Mutation Spectrum and Genotype–Phenotype Analyses in a Pakistani Cohort With Hemophilia B

Abstract

This study aimed to (1) identify F9 genetic alterations in patients with hemophilia B (HB) of Pakistani origin and (2) determine the genotype–phenotype relationships in these patients. Diagnosed cases of HB were identified through registries at designated tertiary health-care centers across the country. Consenting patients were enrolled into the study. The factor IX (FIX) coagulation activity (FIX:C) and key clinical features were recorded. Direct sequencing of F9 was carried out in all patients. All the variants identified were analyzed for functional consequences employing in silico analysis tools. Accession numbers from National Center of Biotechnology Information ClinVar database were retrieved for the novel variants. Genotype–FIX:C relationships were determined followed by FIX:C clinical phenotype assessment. A total of 52 patients with HB from 36 unrelated families were identified, which mainly comprised patients with moderate HB (n = 35; 67.3%). Among these, 35 patients from 22 unrelated families could be contacted and enrolled into the study. Missense variants were the most frequent (58.8%), followed by nonsense variants (17.6%). A missense, a short insertion, and a nonsense novel variants in exon 2, 6, and 7, respectively, were also identified. The disease manifested FIX:C heterogeneity in relation to the corresponding mutation in a significant number of cases. Clinical phenotype heterogeneity was also observed in relation to FIX:C-based severity assessment. We concluded that the registered FIX-deficient population of Pakistan mainly comprises moderate HB. F9 mutation spectrum in Pakistani patients with HB is heterogeneous. The HB population of Pakistan manifests a significant amount of genotype–FIX:C and FIX:C–clinical phenotype heterogeneities.

Introduction

Factor IX (FIX) deficiency or hemophilia B (HB), a coagulopathy, is among the first ones to have been studied genetically. Since its identification in 1952, a total of 1253 mutations have been reported in the FIX gene (F9) till date.¹ The spectrum of mutations that has been described is heterogeneous globally.²

Mutations have been identified in all FIX domains and auxiliary segments. These are mostly point mutations (73.0%), missense in the majority of the cases (74.4%).³ Simultaneous short insertions and deletions (indels) constitute about 1.5% of the mutations, whereas gross gene deletions account for about 16.3% of the cases. Another group of mutations that occur in the promoter region of the FIX gene result in HB Leyden.⁴ A significant number of FIX deficiency cases are due to founder mutations.⁵

HB also exhibits phenotypic heterogeneity, which is generally described at 2 levels. In the first case, patients bearing the same levels of FIX coagulation activity (FIX:C) show different clinical outcomes in terms of severity.⁶ The second disparity lies among cases with identical F9 mutations. Patients with similar mutations have been reported to bear significantly different levels of FIX:C assigning them into different severity classes.^7,8

The World Federation of Hemophilia (WFH) Annual Global Survey 2014 reports a total of 549 cases of hemophilia in Pakistan, comprising 469 patients with hemophilia A (HA) and 80 patients with HB. For a total population of 196174380, this translates to be a prevalence of 0.44 and 0.076 per 100000 males, respectively.⁹ Figures for prevalence of HB in Pakistan, reported annually by WFH since 1998, are very inconsistent, ranging from 2.02 in 2000 to 0.06 in 2003, the reason being lack of a centralized national registry.¹⁰ To date, only 1 study has been performed on genetic analysis of patients with HB from Pakistan, identifying 11 different mutations.¹¹ The current study broadens spectrum of mutations in Pakistani patients with HB. In addition, the study provides a further insight into the genotype–FIX:C and the FIX:C–clinical phenotype heterogeneity.

Materials and Methods

Study Participants

The study was approved by the ethical committee of Khyber Medical University, Pakistan. Eligible patients were those with documented FIX deficiency based on FIX:C levels. In the preliminary phase of the study, patients with HB were identified through the hemophilia registries maintained at 10 designated health-care centers in Islamabad (the federal capital) and the provincial capitals of Punjab, Khyber Pakhtunkhwa, and Sindh. The patients were contacted through their phone numbers recorded in the registries. Consenting patients were enrolled into the study. Those with active concomitant medical illness were excluded.

