Human platelet antigen 1, 2 and 5 gene polymorphisms in Egyptians and their potential association with susceptibility to immune thrombocytopenic purpura in Egyptian patients

ABSTRACT

Objectives: This study determined the incidence of HPA1, HPA2 and HPA5 polymorphisms in 120 Egyptian immune thrombocytopenic purpura (ITP) patients and 120 healthy Egyptian subjects.

Methods: Human platelet antigen (HPA) genotyping was done using the polymerase chain reaction-restriction fragment length polymorphism.

Results: The frequency of HPA1 allele a and b was 78.75 and 21.25% in controls, 80.8 and 19.2% in ITP, respectively. HPA2 allele a and b frequency was 86.25 and 13.75% in controls and of 74.6 and 25.4% in patients, respectively. HPA5 allele a and b frequency was 87.5 and 12.5% in controls, in patients it was 85 and 15%, respectively. With the exception of HPA2, no other significant difference was encountered in HPA allele frequency between controls and ITP patients.

Discussion: Egyptian HPA profile is closely linked to Middle East and neighboring Arabs. The current study noted that in all the studied HPA systems 1, 2 and 5, the ‘a’ allele is more prevalent than the b allele; the most frequent genotype was the homozygous a/a genotype. HPA2b frequency, homo- and hetero-zygous HPA2b genotype frequencies were significantly higher in ITP patients compared to controls.

Conclusion: HPA 2b are 2.37 times more likely to develop ITP compared to those without this allele. The relatively high allele frequency of the HPA-1b in the Egyptian population suggests that this ethnic group has a higher risk of alloimmunization.

KEYWORDS:

• ITP
• HPA polymorphism
• HPA allele/genotype frequency
• population comparison

Introduction

Immune thrombocytopenic purpura (ITP) is characterized by the production of autoantibodies targeting platelet membrane glycoproteins (GP) [1].

Diversity of human platelet antigens (HPA) is based on a single base pair substitution in the GP genes, HPAs assembled into six biallelic systems [2]. HPA are involved in neonatal alloimmune thrombocytopenia, post-transfusion purpura, platelet refractoriness, passive alloimmune thrombocytopenia, drug-induced and transplantation-associated thrombocytopenia [3].

A link between HPA allele polymorphisms and ITP was suggested [1]; however, studies in this regard are scarce. To date, the HPA polymorphisms were not assessed in context with ITP among Egyptians.

Consideration of the genetic amalgamation among Egyptians being centrally positioned to Africa, Europe and Asia together with the wide diversity of HPA genotypes in different ethnic groups such as Asian, European and North American inhabitants [4–6], has led us to design this study to determine the frequency of HPA polymorphisms (HPA-1, HPA-2 and HPA-5) in Egyptian ITP patients and in healthy Egyptian subjects, to assess the potential associations between HPA polymorphisms and the development of ITP and also to reflect the frequency of these HPA systems’ polymorphism among Egyptians.

Subjects and methods

This is a prospective, case control study of 120 Egyptian ITP patients and 120 age- and sex-matched healthy Egyptian subjects; as controls, informed consent was obtained from each participant/guardian. The study was approved by the ethics committee of Cairo University.

Detailed demographic, clinical and laboratory data of ITP patients are summarized in Table 1. Among the studied controls, the age ranged from 4 to 67 with a median age of 29.0 years, 45 (37.5%) below the age of 16 years, 60 (50%) were females.

Table 1. Clinical and laboratory data of the 120 studied ITP patients.

Variable	Value
Sex, number (%)
• Male	44 (36.7%)
• Female	76 (63.3%)
Age, years
• Median, range	20.29 years (2–63)
Age groups
• Children ≤16 years	54 (45.0%)
• Adults >16 years	66 (55.0%)
Platelet at sampling, ×10^9/l
• Mean ± SD	59.15 (6–180)
Hemoglobin at sampling, g/dl
• Median, range	11.01 (8–14)
TLC at sampling, ×10^3/µl
• Mean ± SD	9.74 (4–88)
Phase
• Acute (newly diagnosed)	52 (43.3%)
• Persistent	20 (16.7%)
• Chronic^a	48 (40%)
Disease state at sampling
• Activity	98 (81.7%)
• Remission	22 (18.3%)
Severity at sampling
• Bleeder	25 (20.8%)
• Non-bleeder	95 (79.2%)
Treatment
• Steroids	103(85.8%)
• IV Ig	5 (4.2%)
• Cyclosporine	1(0.8%)
Quality of laboratory response at follow-up (FU)^b
• Complete response	6/77 (7.8%)
• Partial response	70/77 (90.9%)
• No response	1/77 (1.3%)
Quality of clinical response at FU^b
• Response	71/74 (95.9%)
• No response	3/74 (4.1%)

^aLong FU ≥ 1 year.

^bShort FU, (3–6 months after diagnosis).

ITP was diagnosed according to the guidelines of the American Society of Hematology [7]. At the time of sampling, newly diagnosed; persistent and chronic ITP was defined according to the international working group [8]. The severity was assessed according to the absence or presence of significant bleeding, disease state (activity/remission) was defined according to Zhou et al. (2009) [9]. Assessment of laboratory response (Platelets count) to the applied line of management was applied according to the international working group [8].

HPA genotyping was done using the polymerase chain reaction-restriction fragment length polymorphisms as described before [10]. DNA was extracted using the DNAzol Reagent (GIBCO BRL Life Technologies, Eggenstein, Germany). HPA-1, 2 and 5 polymorphic sequences were amplified using the previously recommended primers, and protocol [10], then digested using Thermoscientific restriction enzymes; Nci I (Catalog number: FD0064), BsaH I (Catalog number: FD0474) and Mnl I (Catalog number: FD1074), respectively. Gel interpretation was done on gel documentation system (BioDoc, Analyser, Biometra, Gottingen, Germany) in reference to a molecular weight marker. Ten percent of the samples were amplified twice for the validation of results.

