Frequency of T Regulatory Cells Subpopulations in Hemophilia A Patients with Inhibitors

Abstract

Hemophilia A is a hemorrhagic disease caused by a quantitative/qualitative deficiency of factor VIII. It is classified as severe, moderate, or mild based on its residual procoagulant activity. Long-term administration of FVIII promotes the development of neutralizing antibodies (inhibitors) in almost 30% of patients with the severe form of the disease. Currently, the role of regulatory T cells in the development of these antibodies is conflicting. Accordingly, the aim of this study was to determine the percentage of regulatory T cells subpopulations by flow cytometry in 10 healthy subjects, 15 patients with severe hemophilia without inhibitors, and 8 with inhibitors. No significant differences in the frequency of regulatory T cells subpopulations were found between hemophilia A patients with inhibitors versus hemophilia A patients without inhibitors or healthy subjects. Our results suggest that the role of the regulatory T cells populations on the development of inhibitors in adult patients with hemophilia A is questionable. However, further analysis of the etiological relevance of these cells requires future research.

Keywords

hemophilia - inhibitors - regulatory T cells

Introduction

Hemophilia A (HA) is an X-linked recessive disease in which the concentration or function of the blood coagulation factor VIII (FVIII) is abnormal. In 70% of patients, it is a heritable disease; however, in the rest of cases it is due to a de novo mutation. The propositus will inherit to its offspring the same X-linked inheritance.[1] Based on the activity of FVIII, HA is classified as severe (FVIII: <1%), moderate (FVIII: 1–5%), and mild (FVIII: 5–40%).[2]

Due to the total or partial absence of FVIII, the treatment is based on replacement therapy with plasma derived or recombinant FVIII.[1] However, in some individuals, long-term administration of FVIII promotes the development of neutralizing antibodies called inhibitors, which are defined as anti-FVIII alloantibodies that neutralize the function of infused clotting factor concentrate and are detected by the Nijmegen-modified Bethesda assay, thus affecting the procoagulant activity of FVIII.[3] The main mechanism of action of these alloantibodies is by blocking the interaction sites of FVIII with activated coagulation factors (IXa, IIa, and Xa), as well as to phospholipids.[4] Moreover, it has been described that some of these antibodies are able of degrade FVIII (catalytic antibodies) or increase the clearance of this protein in the kidney.[5] Although, in general, inhibitors are usually polyclonal IgG antibodies, IgA and IgM isotypes have been reported. The most frequent IgG subclasses are IgG4 and IgG1.[3] [4] [6] [7] The estimated prevalence of inhibitors in severe HA patients varies between 20 and 33% while in moderate and mild HA patients it ranges from 3 to 13%.[3] [4]

Regulatory T cells (Tregs) were first described in the 1970s and, lately, they were identified as suppressor T cells.[8] In 1995, these cells were identified as CD25⁺ T cells in mice.[9] The main function of Tregs is to avoid the activation and expansion of conventional T cells although they may also suppress NK and NKT cells, antigen-presenting cells (APC), dendritic cells, monocytes, and macrophages.[10] [11] Effects of Tregs are mediated through four main mechanisms: (1) Suppression of APC; (2) secretion of immunosuppressor cytokines such as tissue growth factor-α (TGF-α), IL-10, and IL-35; (3) cytotoxicity mediated through secretion of granzyme and perforin; and (4) metabolic disruption secondary to generation of adenosine.[11] [12]

The role of Tregs in the development of FVIII inhibitors in HA patients has been previously explored. In 2014, the amount of Tregs (CD4⁺CD25^hi) among healthy subjects (HS) and 12 HA patients (6 with inhibitors and 6 without inhibitors) was explored. The authors were unable to find differences in the amounts of Tregs between HS and HA patients without inhibitors but there was a higher frequency of CD4⁺CD25^hi Tregs in HA patients with inhibitors as compared with HS.[13] In 2016, the amount of Tregs (CD4⁺CD25⁺CD127⁻) was explored in 45 HA pediatric patients. The authors found that HA patients had fewer amounts of Tregs in comparison with HS; these lower values were more marked when HA patients were classified according to their inhibitor status since patients with inhibitors had fewer Tregs versus patients without inhibitors.[14] In 2021, the amount of Tregs and other immunoregulatory cells were evaluated in HA patients with and without inhibitors and no differences were found in the amounts of Tregs (CD4⁺CD25⁺FoxP3⁺) between the groups.[15] Finally, in 2023, a study reported on 8 HA patients with inhibitors and 24 HA patients without inhibitors and 24 HS. The frequency of CD4⁺CD25⁺FOXP3⁺ and CD4⁺CD25 FOXP3⁺ Tregs in all populations was explored but the differences were not significant. Moreover, when they only analyzed CD4⁺CD25 FOXP3⁺ Tregs, HA patients with inhibitors had reduced but non-significant number of these cells as compared with HA patients without inhibitors or HS. Finally, HA patients without inhibitors had a greater frequency of CD4⁺CD25 FOXP3⁺ Tregs versus HS although the difference was not significant in either of them.[16] In conclusion, the role of Tregs in the development of inhibitors in HA patients remains unclear.