Clinical Severity

A comprehensive questionnaire depicting the demographic and clinical details was filled out for each patient by a previously trained physician in a face-to-face interview. The clinical assessment parameters included age at first bleed, number of target joints (defined as joints that had ≥3 spontaneous bleeds within any consecutive 6-month period), number of permanently disabled or disfigured joints (objectively perceivable), annual bleed rate (ABR; number of bleeds experienced in the last 1 year), current primary treatment modality, and the common treatment regimen over the last 1 year.

Blood Sample Collection and Processing

A volume of 8 to 10 mL of blood was collected in plain, EDTA, and sodium citrate–containing collection tubes BD Vacutainer (Becton, Dickinson, Plymouth, UK), following the guidelines of the Clinical and Laboratory Standards Institute (CLSI).¹² Plasma was separated by centrifuging blood at 1700g for 15 minutes at room temperature. The supernatant plasma was extracted and put into aliquots of 1-mL Eppendorf tubes (Eppendorf, Hamburg, Germany) and stored at −70°C until final analysis. The CLSI guidelines (document H21-A5) were adopted while processing and transporting the samples.¹³ The samples were analyzed at National Institute of Blood Diseases, Karachi.

Hemostasis Investigations

The diagnoses of enrolled patients were reconfirmed at a centralized reference laboratory. First-line coagulation tests (prothrombin time, activated partial thromboplastin time [aPTT], and fibrinogen assay) and FIX activity (FIX:C) assay were performed on STA Compact Max (Stago, Parsippany, New Jersey) hemostasis analyzer. The machine employs 1-stage aPTT-based assaying technique in measuring FIX:C. All the samples were cross-checked on Sysmex CA 1500 coagulation analyzer (Sysmex, Kobe, Japan) to establish reliability. The FIX:C deficiency was classified as mild (FIX:C 5%-40%), moderate (FIX:C 1%-5%), and severe (FIX:C <1%). Inhibitor screen was performed by aPTT-based 1:1 mixing test, as described by Dubey et al.¹⁴

Mutation Detection

Genomic DNA was extracted using QIAamp Blood DNA Mini Kit (Qiagen Ltd, Crawley, United Kingdom). Each of the 8 exons along with flanking intronic regions, the promoter region, and polyadenylation site were amplified in 11 polymerase chain reaction (PCR) reactions. The product was then sequenced on an ABI Prism 3500 Genetic Analyzer (Applied Biosystems, Foster City, California, USA). Variation in sequence was confirmed in PCR-amplified products of both forward and reverse primers, processed independently. Sequences were compared with the normal/wild-type F9 sequence (GeneBank accession no. K02402), using Clustal Omega alignment tool (http://www.ebi.ac.uk/Tools/msa/clustalo/).¹⁵

Where identified, the variants were looked for in affected (where available) and healthy family members, and co-segregation was confirmed. The identified mutations were searched in the FIX mutation databases, that is, CDC Hemophilia B Mutation Project, Factor IX Gene Variant Database (FIXVDB), and Human Gene Mutation Database.^1,3,16 All the variants (novel and previously known) were checked for pathogenicity according to the guidelines provided by United Kingdom Haemophilia Centre Doctor’s Organisation.¹⁷ Variants were described per Human Genome Variation Society guidelines.¹⁸

In Silico Analyses

Pathogenicity of all missense variants found in the study (including previously known) was predicted using the online tools Polymorphism Phenotyping v2 (PolyPhen-2) and Sorting Intolerant from Tolerant (SIFT).^19,20 These were assessed with Ensembl’s Variant Effect Predictor tool as well.²¹ The effect of splice site variants were predicted with the help of Human Splicing Finder (HSF) on the basis of HSF Matrices (based on Position-Based Matrices) and MaxEntScan (Maximum Entropy) algorithms.²² Structural analyses of proteins were carried out using I-TASSER online software.²³ The models retrieved from I-TASSER were viewed using Molsoft ICM-Browser (http://www.molsoft.com/icm_browser.html). To confirm the conservation of variants, comparative analysis of F9 orthologs among multiple species was performed.

