patients

Univariate analysis revealed that stage (P = 0.001), B symptoms (P = 0.006), ECOG PS score (P = 0.009), LDH levels (P = 0.044), IPI score (P = 0.001), prealbumin levels (P < 0.001), albumin levels (P = 0.010), platelet count (P = 0.014), β2-MG (P = 0.009), neutrophil count (P = 0.023), POD12 (P < 0.001), POD24 (P < 0.001) and disease remission (P < 0.001) significantly influenced OS in patients with PTCL. The Cox multivariate analysis demonstrated that POD24 status (P = 0.001) and disease remission (P = 0.012) were independent prognostic predictors in PTCL (Table 3).

Table 3. Univariate and multivariate prognostic analysis of Overall Survival (OS) in all patients.

	Univariate		Multivariate
	Hazard ratio (CI)	p value	Hazard ratio (CI)	p value
Age > 60 years	1.523 (0.838-2.769)	0.168
Sex Male/Female	0.949 (0.539-1.673)	0.857
Stage III, IV	3.368(1.668-6.804)	0.001	2.100(0.596-7.401)	0.248
B symptom	2.260 (1.270-4.024)	0.006	0.762(0.282-2.065)	0.594
ECOG PS 2–4	2.189 (1.219-3.931)	0.009	1.279(0.441-3.707)	0.650
LDH > ULN	1.886(1.016-3.501)	0.044	0.996(0.338-2.937)	0.994
IPI 3–5	2.732 (1.548-4.823)	0.001	0.458(0.141-1.484)	0.193
PA < LLN	3.065 (1.705-5.509)	<0.001	0.932(0.268-3.242)	0.912
ALB < LLN	2.265(1.216-4.216)	0.010	2.767(0.716-10.692)	0.140
HB < 120 g/L	1.507(0.859-2.644)	0.152
PLT < 100 × 10^9/L	2.265(1.178-4.352)	0.014	0.424(0.097-1.856)	0.255
β2-MG > ULN	2.792(1.285-6.063)	0.009	0.648(0.164-2.561)	0.536
WBC < 3.5 × 10^9/L	1.553(0.727-3.316)	0.256
NEC < 1.8 × 10^9/L	2.325(1.124-4.815)	0.023	0.666(0.147-3.018)	0.598
LYC < 1.0 × 10^9/L	1.068(0.581-1.963)	0.833
bone marrow invasion	1.884(0.795-4.467)	0.150
extranodal invasion	1.037(0.575-1.869)	0.904
POD24	29.306 (8.694-98.786)	<0.001	52.622(5.225-529.939)	0.001
POD12	9.596(4.895-18.809)	<0.001	2.135(0.556-8.200)	0.269
disease remission	0.102(0.055-0.190)	<0.001	0.200(0.057-0.703)	0.012

Notes: p < 0.05 is indicated in bold.

OS, overall survival; ECOG PS, Eastern Cooperative Tumor group performance status; LDH, lactate dehydrogenase; ULN, upper limit of normal; LLN, lower limit of normal; PA, prealbumin; ALB, albumin; HB, hemoglobin; PLT, platelet; β2-MG, β2-microglobulin; WBC, White blood cell; NEC, neutrophil count; LYC, lymphocyte count; POD24, progression of disease within 24 months.

Prognostic factors for OS in the POD24 group

We conducted an evaluation of various prognostic factors that may impact OS in the POD24 group. Univariate analysis revealed that patients with stage III or IV or decreased albumin levels exhibited shorter OS compared to patients with stage I or II or normal albumin levels, respectively, though this did not reach statistical significance (P = 0.117 and P = 0.090, respectively) (Figure 4).

Figure 4. Prognostic factors for OS in the POD24 group. (A) OS of patients with different stages. (B) OS in ALB normal group and ALB reduced group.

Discussion

PTCL is a relatively uncommon form of NHL, comprising approximately 10–15% of all cases [8]. In this cohort, the 5-year OS rate for all patients was 48.1%, which aligns with previous reports [9–12]. Among the various pathological subtypes of PTCL, PTCL-NOS, AITL, ENKTL and ALCL are commonly observed, among which ALCL exhibits a more favorable prognosis compared to other subtypes [12,13]. Consistent with this, the 3-year OS rate for ALCL in this study was 81.8%, significantly higher than that of other subtypes.

POD24 was established as a prognostically significant outcome in high-risk patients with follicular lymphoma by Casulo et al. in 2015, and it has since has also been found to be of value in other aggressive lymphomas [2,3,14]. Suzuki et al. [7] noted the prognostic significance of POD24 in PTCL excluding ENKL and adult T-cell leukemia/lymphoma. They found the median OS of patients with POD24 was 1.2 years, which was significantly lower than that of patients without POD24. Advani et al. [5] analyzed a cohort of 282 patients with AITL and reported that 27% of them experienced POD24. The 5-year OS rate for all AITL patients in their study was 44%, while the 5-year OS rates for the POD24 and non-POD24 groups were 6% and 63%, respectively. In our study, due to the limited number of AITL patients, a subgroup analysis was not performed, but the 5-year OS rate of patients with AITL was found to be 37.4%, which aligns with the findings reported by Advani et al.