Due to their immunosuppressive nature, our initial hypothesis was that patients with HA and inhibitor may have significantly lower levels of Tregs as compared with HA patients without inhibitor or controls, a reduction that could be involved in a likely abnormal immune status contributing to the development of alloantibodies to FVIII. Therefore, the aim of this research was to determine the frequency of Tregs subpopulations CD4⁺CD25⁺CD127⁻,[14] CD4⁺CD25⁺FoxP3⁺,[15] CD4⁺CD49D⁻CD127⁻,[17] CD4⁺CD25⁺CD127⁻FoxP3⁺,[18] CD4⁺CD25⁺49d⁻CD127⁻FoxP3⁺, and conventional T cells CD4⁺CD25⁻CD127⁺ in HA patients with or without inhibitors and HS. In addition to the abovementioned Tregs immunophenotypes, we analyzed the expression of CD49d to CD4⁺CD25⁺CD127⁻FoxP3 yielding the immunophenotype CD4⁺CD25⁺CD49d⁻CD127⁻FoxP3⁺. To our knowledge, this is the first attempt to associate multiple Tregs immunophenotypes with the FVIII inhibitor development in adult HA patients.

Methods

HA Patients

This study included 30 adult HA patients (15 without inhibitors and 15 with inhibitors), treated at the Instituto Mexicano del Seguro Social (IMSS), and 10 HS. Clinical data were obtained from the clinical charts of the patients. Laboratory data were obtained from the results or the samples obtained at entry. When FVIII trough levels and inhibitors were evaluated in patients with and without inhibitor, the last dose of standard FVIII concentrates was indicated at least 72 hours before sampling. Patients with chronic diseases such as diabetes mellitus, arterial hypertension, and autoimmune diseases, and liver with renal failure, cancer, or known atherothrombotic diseases were excluded. None of the patients with or without inhibitors were carriers of hepatitis B, C, or HIV. The protocol was approved by the National Commission for Scientific Research of IMSS (registration number: CNCI-2016–3609–378). The protocol agreed with the ethical guidelines of the Declaration of Helsinki. All the subjects included in this research gave their informed consent before their participation.

Sample Processing

To measure metabolic parameters in an automated Abbott Architect c4000 (Chicago, IL, USA) 5 mL of blood was collected in tubes without anticoagulant (Vacutainer, Beckton Dickinson, Rutherford, NJ, USA). Another 5 mL was collected in tubes containing EDTA (Vacutainer) to assess the hematological parameters using an automated LH780 (Beckman Coulter, Pasadena, CA, USA). Finally, 10 mL blood was collected in citrated tubes (Vacutainer) for the hemostatic (STA Compact, Stago, Asnieres, France) and Tregs evaluations. Samples collected in tubes without anticoagulant and with citrate were centrifuged at 2,200g for 15 minutes to obtain platelet-poor plasma (PPP) as well as the buffy coat. PPP was frozen at −70°C until processing.

Buffy coats were transferred to 15 mL conical tubes and diluted 1:2 with isotonic saline solution (ISS), then peripheral blood mononuclear cells (PBMCs) were isolated using Lymphoprep Density Gradient Medium at 2:1 proportion (2 parts of buffy coat diluted in 1 part of Lymhoprep) (Stemcell Technologies, Cologne, Germany). Tubes were centrifuged at 400g for 30 minutes without brake using an ALC 4236A centrifuge (ALC International, Italy). PBMCs were harvested and washed twice with ISS at 2,200g for 15 minutes in the same centrifuge. Then, 10 µL of PBMCs were diluted 1:10 in trypan blue and counted in a Neubauer chamber to assess their viability. In all HA patients and HS, the viability of PBMCs was >95%.

FVIII Activity

A coagulometric assay was used based on the activated partial thromboplastin time test (aPTT). Briefly, diluted samples (1:10 with Owren Koller buffer) from HS and HA patients were mixed with an equal volume of FVIII-deficient plasma. This later contains all coagulation factors except for FVIII; therefore, the result of the aPTT in the plasma is a function of the FVIII activity provided by the plasma from HS or HA patients. Evaluation of results was conducted using the STA Compaq equipment software and was based on a calibration curve of FVIII activity constructed with calibrated plasma.

Inhibitor Quantification

FVIII inhibitors were quantified based on the Bethesda method. Briefly, plasma of the patients was diluted with 0.1 M imidazole buffer in serial dilutions (1:2 and 1:4); then, a mixture with an equal volume of pooled normal plasma was attempted. At the same time, a control mixture was prepared by incubation of an equal volume of pooled normal plasma and imidazole buffer. For both the mixtures, FVIII activity was calculated after a 2-hour incubation period at 37°C to measure the FVIII residual activity. Inhibitor titer in the plasma of the patients was obtained from a theoretical inhibitor graph by interpolating the percentage of FVIII residual activity versus Bethesda U/mL. As accepted worldwide, we considered an inhibitor to FVIII as positive when the Bethesda titer was >0.6. Moreover, low-responding or high-responding inhibitors were considered when the titer inhibitor was <5.0 or ≥5.0 Bethesda units, respectively.

Tregs Quantification

PBMCs were stained with anti-CD4/PerCP (clone OKT4), anti-CD25/AF488 (clone M-A251), anti-CD49d/BV421 (clone 9F40), and anti-CD127/APC (clone A109D5) for 30 minutes at 4°C. Then, after two washing steps, cells were fixed and permeabilized with a FOXP3 Fix/Perm Buffer Set kit (catalog number 421403), according to the manufacturer's recommendations. Then permeabilized cells were stained with an anti-FoxP3/PE (clone 206D) for 30 minutes at 4°C, after two washing steps. Finally, cells were resuspended in 250 μL PBS. All these antibodies were purchased from Biolegend (Biolegend, San Diego, USA). Samples were analyzed in a FACS ARIA III flow cytometer (Becton Dickinson, New Jersey, USA) using the BD FACS Diva^TM software (Becton Dickinson, v 6.1.3). The analysis was performed with the FlowJo^TM software version 10.0 (Tree Star Software, Ashland, OR) using 50,000 CD4 ⁺ cells.