Statistical Analyses

Frequencies were calculated in case of categorical data; mean values and standard deviation (SD) were determined for continuous normally distributed data. Comparisons of clinical severity parameters among various HB severity classes were carried out employing exact Kruskal-Wallis test. The statistical analyses were performed with SPSS v22.0 (SPSS Inc, Chicago, Illinois).

Results

Selection of Patients

A total of 52 patients with HB, from 36 unrelated families, were identified. These mainly comprised of patients with moderate HB (n = 35; 67.3%), followed in respective order of frequency by those with severe (n = 11; 21.2%) and mild (n = 6; 11.5%) disease. A total of 41 patients could be contacted, 3 had active hepatitis C infection and were under treatment for it; 35 patients from 20 unrelated families agreed to participate and were enrolled into the study after acquisition of informed consent in writing. The rest either had not left any contact information or could not be reached over the provided phone numbers.

Phenotypic Assessment

The mean age of the study patients was 13.25 (10, SD) years, with a range of 1 to 33 years. All the cases had an isolated prolonged aPTT. The FIX:C assay revealed 6 (17.1%) cases of severe, 26 (74.2%) cases of moderate, and 2 (5.7%) cases of mild disease. None of the patients had evidence of an FIX inhibitor.

The ABR and the total number of days under treatment in the last 1 year in those with mild disease were significantly lesser in comparison to those with moderate and severe disease (Table 1). The cases with mild HB had an accordingly milder clinical severity of the disease. The cases with moderate and severe HB did not have any significant statistical difference, on cumulative scale, for any of the mentioned clinical parameters.

Table 1.

FIX:C Phenotypes in Relation to Clinical Severity Parameters.

Clinical Parameter	HB Severity (n)	Median	Mean Rank	P Valuea
Age, years	Mild (2)	8	14.25
	Moderate (27)	11	18.33	.87
	Severe (6)	10	17.75
Age at first bleed, months	Mild (2)	15.5	18.25
	Moderate (27)	6.0	16.93	.476
	Severe (6)	12.0	22.75
Annual bleeding rate	Mild (2)	1.5	2.25
	Moderate (27)	10.0	19.63	.042
	Severe (6)	4.5	15.92
Total no. of days under treatment	Mild (2)	2.0	1.50
	Moderate (27)	16.0	19.67	.030
	Severe (6)	8.5	16.00
No. of disabled joints	Mild (2)	0	8.50
	Moderate (27)	1	18.63	.315
	Severe (6)	0.5	15.58
Target joints	Mild (2)	0	8.50
	Moderate (27)	1	18.74	.315
	Severe (6)	0.5	17.83

Abbreviations: HB, hemophilia B; n, number of patients; No., number.

^aP values refer to exact Kruskal-Wallis test with .05 as the level of significance.

Fresh frozen plasma (FFP) was the primary therapeutic modality in 22 patients, cryosupernatant (CS) in 8 patients, plasma-derived factor IX concentrates in 3 patients, and fresh whole blood in 2 patients. The treatment regimen was “on demand” in 28 cases, while it was “tertiary prophylactic” (ie, regular treatment commenced after the onset of joint disease) in 7 cases (5 moderate and 2 severe). Of particular interest were the 4 cases with severe HB treated on demand with FFPs, CS, and fresh whole blood. None of the participants reported spontaneous articular bleeding except for those with target joints.

