key: cord-0027113-7ccii67v
authors: Lunghi, Barbara; Morfini, Massimo; Martinelli, Nicola; Linari, Silvia; Castaman, Giancarlo; Bernardi, Francesco
title: Combination of CLEC4M rs868875 G-Carriership and ABO O Genotypes May Predict Faster Decay of FVIII Infused in Hemophilia A Patients
date: 2022-01-29
journal: J Clin Med
DOI: 10.3390/jcm11030733
sha: 566d09a9d1aa7a7aafa1f5a92c6ca63dbba39b10
doc_id: 27113
cord_uid: 7ccii67v

The C-type lectin CLEC4M binds and internalizes factor VIII (FVIII). Common CLEC4M variants have been associated with FVIII pharmacokinetic (PK) profiles in hemophilia A (HA) patients. The two-compartment PK analysis of plasma-derived (pd-) and full length recombinant FVIII concentrates was conducted in twenty-six patients (FVIII:C ≤ 2 IU/dL). F8, ABO blood-groups, and the CLEC4M rs868875A/G polymorphism were genotyped. CLEC4M genotype groups differed for the elimination rate constant K 1-0 (p < 0.001), half-life (K 1-0 HL), and the Beta rate constant. Patients treated with pd-FVIII also differed in the Alpha phase. In linear regression models, the contribution of the CLEC4M genotypes to FVIII PK parameters remained significant after correction for ABO, age, and VWF antigen levels at PK. Combined CLEC4M rs868875A/G and ABO genotypes displayed significant interaction (K 1-0, p = 0.014). Compared to other combined genotypes, the G-carriers/O genotypes showed half-reduced K 1-0 HL (p = 0.008), and faster FVIII clearance (mean 7.1 ± 2.2 mL/h/kg SE) than in the G-carriers/non-O (mean 2.4 ± 0.3 mL/h/kg SE), (p = 0.038). Comparison in HA patients recruited in several countries suggests that CLEC4M genotypes coherently influence infused FVIII half-life and clearance. Our analysis supports substantially faster FVIII decay associated with the rs868875 G-carrier/ABO O genotypes, which has potential implications for genetically tailored substitutive HA treatment.

Genetic components, together with environmental factors [1] [2] [3] [4] [5] [6] , provide explanation only for a small portion of the large inter-patient variability of factor VIII (FVIII) pharmacokinetics (PK) in hemophilia A (HA) patients. Age, von Willebrand factor (VWF) levels, and the ABO blood group altogether explain 30% of such variability in severe HA [1] [2] [3] [4] [5] .

Among the candidate scavenger receptors [7] for circulating FVIII/VWF, the lowdensity lipoprotein receptor (LDLR), the stabilin 2 (STAB2), and the asialoglycoprotein receptor minor subunit gene (ASGR2) have been recently associated with FVIII pharmacokinetics (PK) through gene variation in small size HA cohorts [8] [9] [10] , independently from ABO blood-group for the LDLR and ASGR2 [8, 10] genotypes.

Among receptors, the C-type lectin domain family 4 member M (CLEC4M), also termed L-SIGN or DC-SIGNR, is expressed on the sinusoidal endothelial cells of the liver, the main source of circulating FVIII, and thus, is of noticeable interest both for biological and clinical reasons. Its extra-cellular binding domain consists of an extended neck region, which contains tandem repeats, followed by a C-terminal C-type carbohydrate-recognition domain, which complexes carbohydrates of a high-mannose type, well represented in FVIII [11, 12] and VWF molecules [13, 14] .

In cellular and animal models, CLEC4M binds and internalizes FVIII in a VWF-dependent and -independent manner [15] . CLEC4M can also act as a cell-adhesion and pathogenrecognition receptor, which could provide a link between coagulation and infection [16] .

Common variants at the CLEC4M locus were associated with the FVIII PK profile in a population of severe pediatric HA patients, and in populations of adult moderate/severe HA patients [9, 17, 18] . These observations have been obtained by the infusion of different plasma-derived (pd-) or recombinant (r-) standard half-life FVIII concentrates, and by different PK analysis approaches, the PopPK [17] , and the model-independent method individual PK [9, 18] .