Tregs were classified according the following immunophenotypes: CD4⁺CD25⁺CD127⁻,[14] CD4⁺CD25⁺FoxP3⁺,[15] CD4⁺CD49D⁻CD127⁻,[17] CD4⁺CD25⁺CD127⁻FoxP3⁺,[18] CD4⁺CD25⁺49d⁻CD127⁻FoxP3⁺, as well as T conventional cells CD4⁺CD25⁻CD127⁺. The algorithm used to define these immunophenotypes is shown in [Fig. 1].

Fig. 1 Gating strategy used to enumerate conventional T cells and Tregs subpopulations. Representative plots of healthy subjects (HS) are shown. (A) To exclude aggregates, single events were selected based on area/height proportion. (B) Based on their scatter properties, lymphocytes were selected. (C) CD4⁺ lymphocytes were chosen to distinguish between T helper cells and other lymphocytes subpopulations. (D) Lymphocytes expressing CD127 and lacking CD25 were designated as conventional T cells. (E) CD4⁺CD25⁺CD127⁻ Tregs. (F) CD4⁺CD49d⁻CD127⁻ Tregs. (G) CD4⁺CD25⁺FoxP3⁺ Tregs. (H) CD4⁺CD25⁺FoxP3⁺CD127⁻ Tregs. (I) CD4⁺CD25⁺FoxP3⁺CD127⁻CD49d⁻ Tregs.

Evaluation of Treg-promoting Cytokines in the Serum of Patients and Healthy Subjects[19]

Three cytokines strongly linked to the function of Tregs were evaluated by an ELISA technique, namely, transforming growth factor b-1 (TGF-β1) (reference values: 31–2,000 pg/mL) (catalog number BMS249–4, Invitrogen Thermo Fisher Scientific, Vienna, Austria), Interleukin-2 (IL-2) (reference values: 18.8–1,200 pg/mL) (catalog number BMS221–2, Invitrogen Fisher Scientific), and Interleukin 10 (IL-10) (reference values: 15.36–600 pg/mL) (catalog number EHIL10, Invitrogen Fisher Scientific). Frozen serum samples were thawed and assessed in duplicate using an SPECTROstar spectrophotometer (BMG Labtech, Ortenberg, Germany), at 620 nm.

Statistical Analysis

Quantitative variables are shown as mean ± standard deviation. One-way ANOVA with Holm-Sidak as a post hoc test was used to calculate the significance among the metabolic, hematological, and hemostatic parameters. ANOVA with Holm-Sidak as a post hoc test was used to calculate the significance between the frequency of Tregs. Statistical analysis was performed with a GraphPad Prism version 8.0 (Graphpad Software, San Diego, CA, USA).

Results

Clinical Data of HA Patients and HS

The study included 15 severe HA patients without inhibitors, 15 HA patients with inhibitors, and 10 HS. One patient in the group without inhibitors and two patients in the group with inhibitors had moderate HA. General characteristics of the study population are shown in [Table 1]. A total of 9 out of 15 patients with HA without inhibitor and 11 out of 15 HA patients with inhibitor were under prophylaxis; all other patients were on-demand treatment. All HA patients received either plasma-derived FVIII or standard half-life recombinant FVIII. After diagnosis of the inhibitor was performed, prophylaxis of the patients was carried out with increasing doses of concentrates of FVIII (n = 8), bypassing agents (n = 4), a combination of FVIII concentrates plus bypassing agents (n = 2), or emicizumab (n = 1). No patient was under immune tolerance induction. At entry to the study, none of the patients with or without inhibitors were receiving immunosuppressive treatment.

Table 1 Characteristics and clinical features of HA patients and HS at the time of sampling patients

	HS (n = 10)	HA (n = 15)	HA + inhibitor (n = 15)	P
Age, years (mean ± SD)	27.4 ± 5.0	23.27 ± 6.7	28.1 ± 10.8	NS
FVIII activity at diagnosis of HA (%) (mean, ranges)	−	0.9 (0.6–3.4)	0.8 (0.6–3.0)	NS
History of prophylaxis (n, %)	−	9 (60.0)	11 (73.3)	NS
With plasma-derived FVIII		1 (6.7)	3 (20.0)	NS
With recombinant FVIII		8 (53.3)	8 (53.3)	NS
History of on-demand treatment (n, %)	−	6 (40.0)	4 (26.7)	NS
With plasma-derived FVIII		3 (20.0)	2 (13.3)	NS
With recombinant FVIII		3 (20.0)	2 (13.3)	NS
Higher historical inhibitor titer (BU) (mean, ranges)	−	1.8 (1.1–3.2)	25.6 (10.6–49.8)	0.0002*
Inhibitor titer at study entry (BU) (mean, ranges)	−	<0.6	22.6 (9.1–47.8)	0.0001*
Treatment after inhibitor diagnosis (n, %)
Increasing doses of concentrates of FVIII			8 (53.3)
Bypassing agents			4 (26.7)
Combination of FVIII + bypassing agents			2 (13.3)
Emicizumab			1 (6.7)
Weight (kg) (mean, ranges)	78 (67–95)	81 (60–108)	83 (69–113)	NS
Height (cm) (mean, ranges)	175 (163–191)	172 (169–184)	180 (171–187)	NS
Time to high titer inhibitor diagnosis (months) (median, ranges)	−	−	144 (101–186)
Clinical arthropathy (n, %)	−	4 (26.7)	7 (46.7)	0.01*
US arthropathy (n, %)	−	6 (40)	10 (66.7)	0.002*

Abbreviations: HA, hemophilia A; HS, healthy subjects.