Mutation Analysis

The F9 was sequenced for the index patient in each family. A total of 17 different mutations were found in 34 patients from 21 families (Table 2). Point mutations were identified in 13 (76.5%) cases; 3 of these were nonsense, whereas 10 were missense mutations. One family with 7 patients were found to have frameshift mutation in exon 6 with a premature stop codon. In 2 first cousins, from another family, a splice site mutation was found in intron 4. Two brothers with moderate HB had a branch point mutation. Three mutations found in 4 individual patients were novel.

Table 2.

Description of 17 Different F9 Mutations Identified in 34 Pakistani Patients With Hemophilia B.^a

Nucleotide Change (HGVS)	Consequence	Location Exon/Domain	CpG	Nature of Mutation	FIX: C Levels, %	No. of Patients/Family b	Conservationc (Amino Acid)	SIFT Score	PolyPhen2 Score
c.127C>T	p.Arg43Trp	2/Propeptide	Yes	Missense	2.0, 12.0	2/2	RRRRRRRRR	0	0.977
c.141T>A	p.Tyr47*	2/Gla	No	Nonsense	1.8	1/1	–	–	–
c.160G>A	p.Glu54Lys	2/Gla	No	Missense	1.0	1/1	EEEEEEEEE	0	0.924
c.223C>T	p.Arg74*	2/Gla	Yes	Nonsense	1.1, 1.3	2/1	–	–	–
c.253-25A>G	N/A	Intron 2	No	Branch point	1.5, 1.3	2/1	–	–	–
c.391+1G>A	N/A	Intron 4	No	Splice site	1.1, 1.2	2/1	–	–	–
c.415G>A	p.Gly139Ser	5/EGF-2	No	Missense	1.2, 0.9	2/2	GGGGGGGGG	0	1
c.427C>A	p.Gln143Lys	5/EGF-2	No	Missense	1.3, 1.4, <0.01, <0.01, <0.01	5/2	QQQQQHHHH	0.02	0.277
c.434G>A	p.Cys145Tyr	5/EGF-2	No	Missense	2.6	1/1	CCCCCCCCC	0	1
c.519dupA	p.Val174SerFx13	6/Linker	No	Frameshift- truncation	0.9, 0.8	2/1	–	–	–
c.540-541delAG	p.Arg180SerFx8	6/Linker	No	Frameshift- truncation	1.9, 1.6, 1.4, 1.4, 1.6, 1.4, 1.5	7/1	–	–	–
c.756T>G	p.Cys252Trp	7/S-Protease	No	Missense	1.0	1/1	CCCCCCCCC	0	1
c.872A>G	p.Glu291Gly	8/S-Protease	No	Missense	2.3	1/1	EEEEEEEEE	0	1
c.1064G>T	p.Gly355Val	8/S-Protease	No	Missense	1.2	1/1	GGGGGGGGG	0	1
c.1150C>T	p.Arg384*	8/S-Protease	Yes	Nonsense	1.8	1/1	–	–	–
c.1157C>G	p.Thr386Arg	8/S-Protease	No	Missense	2.2	1/1	TTTTTTSSS	0	0.993
c.1214A>G	p.Asp405Gly	8/S-Protease	No	Missense	2.0, 1.5	2/2	DDDDDDDDD	0	1

Abbreviations: CpG, 5′-cytosine-phosphate-guanine-3′ dinucleotide sequence; No., Number.

^aLight gray color highlights the 3 novel mutations found in the current study.

^bData from the current study.

^cAmino acid residue conserved through human, chimpanzee, dog, rhesus monkey, Norway rat, chicken, zebra fish F9a, zebra fish F9b, and frog (in respective order).

Mutations were found throughout the gene with no specific hot spot, mainly in the serine protease domain. Sequences were also inspected for neutral variants. None of the patients was found to have more than 1 mutation. In an isolated case, we could not identify any variation in sequence except for the Malmo polymorphism (c.580A>G).

Four missense mutation recurred, in 2 unrelated families each time. These families belonged to different ethnic subgroups settled in different geographical areas. Their family histories did not indicate any relationship. Disease severity was found to be identical for the pair on only 1 occasion. The other 3 mutations, however, did not strictly correlate with the phenotype (Table 3).