Prompted by these observations, and aimed at understanding specific components of interest for substitutive treatment in HA patients, we have analyzed association of CLEC4M genotypes with two-compartment model full length (FL) FVIII PK parameters, and compared the degree of association with those observed in recently published studies. We focused on the interaction between CLEC4M gene variation and ABO genotypes, which are well recognized genetic components of FVIII PK outcomes [1, 8, 19, 20] .

This study was conducted in accordance with the principles of Helsinki Declaration, and with Good Clinical Practice. This study reports an investigator-initiated IRB-approved retrospective chart review of FVIII PK performed in the Center for Bleeding Disorders, Careggi University Hospital, Florence, Italy, for optimization of replacement treatment. Patients expressed their oral and written informed consent for PK execution and genotyping, originally performed for F8 mutation detection.

Patients with severe or moderately severe HA (FVIII:C ≤ 2 IU/dL), treated with plasma-derived (pd)-FVIII or r-FVIII SHL concentrates, were selected. Patients affected by full-blown AIDS (CD4 < 200/mcL) were not included in the study. Patients who had the FVIII inhibitor test by Nijmegen assay > 0.4 IU/dL or INR > 1.3 were excluded.

Twenty-six patients (median age 39 years, SD 14 years, range 14-67 years) were investigated for FVIII PK and CLEC4M genotype. The PK study was conducted as previously described [10] . After a 3-4 day wash out, HA patients were infused with a single dose of FVIII products (22.7-51.8 IU/kg). Twenty-two patients were treated with pd-FVIII, and eleven with r-FVIII concentrates. For seven patients, PKs were obtained with both FVIII concentrates. Twelve patients (46%) underwent repeated PKs (2-5).

Platelet poor plasma was stored at −40 • C in 0.5 mL aliquots. FVIII coagulant activity assays (One-Stage Method) were done at the same time on three duplicate dilutions of baseline and post-infusion samples [21] . VWF antigen (VWF:Ag) determination was performed as previously described [22] , and was available for n = 24 patients.

The blood samples were collected before infusion and after 0.25, 0.5, 1, 3, 6, 9, 24, 28, 48, and 72 h, for better evaluation of the Beta phase. Each PK decay was analyzed according to the Two-Compartment Model (TCM) by WinNonlin 7.0 (Phoenix 64, Pharsight, Mountain View, CA, USA). From 54 PKs, resulted fitting the TCM better, we considered specific final and secondary PK parameters listed in the legend to Table 1. In the twelve patients who underwent repeated PKs, mean PK parameter values were calculated and used as single case values for association studies with genetic polymorphisms. 

F8 mutations were found by direct sequencing [23] , F8 intron 22 inversion (IVS 22), and ABO blood-group, as previously described [2, 24] . The rs868875 A/G polymorphism of CLEC4M gene was investigated by TaqI restriction analysis of a PCR fragment (215 bp) obtained by using the mutagenized forward primer (5 -GTGTGATGTGACTTTACTTGAGTT-ATC-3 ) and the reverse primer (5 -AGGAGTCCTGGCTCCATCTCT-3 ) that introduced a TaqI restriction site in the G allele (189 and 26 bp).

We compared the results obtained in the present study with three published studies: (i) Swystun et al. [9] reported the TCIWorks PK analysis of 43 pediatric HA patients infused with r-FVIII products; (ii) Garcia-Martinez et al. [17] reported the myPK-Fit PopPK analysis of 43 pediatric/adult HA patients infused with r-FVIII (ADVATE) products; (iii) Ogiwara et al. [18] reported the PKRD (PharmacoKinetics Repeated Doses) or the TCIWorks PK analysis of 43 adult HA patients infused with r-FVIII (80%) and pd-FVIII (20%) products.