Notes: *HA patients without inhibitors versus HA patients with inhibitors.

Hematological and Hemostatic Parameters in HA Patients and HS

Comparison of metabolic and hematological parameters are shown in [Tables 2] and [3]. As observed, although significant differences among the groups were found, in general, the three populations studied were quite homogeneous.

Table 2 Metabolic parameters in HA patients and HS

Variable	HS (n = 10)	HA (n = 15)	HA + inhibitor (n = 15)
Glucose (mg/dL)	85.0 ± 5.8	85.0 ± 15.1	88.6 ± 10.4
Urea nitrogen (mg/dL)	14.8 ± 3.6	14.7 ± 3.7	13.5 ± 4.1
Urea (mg/dL)	31.6 ± 7.6	31.4 ± 7.9	29.9 ± 9.3
Creatinine (mg/dL)	0.9 ± 0.1	0.8 ± 0.1	0.9 ± 0.4
TC (mg/dL)	179.4 ± 36.4	151.9 ± 23.1	168.4 ± 37.2
TGL (mg/dL)	135.3 ± 44.3	87.5 ± 34.9	121.6 ± 46.5
HDL-C (mg/dL)	38.7 ± 7.9	45.7 ± 8.6	44.6 ± 10.4
LDL-C (mg/dL)	110.89 ± 40.13	88.4 ± 18.9	92.5 ± 30.7
AST (U/L)	30.5 ± 10.3	27.2 ± 12.4	29.9 ± 10.5
ALT/TGP (U/L)	42.2 ± 20.6	31.2 ± 35.3	32.8 ± 15.8
TB (mg/dL)	0.8 ± 0.2	0.8 ± 0.4	0.9 ± 0.3
DB (mg/dL)	0.1 ± 0.1	0.2 ± 0.1	0.1 ± 0.5
IB (mg/dL)	0.7 ± 0.2	0.6 ± 0.3	0.6 ± 0.8
LDH (U/L)	144.2 ± 16.5	182.6 ± 35.7	242.7 ± 65.2[b]
AP (U/L)	72.5 ± 23.7	93.1 ± 16.3	112.6 ± 30.1
TP (g/dL)	7.3 ± 0.4	7.4 ± 0.40	7.9 ± 0.3
Albumin (g/dL)	4.4 ± 0.4	4.9 ± 0.4	4.7 ± 0.5
Globulins (g/dL)	2.8 ± 0.1	2.5 ± 0.3	3.1 ± 0.2
A/G ratio	1.5 ± 0.2	1.9 ± 0.2[a]	1.4 ± 0.8[c]
GGT (U/L)	38.0 ± 43.3	24.3 ± 16.9	40.9 ± 30.2

Abbreviations: A/G ratio, albumin/globulin ratio; ALT, alanine aminotransferase; AP, alkaline phosphatase; AST, aspartate aminotransferase; DB, direct bilirubin; GGT, gamma glutamyl transferase; HA, hemophilia A; HDL-C, high density lipoprotein cholesterol; HS, healthy subjects; IB, indirect bilirubin; LDH, lactate dehydrogenase; LDL-C, low density lipoprotein cholesterol; TB, total bilirubin; TC, total cholesterol; TG, triglycerides; TP, total proteins.

Notes: a p <0.05 between HA patients without inhibitors and HS.

^b p <0.05 between HA patients with inhibitors and HS.

^c p <0.05 between HA patients with and without inhibitors.

Table 3 Hematological parameters in HA patients and HS

	HS (n = 10)	HA (n = 15)	HA + inhibitors (n = 15)
Erythrocyte (×109/mL)	5.5 ± 0.3	5.2 ± 0.4	5.8 ± 0.7
Hemoglobin (g/dL)	17.2 ± 0.7	16.3 ± 0.9	16.1 ± 1.6
Hematocrit (%)	50.0 ± 1.6	47.1 ± 2.7	45.7 ± 3.9
MCV (fL)	90.6 ± 2.9	90.0 ± 3.4	88.3 ± 4.0
MCH (pg)	31.2 ± 0.9	31.3 ± 1.8	31.2 ± 1.7
MCHC (g/dL)	34.4 ± 0.4	34.5 ± 0.5	34.7 ± 1.0
RDW (%)	11.6 ± 0.4	12.9 ± 0.7[a]	13.2 ± 0.2
Platelets (×109/mL)	204.2 ± 36.7	250.2 ± 17.8	247.1 ± 20.5
MPV (fL)	9.6 ± 0.5	8.8 ± 0.5	9.1 ± 0.9
Leucocytes (×109/mL)	7.4 ± 1.6	5.6 ± 1.4	6.4 ± 1.6
Lymphocytes (%)	36.7 ± 6.6	37.0 ± 13.2	33.5 ± 10.5
Monocytes (%)	7.7 ± 1.2	9.6 ± 3.9	9.9 ± 2.5
Neutrophils (%)	51.6 ± 7.9	51.2 ± 10.9	50.3 ± 8.9
Eosinophils (%)	2.8 ± 1.4	1.6 ± 1.0	1.3. ± 0.9
Basophils (%)	1.0 ± 0.7	0.6 ± 0.5	0.6 ± 0.9
Lymphocytes (×109/mL)	2.7 ± 1.1	2.1 ± 0.4	2.5 ± 0.4
Monocytes (×109/mL)	0.5 ± 0.2	0.4 ± 0.3	0.3 ± 0.7
Neutrophils (×109/mL)	3.7 ± 0.4	2.8 ± 1.2	3.5 ± 1.8
Eosinophils (×109/mL)	0.2 ± 0.1	0.1 ± 0.1	0.1 ± 0.1
Basophils (×109/mL)	0.04 ± 0.1	0 ± 0.0[b]	0.1 ± 0.01

Abbreviations: HA, hemophilia A; HS, healthy subjects; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular hemoglobin; MPV, mean platelet volume; RDW, red cell distribution width.