Table 3.

Phenotypic Heterogeneity^a in Mutations From the Current Study.

S. No.	Mutation	Disease Severityb									Remarks
		In the Current Study, n			In Asian Studies, n			In Non-Asian Studies, n
		m	M	S	m	M	S	m	M	S
1	c.127C>T	1	1	0	0	>5	>5	1	>5	>5	First m phenotype in Asia reported in the current study
2	c.141T>A	0	1	0	0	0	0	0	0	0	Novel mutation
3	c.160G>A	0	1	0	0	0	0	0	1	0	Novel in AsiaCurrent study supports previous information
4	c.223C>T	0	2	0	0	0	>5	0	>5	>5	Current study confirms GPH in Asia
5	c.252-25A>G	0	2	0	0	0	0	0	1	2	Novel in Asia
6	c.391+1G>A	0	2	0	0	0	1	0	0	5	GPH first reported in the current study
7	c.415G>A	0	1	1	0	1	2	1	3	3	Current study supports previous information
8	c.427C>A	0	2	3	0	0	1	0	0	1	GPH first reported in the current study
9	c.434G>A	0	1	0	0	0	1	0	1	1	Current study confirms GPH in Asia
10	c.519dupA	0	0	2	0	0	0	0	0	0	Novel mutation
11	c.540-541del AG	0	7	0	0	0	1	0	0	1	GPH first reported in the current study
12	c.756T>G	0	1	0	0	0	0	0	0	0	Novel mutation
13	c.872A>G	0	1	0	0	0	0	2	3	0	Novel in Asia
14	c.1064G>T	0	1	0	0	0	0	0	0	1	Novel in AsiaGPH first reported in the current study
15	c.1150C>T	0	1	0	0	2	0	0	0	4	M phenotype in Asia vs severe phenotype in non-Asia
16	c.1157C>G	0	1	0	0	0	0	0	0	1	Novel in AsiaGPH first reported in the current study
17	c.1214A>G	0	2	0	0	0	0	0	1	0	Novel in Asia

Abbreviations: m, mild; M, moderate; n, number of patients, GPH, genotype–phenotype heterogeneity; S, severe.

^aSame mutation expressing different disease phenotypes in terms of severity. ^bDisease severity is described in terms of factor IX activity (FIX: C).

Novel mutations

Mutations are a natural insight into genotype–phenotype relationship. Novel mutations, confirmed for their pathogenicity, are always an addition to the biochemical premise of a genetic disease. In the current study, we identified 3 novel mutations. Two of these are substitution point mutations. Variant c.756T>G occurred in catalytic domain of the protein. The patient did not have any other mutation in any of the other exons or flanking intronic regions. The promoter region was also screened for any variation in sequence and was found to have none. Comparative analysis of F9 orthologs among multiple species shows conservation of the Cys252 residue (Table 2). PolyPhen-2 and SIFT analyses also confirmed pathogenicity for the variant. The patient had moderate HB (FIX:C 1.0%), with no inhibitor. Figure 1 elicits this novel variant in comparison to the contemporary wild type.

Figure 1.

Genetic and bioinformatic aspects of novel c.756T>G (p.Cys252Trp) variant in comparison to wild type. A, I-TASSER-derived assessment of secondary protein structure and accessibility of amino acid residues of mutant and wild type. Solid arrow head points at residue 252 (residue 26 in catalytic domain) of FIX protein. B, Ribbon model of wild-type FIX 235 residue catalytic domain displaying cysteine at residue number 26 (C26). Dashed circle focuses beta sheet (arrow plate) in the vicinity. C, Predicted protein model of mutant catalytic domain with W26 substitution shown in stick and ball configuration. Dissolution of beta sheet can be appreciated in the area encircled. D, Mesh representation of mutant catalytic domain. Tryptophan residue (W26) is buried deep, revealed only when toggled.