PK parameter values reported for the most frequent rs868875 AA and AG genotypes [9, 18] were compared in Italian patients. The constant for the elimination rate from the central compartment K 1-0 (1/h), K 1-0 half-life (K 1-0 HL, h), and clearance (mL/h) were compared with the following assumptions and/or limitations: (i) K, half-life, and clearance were according to TCIWorks [9, 18] and myPK-Fit [17] , whereas in the present study, K 1-0, K 1-0 HL, and clearance were according to the 2 CP model PK; (ii) in Garcia-Martinez et al. [17] , the G-allele-related increments are reported for a single allele and for the homozygous GG condition; (iii) clearance data were compared after adjustment for a mean weight of 70 kg.

All statistical analyses were performed using IBM ® SPSS ® Statistics version 23.0 software (IBM Corp., Armonk, NY, USA). Continuous variables with normal distribution (MRT and AUC) were expressed as means with standard error (SE). Skewed variables were logarithmically transformed, and means with SE were reported. Genotype-related differences in PK parameters were analyzed by t-test, Mann-Whitney test, or analysis of variance (ANOVA) for linear trend, as appropriate. The contribution to PK parameters of CLEC4M rs868875 (G-carriers vs. AA), ABO (O vs. non-O) genotypes, age, and VWF:Ag levels were evaluated by linear regression analysis, and the interaction between genotypes was estimated as p (int) by generalized linear model.

The relationship between the CLEC4M rs868875 genotypes and FVIII PK parameters was investigated in the cohort of Italian HA patients (n = 26) infused with pd-and FL r-FVIII products (54 PKs).

In patients grouped by genotypes (AA, n = 12; AG, n = 12, and GG, n = 2), values of PK parameters displayed significant differences for the final elimination rate constant K 1-0 (p < 0.001, ANOVA for linear trend), for the transfer from the central to peripheral plasma compartment rate constant K 1-2 (p = 0.049), the secondary Beta (p = 0.030), and K 1-0 HL (p = 0.011). A trend was observed for the Beta HL (p = 0.054) ( Table 1) .

Patient carriers of the G allele (n = 14) showed K 1-0 values (0.11 ± 0.03 1/h SE) higher than AA homozygotes (0.06 ± 0.00 1/h SE, p = 0.045).

When the analysis was restricted to the patients who underwent at least one PK with a pd-FVIII (n = 22, Table S1 ), PK parameters in genotyped patients (AA, n = 9; AG, n = 11; GG, n = 2) displayed several and significant differences, also for the Alpha distribution phase (Alpha, p = 0.033; Alpha HL, p = 0.007).

Patients grouped by G-allele carriership (n = 13) showed K 1-0 HL values (mean 8.07 ± 0.92 h SE) 28% (approximately 3 h) shorter (p = 0.062) than AA-homozygotes (mean 11.2 ± 1.07 h SE).

ABO blood groups are a well-known modulator of FVIII PK [1, 8, 19, 20] . Accordingly, in the Italian cohort, the ABO genotype groups (O, non-O) showed significant differences for several parameters (Table S2) .

We compared the influence on FVIII PK of the CLEC4M and ABO genotypes in linear regression models of PK variables associated with CLEC4M genotypes. The contribution of the CLEC4M rs868875 polymorphism remained significant for K 1-0, Beta, K 1-0 HL, and, as a trend, for Beta HL PK parameters ( Table 2) . For the Beta, Beta HL, and K 1-2, the contribution of the ABO genotypes prevailed ( Table 2) . We observed significant interaction between genotypes, particularly for K 1-0, K 1-0 HL, and Beta HL ( Table 2) .

The significant contribution of CLEC4M rs868875 A/G and ABO genotypes was maintained after inclusion in the model of age and VWF:Ag levels at PK (Table S3 ). The influence of age was observed as a trend for the Beta parameters.

VWF:Ag levels were not associated with CLEC4M G-carriership (AA, 135 ± 9.6% SE; AG + GG, 122 ± 7.3% SE).

Prompted by these observations, we report the distribution of K 1-0 HL ( Figure 1A ) in the 26 patients grouped by combination of CLEC4M and ABO genotypes. The G-carriers/O blood group genotypes showed significantly shorter K 1-0 HL values than all the other genotypes groups. In contrast, in the AA homozygotes, the K 1-0 HL did not differ in relation to the ABO genotypes ( Figure 1A) .