Notes: Multiple t-test with Holm-Sidak method was used to calculate significance.

^a p <0.05 between HA patients without inhibitors and HS.

^b p <0.05 between HA patients with and without inhibitors.

Percentage of CD4⁺ and CD4⁺CD25⁻CD127⁺ Cells

We assessed the percentage of CD4⁺ cells with respect to total lymphocytes. The percentages of CD4⁺ cells were 28.7 ± 6.8% for HS, 31.6 ± 6.3% for HA patients without inhibitors, and 34.8 ± 3.3% for HA patients with inhibitors. Percentages of CD4⁺ were not different (p >0.05) among the study groups ([Fig. 2A]). Then, based on the immunophenotype CD4⁺CD25⁻CD127⁺, conventional T cells were quantified in HA patients and HS. Percentages of conventional T cells were 86.5 ± 6.7% in HS, 82.0 ± 9.1% in HA patients without inhibitors, and 83.8 ± 4.2% in HA patients with inhibitors ([Fig. 2B]). As observed for CD4⁺ cells, the percentages of CD4⁺CD25⁻CD127⁺ cells were not significantly different between the groups (p >0.05).

Fig. 2 No difference in the percentages of Tregs subpopulations between hemophilia A (HA) patients and healthy subjects (HS). (A) Flow cytometry was used to evaluate CD4⁺ cells regarding the total lymphocytes. (B) Then T conventional cells CD4⁺CD25^-CD127⁺ were assessed. Multiple Tregs immunophenotypes were evaluated: (C) CD4⁺CD25⁺CD127^- Tregs; (D) CD4⁺CD49d^-CD127^- Tregs; (E) CD4⁺CD25⁺FoxP3⁺ Tregs; (F) CD4⁺CD25⁺FoxP3⁺CD127^- Tregs; and (G) CD4⁺CD25⁺FoxP3⁺CD127⁻CD49d⁻ Tregs. All Tregs immunophenotypes are expressed as percentage of CD4⁺ cells. One-way ANOVA with Holm-Sidak as a post hoc test was used to calculate significance. p <0.05 was considered significant.

Percentage of Tregs Subpopulations

Multiple Tregs immunophenotypes were also assessed. First, we determined the percentages of CD4⁺CD25⁺CD127⁻ Tregs as previously evaluated in pediatric HA patients.[14] Percentages of Tregs were 1.9 ± 0.5% for HS, 2.1 ± 0.9% for HA patients without inhibitors, and 1.9 ± 0.3% for HA patients with inhibitors ([Fig. 2C]). A previously described, second CD4⁺CD49d⁻CD127 immunophenotype was used to avoid contamination with effector T cells.[17] By using this phenotype, we found a percentage of Tregs of 3.8 ± 0.6% for HS, 3.0 ± 1.9% for HA patients without inhibitors, and 2.9 ± 1.0% for HA patients with inhibitors ([Fig. 2D]). The third phenotype was CD4⁺CD25⁺FoxP3⁺, as previously used to evaluate the evolution of Tregs percentages during immunotolerance induction in HA patients.[17] Using this immunophenotype, we found that percentages of Tregs were 1.2 ± 0.2% for HS, 1.3 ± 0.6% for HA patients without inhibitors, and 1.1 ± 0.2% for HA patients with inhibitors ([Fig. 2E]). In addition, percentages of Tregs of 1.2 ± 0.8% in HS, 1.3 ± 0.6% in HA patients without inhibitors, and 1.0 ± 0.7% for HA patients with inhibitors were found when a fourth immunophenotype CD4⁺CD25⁺FoxP3⁺CD127⁻¹⁸ was used ([Fig. 2F]). Finally, because the lack of expression of CD49d to an accepted Tregs CD4⁺CD25⁺FoxP3⁺CD127⁻ immunophenotype[17] yields the CD4⁺CD25⁺FoxP3⁺CD127⁻CD49d⁻ immunophenotype, we searched for the percentage of its expression in Tregs: 0.8 ± 0.6% for HS, 0.8 ± 0.4% in HA patients without inhibitors, and 0.9 ± 0.6% in HA patients with inhibitors ([Fig. 2G]).

As observed in [Fig. 2], no statistical differences regarding the frequency of CD4⁺ cells, conventional T cells, or multiple Tregs immunophenotypes between HA patients and HS were found. Lastly, the median fluorescence intensity (MFI) of FoxP3 in all Tregs subpopulations was also evaluated. However, no significant differences in the MFI for FoxP3 in the Tregs subpopulations between HA patients and HS were found (data not shown).

Analysis of Treg-promoting Cytokines in the Serum of Patients and Healthy Controls

Because no significant differences in the percentages of Tregs were found among the three groups analyzed, we attempted to evaluate the cytokine response in the serum of the patients and controls. Three cytokines strongly associated with the Treg function were evaluated, namely, TGF-b1, IL-2, and IL-10. As shown in [Table 4], no significant differences were found among the groups in terms of the serum concentrations of these cytokines suggesting that not only the percentages but also the function of these cells was preserved in all individuals analyzed.