In our study, 62% of our patients experienced POD24. Notably, patients who experienced POD24 had a median OS of 11.9 months and a 3-year OS of 16.0%, which was significantly lower compared to patients without POD24. Furthermore, the group of patients who experienced POD12 had a median OS of 8.7 months and a 3-year OS of 15.2%, which was significantly worse than that of the group without POD12. In a study by Ruiz et al. [15], the prognostic significance of POD12 was analyzed in diffuse large B-cell lymphoma (DLBCL), where they reported a median OS of the POD12 group and non-POD12 group was 14.9 months and not reached, respectively. The authors also observed significant differences in various clinical characteristics between POD12 and non-POD12 patients, including clinical staging, IPI score, presence of extranodal involvement, and presence of bone marrow infiltration, LDH levels, albumin levels, and HB levels. Indeed, in our study, there were significant differences in clinical staging, ECOG PS score, IPI score, LDH levels, β2-MG levels, and prealbumin and albumin levels between the POD12 group and the non-POD12 group, as well as between the POD24 group and the non-POD24 group. However, Suzuki et al. [7] demonstrated that there were no statistically significant variations in clinical staging, ECOG PS score, IPI score, or LDH levels between the POD24 group and the non-POD24 group in PTCL. It is worth noting that this lack of difference may be attributed to the exclusion of patients with ENKTL from their study.

Serum albumin levels have found to be associated with the prognosis and recurrence of non-hematological malignancies [16–18]. In addition to reflecting nutritional status, albumin also serves as an indicator of inflammation severity. Notably, low serum albumin levels have been identified as an unfavorable prognostic factor for patients with lymphoma [4,19,20]. For example, in ENKTL, hypoalbuminemia was an independent prognostic factor for OS in patients with POD24. In the context of autologous hematopoietic stem cell transplantation in lymphoma patients, it has been observed that low levels of albumin prior to transplantation are associated with an increased risk of PFS, while albumin level below 35 g/L are associated with an increased risk of OS. Moreover, in DLBCL, pre-treatment albumin levels demonstrate independent predictive value for OS. Among patients with high NCCN-IPI and albumin level below 3.5 g/dl, the 5-year OS was 29.2%, whereas patients with albumin level above 3.5 g/dl had a 5-year OS rate of 60% (P  = 0.022) [20]. Although albumin was associated with OS according to univariate analysis in this study, it was not confirmed by multivariate analysis. However, in the POD24 group, a univariate analysis also revealed that patients with diminished albumin levels exhibited a more unfavorable prognosis in comparison to those with normal albumin levels. However, this disparities did not achieve statistical significance (P = 0.09), potentially due to the limited size of sample.

In our study, we observed the median OS, median PFS, and 5-year OS rate were 60.7, 12.3 months and 48.4%, respectively. It was poorer than previously reported. Advani et al. reported that at a median follow-up of 58 months, the 5-year OS and PFS estimates for the entire patient cohort were 44% (95% confidence interval [CI], 15–54%) and 32% (95% CI, 17–39%), respectively [5]. These are more favorable compared to a study from the Swedish Lymphoma Registry that observed 5-year OS and PFS estimates of 32% and 20%, respectively [21]. We hypothesized that the differences may in part reflect the median age and differences in baseline patient characteristics.

Our study possesses several limitations. Primarily, it conducted as a single-center study with a relatively small sample size. Additionally, the classification of ALCL into ALK-ALCL and ALK + ALCL was not feasible due to the scarcity of ALCL cases. In future investigations, we intend to expand the sample size and undertake multi-center studies.

In summary, POD24 is predictive of outcomes in patients with PTCL and may serve as an important prognostic tool to stratify high-risk patients. Substantial differences were observed in various parameters including clinical staging, ECOG PS score, IPI score, LDH levels, albumin levels, prealbumin levels, HB levels, and β2-MG levels between the POD24 group and the non-POD24 group. Notably, decreased albumin levels and advanced clinical stage may serve as independent prognostic factors for unfavorable OS in patients with POD24, although additional clinical validation is warranted.

Author contributions

HC (Huimin Chen) and RM are co-first authors and contributed to the experiment conduction, data analysis and manuscript preparation. QZ, FL, YM and JZ collected the data. JC, KQ, ZY, WS, FZ, HS, DL, ZL provided the detailed information of patients. HC (Hai Cheng), KX, and WC contributed to study design and revised the manuscript. RB revised the manuscript.

Ethics approval and informed consent

The study was approved by the ethics committee of The Affiliated Hospital of Xuzhou Medical University (XYFY2023-KL005-01). All participants signed a term of informed consent prior to study commencement.

Acknowledgements

We thank all the researchers and study participants for their contributions. We appreciated Dr. Rahul Bhansali (Department of Medicine, Hospital of the University of Pennsylvania) for his thorough editing spelling, grammar, and syntax.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The dataset used during the study is available from the corresponding author on a reasonable request.

Additional information

Funding

This study was supported by grants from Natural Science Foundation of Jiangsu Province (BK20161177, BK20221218), Scientific Research Project of the Jiangsu Provincial Health and Family Planning Commission/International (JSH-2017-008), China Postdoctoral Science Foundation project (2016M590507, 2018T110557), Xuzhou clinical Backbone Training Project (2018GG006). Suqian Key Laboratory of Hematology (M202111). Development Fund of Affiliated Hospital of Xuzhou Medical University (XYFY2021001).