To favor comparison with previous studies, we also analyzed the clearance distribution in the CLEC4M/ABO genotypes groups. Although with overlapping of values, mean clearance of infused FVIII was higher (p = 0.038) in the G-carriers/O blood group (mean 7.13 ± 2.23 mL/h/kg SE) than in the G-carriers/non-O (mean 2.40 ± 0.34 mL/h/kg SE) ( Figure 1B ). Prompted by these observations, we report the distribution of K 1-0 HL ( Figure 1A ) in the 26 patients grouped by combination of CLEC4M and ABO genotypes. The G-carriers/O blood group genotypes showed significantly shorter K 1-0 HL values than all the other genotypes groups. In contrast, in the AA homozygotes, the K 1-0 HL did not differ in relation to the ABO genotypes ( Figure 1A) .

To favor comparison with previous studies, we also analyzed the clearance distribution in the CLEC4M/ABO genotypes groups. Although with overlapping of values, mean clearance of infused FVIII was higher (p = 0.038) in the G-carriers/O blood group (mean 7.13 ± 2.23 mL/h/kg SE) than in the G-carriers/non-O (mean 2.40 ± 0.34 mL/h/kg SE) ( Figure  1B ). 

Previous studies have detected association of the CLEC4M rs868875 A/G polymorphism with clearance [9, 17, 18] , and with half-life [17] FVIII PK parameters (Table 3 ), detected by different PK analysis approaches. Aimed at improving our knowledge of main components of the association between CLEC4M genotypes and FVIII PK parameters, which could provide useful information for HA patients' treatment, we contribute a two compartment (2CP) model PK analysis of patients with severe/moderate HA.

We compared the association of the rs868875 CLEC4M genotypes with FVIII PK parameters in the different studies (Table 3) , with reasonable assumptions permitting value paralleling. A substantial concordance of PK parameter distribution was observed in relation to the CLEC4M genotypes in HA patients, particularly for the elimination constant 

Previous studies have detected association of the CLEC4M rs868875 A/G polymorphism with clearance [9, 17, 18] , and with half-life [17] FVIII PK parameters (Table 3 ), detected by different PK analysis approaches. Aimed at improving our knowledge of main components of the association between CLEC4M genotypes and FVIII PK parameters, which could provide useful information for HA patients' treatment, we contribute a two compartment (2CP) model PK analysis of patients with severe/moderate HA.

We compared the association of the rs868875 CLEC4M genotypes with FVIII PK parameters in the different studies (Table 3) , with reasonable assumptions permitting value paralleling. A substantial concordance of PK parameter distribution was observed in relation to the CLEC4M genotypes in HA patients, particularly for the elimination constant rate (K/K 1-0) and the half-life/K 1-0 HL. FVIII clearance was found to be 25-40% increased in G-carriers, a noticeable observation for the critical parameter to tailor prophylaxis or continuous infusion. This observation was coherent with the lower K/K 1-0 elimination rate constant, and with the shorter HL/K 1-0 HL values, albeit with one discrepancy (Ogiwara et al. [18] , Table 3 ). A comparison of parameter values in the different studies suggests that the presence of at least one G-allele may shorten the FVIII half-life. We believe that the observation of similar results in several studies with small size HA cohorts strengthens these findings. Table 3 . CLEC4M rs868875 genotypes and FVIII PK parameters.

Comparison of association between the CLEC4M rs868875 genotypes and FVIII PK parameters estimated from published studies. G-allele-related increments estimated from Garcia-Martinez et al. are reported. r-FVIII, recombinant FVIII; pd-FVIII, plasma derived FVIII concentrates. • Values obtained with only pd-FVIII. § after adjustment for a mean weight of 70 kg. K, half-life, and clearance according to TCIWorks [9, 18] and myPK-Fit [17] ; K 1-0, K 1-0 HL, and clearance of the present study according to WinNonlin 2CP model. K 1-0, the elimination rate constant from the central compartment. For the present study: p, ANOVA; * p, t-test analysis. ns, not significant value. The arrows indicate decreased (↓) or increased (↑) values in relation to the AA genotypes, and are meant to quickly compare parameter values in the different PK studies. Double arrows, statistically significant association; single arrow, not significant (ns).