Table 4 Treg-promoting cytokines in patients with hemophilia and controls

	HS (n = 10)	HA (n = 15)	HA + inhibitors (n = 15)	NS
TGF-β1 (pg/mL)	1,326 ± 125	1,480 ± 147	1,432 ± 122.9	NS
IL-2 (pg/mL)	733.8 ± 69.8	776.3 ± 70.56	789.8 ± 89.3	NS
IL-6 (pg/mL)	422.5 ± 85.4	401.5 ± 99.7	399.7 ± 65.0	NS

Abbreviations: HA, hemophilia A; HS, healthy subjects; IL-2, interleukin 2; IL-10, interleukin 10; TGF-β1, transforming growth factor β1.

Discussion

FVIII inhibitors remain the most significant and feared complication associated with factor replacement therapy in HA patients.[20] Inhibitors affect almost 25% and 3 to 13% of severe and mild or moderate HA patients, respectively.[3] [4] Control of the ongoing immune response is conducted by several regulatory cells such as regulatory macrophages, regulatory dendritic cells, regulatory natural killer cells, regulatory B cells, and Tregs.[21] Based on their FoxP3 expression, multiple Tregs have been described.[22] FoxP3⁻ Tregs are constituted by either Tr1 cells which can produce IL-10 and TGF-α,[23] as well as T_H3 cells which express a membrane-bound form of TGF-α called latency-associated peptide.[24] Moreover, it has been described in some FoxP3⁺ immunophenotypes such as CD4⁺CD25⁺FoxP3⁺,[15] CD4⁺CD25⁺CD127⁻FoxP3⁺, and CD4⁺FoxP3⁺ Tregs.[25]

In this study, different Tregs immunophenotypes were analyzed in HA patients and HS: CD4⁺CD49d⁻CD127⁻, CD4⁺CD25⁺CD127⁻, and CD4⁺CD25⁺CD127⁻FoxP3⁺. Considering this last traditional Tregs immunophenotype, we added the lack of expression of CD49d: CD4⁺CD25⁺CD49d⁻CD127⁻FoxP3⁺ to ensure that the studied population had immunosuppressive activity avoiding the contamination of effector cells. In none of the immunophenotypes analyzed in HA patients, with or without inhibitors, statical differences were observed as compared with HS. Again, after analyzing the MFI of FoxP3 of each immunophenotype no significant differences were seen between HA patients and HS.

The frequency of CD4⁺CD25⁺CD127⁻ Tregs in 45 pediatric HA patients (median age 7.6 ± 4.1 years old) was published. It was described that HA patients with inhibitors had fewer Tregs compared with HA patients without inhibitors and HS.[14] For example, in accordance with a previous publication, our results show no differences in the Tregs percentage between HA patients and HS independently of their inhibitor status even though the same Tregs immunophenotype was used.[15] In pediatric HA patients (7.6 ± 4.1 years old), the differences were significant; however, in our study with adult HA patients we were unable to find significant differences when patients with or without inhibitors or HS (age at study entry 28.1 ± 10.8, 23.3 ± 6.7, and 27.4 ± 5.0 years old, respectively) were analyzed. We may hypothesize some explanations for this discrepancy between children and adult patients. First, an explanation for this finding could be that older patients had higher exposure days than non-inhibitor HA patients; consequently, they may have a higher risk of inhibitor development. However, this was dismissed in a study showing that most inhibitors develop before 70 cumulative exposure days.[26] Second, changes in the immune modulatory response associated with aging. Third, the effect of multiple additive comorbidities in adults, which may modify the immune response as compared with the patterns of children. Fourth, the differences are simply due to the non-comparable nature of both populations. More research regarding the effect of age on the frequency of Tregs and the development of inhibitors in HA patients is warranted.

In another study, high frequencies of Tregs CD4 + 25hi in HA patients with inhibitors were described as compared with HS; however, the Tregs immunophenotype (CD4⁺CD25^hi) was more related to activated T effector cells rather than Tregs cells. It has been demonstrated that the activation of PBMC with bryostatin and ionomycin results in the induction of FoxP3 and CD25 without conferring regulatory functions.[27]

On the other hand, it is possible that the defect in the immune regulation (including FVIII) may be located in the secondary lymphoid organs, more specifically the spleen, because the splenic marginal zone is one of the inductive places where the immune response begins.[28] [29] Specifically, the immune defect may be located at the splenic T regulatory cells because the existence of a splenic Tregs population has been described.[30] However, its effects on the regulation of the immune response against FVIII have not been reported yet. Of course, evaluation of these cells in HA patients would be ethical and technically cumbersome in patients with an increased tendency for bleeding. One approach would be the quantification of recirculating spleen Tregs as previously performed.[30] Briefly, circulating follicular regulatory T cells (T_FR) were evaluated in patients with multiple sclerosis based on the CD4⁺CD25⁺CD127⁻CXCR5⁺PD-1⁺ immunophenotype. A decreased frequency of circulating T_FR and increased ratio of circulating T follicular cells/circulating T follicular regulatory T cells in patients with multiple sclerosis as compared with HS was found.[31]

Multiple studies have used genetically engineered Tregs to induce tolerance in HA models. A B cell–targeting Ab receptor (BAR)-transduced human CD4⁺ Tregs (BAR hTreg), in which the extracellular domain of the BAR contains the immunodominant FVIII A2 or C2 domains, has been created. In HA mice, these BAR Tregs suppressed the generation of anti-FVIII antibodies in response to the blockage of B cell differentiation up to antibody-secreting B cells (ASC), both prophylactically as well as in HA mice with preexistent antibodies.[32] This group described the mechanism for this suppressant activity; they found that BAR Tregs acted on FVIII-specific memory B cells in a cell contact-dependent manner.[33] Strikingly, there is a species-dependent variation in the expression of cytotoxic markers. In human Tregs, low expression of cytotoxicity markers such as granzyme B and perforin was described in comparison with BAR T conventional cells.[32] In contrast, in mouse BAR natural Tregs, more than 96% of BAR Tregs expressed granzyme B.[33] Nonetheless, the authors did not explain whether these mouse BAR Tregs had cytolytic activity. In human and mice CD8⁺ cytotoxic T cells, A2/C2-BAR CD8⁺ T cells could eliminate FVIII-specific B cells, thus preventing anti-FVIII antibody formation.[34]