The second novel variant c.141T>A in exon 2 is nonsense (p.Tyr47*) in character. It occurs at the first amino acid residue of Gla domain of FIX. The patient had moderate HB (FIX:C level 1.8%). FIX inhibitor was not detected. No other mutation was identified in this patient.

Two brothers with severe HB (FIX:C 0.9% and 0.8%) were found to carry a novel F9 variant c.519dupA. This mutation results in frameshift in exon 6 at 174th codon. The insertion was confirmed by sequencing in either directions. No other variation was found in the coding or flanking F9 sequences. Both the patients lacked an inhibitor.

The 3 novel variants were reported to National Center of Biotechnology Information (NCBI) ClinVar database (www.ncbi.nlm.nih.gov/clinvar/). Accession numbers SCV000494046, SCV000494048, and SCV000494048 were assigned to c.141T>A, c.519dupA, and c.756T>G, respectively.

Genotype–Phenotype Heterogeneity

Besides the well-known genetic heterogeneity, HB also exhibits genotype–phenotype heterogeneity, where the same mutation may express multiple phenotypes or disease severities. In the current study, we report genotype–phenotype heterogeneity in 5 different cases for the first time (Table 3). In another 7 cases, the heterogeneity was reported previously, and in 3 cases, the mutations were novel. In 2 cases, heterogeneity could not be elicited, and the findings agreed to those published earlier. Two mutations (c.415G>A and c.427C>A) were found to associate with moderate and severe phenotypes among participants from the current study. Four of 5 patients with mutation c.427C>A were from 2 separate parts of the same family.

Discussion

Globally, around 13% to 33% of the patients with HB are reported to have moderate severity of the disease.^9,24 The current study, however, shows a significantly higher frequency of moderate HB (73.5%) in local population. This increased propensity of moderate HB might relate to the phenomenon of natural selection. Lack of treatment facilities in the developing countries is an adverse survival factor for those with severe phenotype.²⁵ An on-demand treatment regimen in conjunction with FFPs and fresh whole blood as the primary treatment modalities endorse a high mortality rate in settings of severe HB.²⁶

Another explanation might be the fact that 69.5% of the cases with HB reported by the low-income countries lack classification into a severity class.⁹ Appropriate categorization of this nonclassified group might change the current picture.

Regarding frequency of mutation types, our findings are comparable with those from the FIXVDB, with point mutations constituting the majority, comprising mainly missense mutations.³ Deletion and splice site mutations are the next most common.

Recurrent mutations are not uncommon in HB. In some studies, founder effect has been identified to cause this,^27,28 yet other studies advocate a de novo occurrence of these repeats.²⁹ The 4 recurrent mutations found in the present study are seemingly de novo. Haplotype analyses, however, could not be performed to categorically rule out any ancestral ties. Further studies are needed to investigate the true nature of these mutations.

The F9 variant p.Cys252Trp, identified in the current study, is not listed in the data available for normal population in single-nucleotide polymorphism (SNP) databases of NCBI (http://www.ncbi.nlm.nih.gov/projects/SNP/; accessed July 22, 2016) and International HapMap Project (http://www.hapmap.org/; accessed July 22, 2016). Cys-252 is a buried residue in the catalytic domain of FIX which has a disulfide (S-S) bond, with the next cysteine residue Cys-258 in the same domain. S-S bonds are known to have a role in folding and stabilizing proteins.³⁰ Tryptophan substitution at this point abolishes this bond, which is likely to sever geometry of the protein to the point where its catalytic activity is compromised enough to cause the disease. To negate the possibility of neutral variation, absence of this mutation in the normal Pakistani population has not been confirmed. However, absence of any other variation in coding or the promoter regions for the gene and absence in normal population SNP databases strongly support our view. The Cys-252 residue is strongly conserved across species. Finally, 3 other substitution mutations Cys252Tyr, Cys252Ser, and Cys252Arg have been previously reported at the same residue on 13 different occasions, with only 1 being reported from Asia (Turkey).^29,31–33 p.Cys252Tyr and p.Cys252Ser are reported to bear a severe phenotype, whereas p.Cys252Arg is found in association with all 3 phenotypes. Serine shares biochemical properties with cysteine and is thought to be relatively similar, yet a substitution leads to loss of function mutation. Tryptophan, biochemically comparable with tyrosine, is hence likely pathogenic if substituted for cysteine at this point.