Differences in FVIII half-life could be amplified in adults, whereas those in clearance could be more prominent in children and young individuals [17] .

In the Italian patients, linear regression analysis suggested both ABO and CLEC4M genotypes as significant predictors of FVIII-PK, which displayed significant interaction to influence the Beta and elimination rates, independently from age at PK. The effects of the G-allele carriership, a genetic approach that strengthens our findings, were particularly pronounced in patients belonging to the O blood group. The combined genotype CLEC4M rs868875 G-carrier/ABO O, detectable in 15% of patients, produced significantly shorter FVIII PK half-life and increased clearance, major results of our study. Since the mean FVIII half-life was approximately half-reduced as compared with all other combined genotypes, and was reflected in a largely increased clearance, this finding deserves to be explored for personalized replacement therapy/prophylaxis. Since our observations were obtained in a limited number of HA patients, and the infrequent CLEC4M rs868875 GG homozygous condition does not favor the exploration of the potentially higher influence of this genotype, confirmation in larger cohorts is needed. Provided that our findings were confirmed, the practicalities needed to implement the routine clinical practice in FVIII PK analysis in HA would not be particularly demanding in terms of genetic analysis, as in other diseases. Patients with the unfavorable genotype combination would benefit from infusion with a higher FVIII dose or extended half-life FVIII products.

It is worth noting that in cellular and animal models, CLEC4M binds and internalizes FVIII both in a VWF-dependent and -independent manner [15] . This would imply that the modulation on FVIII PK parameters by receptors encoded by different CLEC4M genotypes may be exerted both through binding to FVIII and to VWF-FVIII complex. As a consequence, studies aimed at disentangling FVIII-and VWF-specific PK components could be disfavored by the several genetic/acquired components modulating VWF expression, as supported by the absence of significant differences in VWF:Ag levels in patients grouped by CLEC4M genotypes in the present study, and in Swystun et al. and Garcia-Martinez et al. [9, 17] . In linear regression analysis, the effects of the CLEC4M and ABO genetic components were independent from the VWF:Ag levels, which, however, are strongly associated with the ABO genotypes.

Concerning the FVIII concentrates used for the PK analysis, we observed a remarkable influence of CLEC4M genotypes on PK analysis with pd-FVIII products. Although it is tentative to speculate that "natural" glycosylation of FVIII favors interaction with the receptor, comparison with other studies, which used only recombinant products [9, 17] , suggests that CLEC4M genotypes influence the PK of FVIII concentrates characterized by differences in glycan structure after recombinant expression. This observation will favor further comparison with extended half-life FVIII concentrates, which will require appropriately designed studies.

Overall, observations in patients recruited in several countries support that specific CLEC4M genotypes influence half-life and clearance of different FVIII concentrates infused in HA patients. Our analysis suggests that the combined CLEC4M G-carrier and ABO O genotypes contribute to a faster decay of infused FVIII in HA, which deserves to be explored for personalized replacement therapy/prophylaxis.

The following are available online at: https://www.mdpi.com/article/10.3 390/jcm11030733/s1, Table S1 : Analysis of association between plasma derived FVIII PK parameters and the CLEC4M rs868875 polymorphism. Table S2 : Analysis of association between FVIII PK parameters and the ABO blood group genotypes in the 26 severe/moderate HA patients. Institutional Review Board Statement: The study, an investigator-initiated institutional review board-approved retrospective chart review for optimization of replacement treatment, was conducted according to the guidelines of the Declaration of Helsinki, and approved (date, June 2013) by the Institutional Review Board of the Center for Bleeding Disorders, Careggi University Hospital, Florence, Italy.

Informed consent for the study inclusion was obtained from all subjects involved in the study.