Another possibility to explain the lack of association between the percentage of Tregs and FVIII inhibitors is that regulation of the anti-FVIII response would not be entirely associated with Tregs. The role of regulatory B cells (Bregs) in the FVIII tolerance was described after the phenotype and function of CD19⁺CD24^hiCD38^hi Bregs in HA patients with and without inhibitors was evaluated. A reduction of the frequency of Bregs in patients with inhibitors versus HA patients without inhibitors and HS was found. In contrast, the percentage of Bregs as part of B cells was similar between HA patients with and without inhibitors. Then B cells producing IL-10 were assayed and a reduced percentage of these cells was found in HA patients with inhibitors versus HS. However, the percentage of these cells between HA patients with and without inhibitors was not significant.[35] This was confirmed after showing that the frequency of CD19⁺CD24^hiCD38^hi Bregs was lower in HA patients with inhibitors than HA patients without inhibitors.[15]

Searching for other cells implicated in FVIII tolerance has yielded poor results. The frequency of CD3⁺CD4⁺CD185⁺CD278⁺CD279⁺ follicular T helper cells was analyzed in HA patients with and without inhibitors and no statistical differences were found.[35] The proportion of Lin^–HLA-DR^low CD11b⁺ CD33⁺ myeloid-derived suppressor cells was analyzed in HA patients with and without inhibitors but the frequency of these cells was similar in both groups.[15]

Our study has limitations requiring discussion. First, it must be underlined that an important limitation of this and previous studies is that the differentiation ability from T helper phenotype up to Tregs phenotype was not evaluated, a prominent issue to discriminate between natural Tregs and induced Tregs. Second, we did not include enough HA patients with inhibitors that allowed us to build an association between the titer of the inhibitor and the frequency of Tregs. Third, at first glance, the number of patients included may appear small; however, this is a biomedical investigation rather than a clinical research. Despite this limitation, no significant differences were found either in terms of the frequency of Tregs or in some aspects of their function. Fourth, only adult patients were included in this study, a fact that may be considered a weakness. However, we decided not to include children under the consideration that in this population with a transient immature immunological system, it would be difficult to be sure that inhibitors would never appear during the treatment. On the contrary, including adult patients allowed us to have a population with a fully developed immune system and a long-term exposure to FVIII, reducing the possibility of developing alloantibodies in the future.

In conclusion, we did not find differences between HA patients with or without inhibitors or HS in terms of the frequencies of multiple Tregs phenotypes or MFI of FoxP3 or in the evaluation of Treg-promoting cytokines. In the future, to analyze a higher number of HA patients with and without inhibitors randomly treated with plasma-derived or recombinant FVIII would be desirable to investigate the biochemical and cellular immunoregulatory response networks to the FVIII obtained from various sources. A higher number of HA patients with inhibitors will be useful to make a correlation between the titer of inhibitor and the frequency of cells with immunosuppressive functions such as Tregs, Bregs, and myeloid-derived suppressor cells.

Conflict of Interests

The authors declare that they have no conflict of interest.