Variant c.141T>A (p.Tyr47X), being nonsense, results in a truncated messenger RNA (mRNA). The probable sequela is a subsequent deletion of the product by nonsense-mediated mRNA decay pathway.³⁴ This prompts a severe phenotype. The protein product, lacking all vital protein domains, is believed to be inactive.

Alteration in the nucleotide sequence, observed in the novel c.519dupA variant, results in introduction of 13 unrelated new amino acids into the polypeptide chain, followed by a stop codon. This results in a truncated protein, with no catalytic domain. A severe phenotype is hence likely.

The current study adds to the existing pool of genotype–phenotype heterogeneity related to HB. Phenotypic variation is also reported in other monogenic hematological disorders as well.³⁵ In HB, for most of the times, it is possible to correlate the mutant genotype with a certain FIX:C phenotype. In several instances, however, FIX deficiency has been reported to associate with different severity of the disease against the same mutation.^7,8 The FIX variant databases show that a total of 36 F9 mutations have been reported to associate with all 3 clinical phenotypes of HB.¹⁶ Sixteen different F9 mutations have been found to associate with both severe and mild phenotypes, whereas 116 mutations are associated with severe and moderate phenotypes. Another group of 26 mutations reportedly associate with both mild and moderate severity of the disease.¹⁶

The high frequency of ABR, target joints, and impaired joints in patients with moderate severity is an outcome of inadequate management. Fresh frozen plasma, CS, and fresh whole blood cannot raise the FIX:C levels to the appropriate targets in settings of active bleeds.²⁶ This leads to frequent development of target joints culminating in arthropathies.

In the current study, a milder clinical phenotype is observed in several patients with severe FIX:C deficiency, rendering them indistinguishable from those with moderate severity. This is inconsistent with the established association of adverse clinical phenotype with severe FIX:C deficiency.³⁶ A significant number of patients with severe HB are, however, found to carry a milder clinical phenotype.⁶ The phenotypic variation is especially pronounced in those with moderate severity of the disease.^24,37,38

The basic concept of phenotypic heterogeneity, largely unknown, applies to both FVIII and FIX deficiencies, and any external influence necessarily has a similar bearing in either case. The phenomenon is hypothesized to be a function of multiple factors including modifier genes, epigenetic influences, and environmental effects.⁸ An association between factor V Leiden and milder clinical phenotype in an otherwise severe FVIII deficiency has been reported.³⁹ In another study on patients with HA having intron 22 inversion, almost half the patients had a milder phenotype.⁴⁰ The 2 groups did not significantly vary when compared for various coagulation, prothrombotic, and fibrinolytic factors. Increased levels of tissue plasminogen activator and thrombin-activatable fibrinolysis inhibitors are also found to associate with severe clinical outcomes in hemophilia.⁴¹

Owing to rarity of the disease and the exorbitant treatment costs, prophylactic management of moderate and severe HB with plasma-derived or recombinant FIX concentrates is a major challenge in the developing countries. Identification of factors modulating phenotypic heterogeneity may prove beneficial in management of such cases and will open new avenues in the treatment of HB.

Conclusion

We concluded that the majority of Pakistani population with HB has moderate FIX deficiency. The genetic spectrum for the disease is heterogeneous, mainly comprising missense mutations. The disease manifests a significant amount of genotype–phenotype and FIX:C–clinical phenotype heterogeneity in local HB population. Given the high propensity of moderate disease phenotype, the Pakistani population with HB provides an ideal ground for exploration of factors affecting the phenotypic heterogeneity.