Data Availability Statement: Data supporting reported results can be obtained by request to Dr M. Morfini (drmassimomorfini@gmail.com).

M.M. reports personal fees from Kedrion, grants from Pfizer, personal fees from Novo Nordisk, personal fees from SOBI, personal fees from Bayer, personal fees from Bioverativ, personal fees from Octapharma, personal fees from CSL Behring, outside the submitted work. G.C. reports personal fees from Roche, personal fees from Bayer, personal fees from Shire, grants and personal fees from CSL Behring, grants and personal fees from Sobi, personal fees from Uniqure, grants from Pfizer, personal fees from Kedrion, personal fees from Novo Nordisk, personal fees from Werfen, outside the submitted work. F.B. reports grants from Bayer, during the conduct of the study; grants from Pfizer, outside the submitted work. The other authors state that they have no conflict of interest.

The Half-Life of Infused Factor VIII Is Shorter in Hemophiliac Patients with Blood Group O than in Those with Blood Group A

Juhan-Vague, I. Biological and Genetic Factors Influencing Plasma Factor VIII Levels in a Healthy Family Population: Results from the Stanislas Cohort

Factor VIII Half-Life and Clinical Phenotype of Severe Hemophilia A

Advate Clinical Program Group; et al. Comparative Pharmacokinetics of Plasma-and Albumin-Free Recombinant Factor VIII in Children and Adults: The Influence of Blood Sampling Schedule on Observed Age-Related Differences and Implications for Dose Tailoring

Polymorphisms at LDLR Locus May Be Associated with Coronary Artery Disease through Modulation of Coagulation Factor VIII Activity and Independently from Lipid Profile

Factor VIII Requirement to Maintain a Target Plasma Level in the Prophylactic Treatment of Severe Hemophilia A: Influences of Variance in Pharmacokinetics and Treatment Regimens

iPATH study group Biological Mechanisms Underlying Inter-Individual Variation in Factor VIII Clearance in Haemophilia

Functional Polymorphisms in the LDLR and Pharmacokinetics of Factor VIII Concentrates

Genetic Determinants of VWF Clearance and FVIII Binding Modify FVIII Pharmacokinetics in Pediatric Hemophilia A Patients

The Asialoglycoprotein Receptor Minor Subunit Gene Contributes to Pharmacokinetics of Factor VIII Concentrates in Hemophilia A

N-Linked Glycosylation Modulates the Immunogenicity of Recombinant Human Factor VIII in Hemophilia A Mice

Comparative Glycosylation Mapping of Plasma-Derived and Recombinant Human Factor VIII

Mapping the N-Glycome of Human von Willebrand Factor

An Insight into Glyco-Microheterogeneity of Plasma von Willebrand Factor by Mass Spectrometry

The Endothelial Lectin Clearance Receptor CLEC4M Binds and Internalizes Factor VIII in a VWF-Dependent and Independent Manner

Lectins Enhance SARS-CoV-2 Infection and Influence Neutralizing Antibodies

Common Genetic Variants in ABO and CLEC4M Modulate the Pharmacokinetics of Recombinant FVIII in Severe Hemophilia A Patients

Factor VIII Pharmacokinetics Associates with Genetic Modifiers of VWF and FVIII Clearance in an Adult Hemophilia A Population

Parameters Influencing FVIII Pharmacokinetics in Patients with Severe and Moderate Haemophilia A. Haemophilia

The Importance of ABO Blood Group in Pharmacokinetic Studies in Haemophilia A

Accuracy of FVIII: C Assay by One-Stage Method Can Be Improved Using Hemophilic Plasma as Diluent

Validation of a Rapid Test (VWF-LIA) for the Quantitative Determination of von Willebrand Factor Antigen in Type 1 von Willebrand Disease Diagnosis within the European Multicenter Study MCMDM-1VWD

The Italian AICE-Genetics Hemophilia A Database: Results and Correlation with Clinical Phenotype. Haematologica

Single-Tube Polymerase Chain Reaction for Rapid Diagnosis of the Inversion Hotspot of Mutation in Hemophilia A