• References

• 1 García-Chávez J, Majluf-Cruz A. Hemofilia. Gac Med Mex 2013; 149 (03) 308-321
• 2 Santagostino E, Dougall A, Jackson M. et al. Chapter 2: Comprehensive care of hemophilia. WFH Guidelines for the Management of Hemophilia, 3rd edition. Haemophilia 2020; 26 (Suppl 6): 19-34
• 3 Ragni MV, Berntorp E, Carcao M. et al. Chapter 8: Inhibitors to clotting factor. Haemophilia 2020; 26 (Suppl. 06) 95-107
• 4 Green D. Factor VIII inhibitors: a 50-year perspective. Haemophilia 2011; 17 (06) 831-838
• 5 Lacroix-Desmazes S, Moreau A, Sooryanarayana. et al. Catalytic activity of antibodies against factor VIII in patients with hemophilia A. Nat Med 1999; 5 (09) 1044-1047
• 6 Astermark J. Basic aspects of inhibitors to factors VIII and IX and the influence of non-genetic risk factors. Haemophilia 2006; 12 (Suppl. 06) 8-13 , discussion 13–14
• 7 Lacroix-Desmazes S, Navarrete AM, André S, Bayry J, Kaveri SV, Dasgupta S. Dynamics of factor VIII interactions determine its immunologic fate in hemophilia A. Blood 2008; 112 (02) 240-249
• 8 Gershon RK, Kondo K. Cell interactions in the induction of tolerance: the role of thymic lymphocytes. Immunology 1970; 18 (05) 723-737
• 9 Sakaguchi S, Sakaguchi N, Asano M, Itoh M, Toda M. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J Immunol 1995; 155 (03) 1151-1164
• 10 Bodor J, Bopp T, Vaeth M. et al. Cyclic AMP underpins suppression by regulatory T cells. Eur J Immunol 2012; 42 (06) 1375-1384
• 11 Wardell CM, MacDonald KN, Levings MK, Cook L. Cross talk between human regulatory T cells and antigen-presenting cells: lessons for clinical applications. Eur J Immunol 2021; 51 (01) 27-38
• 12 Vignali DAA, Collison LW, Workman CJ. How regulatory T cells work. Nat Rev Immunol 2008; 8 (07) 523-532
• 13 Ding KY, Ji WC, Wu JS, Li T, Sheng YY. Higher frequency of CD4(+)CD25(high) Treg cells in hemophilia patients with factor VIII inhibitor. Genet Mol Res 2014; 13 (01) 1774-1781
• 14 El-Asrar MA, Hamed Ael-S, Darwish YW, Ismail EAR, Ismail NA. Assessment of the frequency of regulatory T cells (CD4+CD25+CD127-) in children with hemophilia A: relation to factor VIII inhibitors and disease severity. Blood Coagul Fibrinolysis 2016; 27 (01) 42-46
• 15 Schep SJ, Schutgens REG, Fischer K, Voorberg J, Boes M. Role of regulatory cells in immune tolerance induction in hemophilia A. HemaSphere 2021; 5 (05) e557
• 16 Arandi N, Zekavat OR, Shokrgozar N, Shahsavani A, Golmoghaddam H, Kalani M. Altered frequency of FOXP3⁺ regulatory T cells is associated with development of inhibitors in patients with severe hemophilia A. Int J Lab Hematol 2023; 45 (06) 953-960
• 17 Kleinewietfeld M, Starke M, Di Mitri D. et al. CD49d provides access to “untouched” human Foxp3+ Treg free of contaminating effector cells. Blood 2009; 113 (04) 827-836
• 18 Santegoets SJAM, Dijkgraaf EM, Battaglia A. et al. Monitoring regulatory T cells in clinical samples: consensus on an essential marker set and gating strategy for regulatory T cell analysis by flow cytometry. Cancer Immunol Immunother 2015; 64 (10) 1271-1286
• 19 Zong Y, Deng K, Chong WP. Regulation of Treg cells by cytokine signaling and co-stimulatory molecules. Front Immunol 2024; 15: 1387975
• 20 Witmer C, Young G. Factor VIII inhibitors in hemophilia A: rationale and latest evidence. Ther Adv Hematol 2013; 4 (01) 59-72
• 21 Salminen A. Activation of immunosuppressive network in the aging process. Ageing Res Rev 2020; 57 (Jan): 100998
• 22 Chien CH, Chiang BL. Regulatory T cells induced by B cells: a novel subpopulation of regulatory T cells. J Biomed Sci 2017; 24 (01) 86
• 23 Roncarolo MG, Gregori S, Bacchetta R, Battaglia M, Gagliani N. The biology of T regulatory type 1 cells and their therapeutic application in immune-mediated diseases. Immunity 2018; 49 (06) 1004-1019
• 24 Gandhi R, Farez MF, Wang Y, Kozoriz D, Quintana FJ, Weiner HL. Cutting edge: human latency-associated peptide+ T cells: a novel regulatory T cell subset. J Immunol 2010; 184 (09) 4620-4624
• 25 Long SA, Buckner JH. CD4+FOXP3+ T regulatory cells in human autoimmunity: more than a numbers game. J Immunol 2011; 187 (05) 2061-2066
• 26 Gouw SC, van den Berg HM, Fischer K. et al; PedNet and Research of Determinants of INhibitor development (RODIN) Study Group. Intensity of factor VIII treatment and inhibitor development in children with severe hemophilia A: the RODIN study. Blood 2013; 121 (20) 4046-4055
• 27 Kmieciak M, Gowda M, Graham L. et al. Human T cells express CD25 and Foxp3 upon activation and exhibit effector/memory phenotypes without any regulatory/suppressor function. J Transl Med 2009; 7: 89
• 28 Zerra PE, Cox C, Baldwin WH. et al. Marginal zone B cells are critical to factor VIII inhibitor formation in mice with hemophilia A. Blood 2017; 130 (23) 2559-2568
• 29 Navarrete A, Dasgupta S, Delignat S. et al. Splenic marginal zone antigen-presenting cells are critical for the primary allo-immune response to therapeutic factor VIII in hemophilia A. J Thromb Haemost 2009; 7 (11) 1816-1823
• 30 Li C, Muñoz-Rojas AR, Wang G, Mann AO, Benoist C, Mathis D. PPARγ marks splenic precursors of multiple nonlymphoid-tissue Treg compartments. Proc Natl Acad Sci U S A 2021; 118 (13) e2025197118
• 31 Dhaeze T, Peelen E, Hombrouck A. et al. Circulating follicular regulatory T cells are defective in multiple sclerosis. J Immunol 2015; 195 (03) 832-840
• 32 Zhang A-H, Yoon J, Kim YC, Scott DW. Targeting antigen-specific B cells using antigen-expressing transduced regulatory T cells. J Immunol 2018; 201 (05) 1434-1441
• 33 Pohl AP, Venkatesha SH, Zhang AH, Scott DW. Suppression of FVIII-specific memory B cells by chimeric BAR receptor-engineered natural regulatory T cells. Front Immunol 2020; 11 (April): 693
• 34 Parvathaneni K, Scott DW. Engineered FVIII-expressing cytotoxic T cells target and kill FVIII-specific B cells in vitro and in vivo. Blood Adv 2018; 2 (18) 2332-2340
• 35 Boulassel MR, Al-Ghonimi M, Al-Balushi B. et al. Regulatory B cells are functionally impaired in patients having hemophilia A with inhibitors. Clin Appl Thromb Hemost 2018; 24 (04) 618-624

Publication History

Received: 22 December 2024

Accepted: 27 August 2025

Article published online:
02 February 2026