Results

Among the studied ITP patients and the control subjects, the current study noted that in the studied HPA 1, 2 and 5, the ‘a’ allele was more prevalent than the b allele; the most frequent genotype was the homozygous a/a genotype, less frequently was the heterozygous a/b genotype, while homozygous b/b genotype was very rare in all the studied HPA systems.

Comparison between controls and patients groups as regards HPA1 and HPA5 allele and genotype frequencies showed no statistically significant difference. While a statistically significant higher HPA 2 b allele frequency, and HPA 2b/2b and 2a/2b genotype frequencies were encountered in ITP patients compared to controls, detailed allele and genotype frequencies in healthy Egyptian subjects and ITP patients are shown in Table 2.

Table 2. HPA1, 2 and 5 allele, allele phenotype and genotype frequencies in patients and controls.

HPA alleles	Allele frequency			Allele phenotype frequency			HPA genotypes	Genotype frequency
	Patients	control	P-value	Patients	control	P-value		Patients	control	P-value
HPA1-a	194 (80.8%)	189 (78.75%)	0.570	115 (95.8%)	114 (95%)	0.758	HPA1a/1a	79 (65.8%)	75 (62.5%)	0.736
HPA1-b	46 (19.2%)	51 (21.25%)		41 (34.2%)	45 (37.5%)	0.590	HPA1a/1b	36 (30%)	39 (32.5%)
							HPA1b/1b	5 (4.2%)	6 (5%)
HPA2-a	179 (74.6%)	207 (86.25%)	0.001^a	112 (93.3%)	117 (97.5%)	0.123	HPA2a/2a	67 (55.8%)	91 (75.8%)	0.005^a
HPA2-b	61 (25.4%)	33 (13.75%)		53 (44.2%)	30 (25%)	0.002^a	HPA2a/2b	45 (37.5%)	26 (21.7%)
							HPA2b/2b	8 (6.7%)	3 (2.5%)
HPA5-a	204 (85%)	210 (87.5%)	0.426	115 (95.8%)	116 (96.7%)	0.734	HPA5a/5a	88 (73.3%)	92 (76.7%)	0.828
HPA5-b	36 (15%)	30 (12.5%)		32 (26.7%)	28 (23.3%)	0.551	HPA5a/5b	27 (22.5%)	24 (20%)
							HPA5b/5b	5 (4.2%)	4 (3.3%)

^aStatistically significant.

No statistically significant difference was found in HPA1, 2 and 5 allele or genotype frequencies when different patient groups were compared; males/females; children/adults; bleeders/non-bleeders; ITP patients in activity/in remission; newly diagnosed/persistent/chronic ITP.

Comparing ITP patients with different clinical or laboratory responses (complete, incomplete and no response), as regards HPA1, HPA2 and HPA5 allele and genotype frequencies, showed no significant difference.

Comparison between the ‘a’ allele patients with those with the ‘b’ allele in the studied HPA systems, as regards the age, platelet count, hemoglobin level and total leukocyte count (at diagnosis as well as at sampling time), none of these comparative studies was statistically significant. When these comparative studies were applied between patients with different genotypes (a/a), (a/b), (b/b), this was not statistically significant either.

Calculating the Odds ratio (OR) with the 95% CI for the risk of developing ITP in correlation to different studied HPA alleles revealed that individuals with HPA2b are 2.37 times more likely to develop ITP compared to those with other alleles. Individuals with HPA1a and HPA5b allele phenotypes are more likely to present with the disease by 1.21 and 1.20 times respectively; however, this was not statistically significant in Table 3.

Table 3. OR of HPA1, 2 and 5 genotypes in patients and controls and in relation to the presence of chronic disease in patients group.

Allele phenotype	Patients		Controls		OR	95% CI OR
	Allele present	Allele absent	Allele present	Allele absent		Lower	Upper
HPA1a	115	5	114	6	1.21	0.36	4.08
HPA1b	41	79	45	75	0.87	0.51	1.47
HPA2a	112	8	117	3	0.36	0.09	1.39
HPA2b	53	67	30	90	2.37^a	1.37	4.11
HPA5a	115	5	116	4	0.79	0.21	3.03
HPA5b	32	88	28	92	1.20	0.67	2.15
	Chronic		Non-Chronic			95% CI OR
Allele phenotype	Allele present	Allele absent	Allele present	Allele absent	OR	Lower	Upper
HPA1a	46	4	4	0	1.14	0.05	24.93
HPA1b	19	31	0	4	5.57	0.28	109.24
HPA2a	47	3	4	0	1.51	0.07	34.05
HPA2b	17	33	1	3	1.22	0.17	8.99
HPA5a	48	2	4	0	2.16	0.09	52.16
HPA5b	12	38	2	2	0.33	0.05	2.20

^aStatistically significant.

Estimation the OR of developing chronic type of ITP in correlation to different studied HPA alleles revealed that individuals who have HPA1b and HPA 5a are 5.57 and 2.16 times more likely to present with chronic disease, compared to individuals who do not have these alleles. Individuals with HPA1a, 2a and 2b allele phenotypes are more likely to present with chronic disease by 1.15, 1.5 and 1.2 times respectively. However, none of these results are statistically significant. Our results show no increased risk of acquiring chronic disease for individuals having HPA 5b allele phenotype, as shown in Table 3.

The Hardy Weinberg output for observed versus expected HPA-1, HPA-2 and HPA-5 genotype frequencies in the healthy participants and ITP patients revealed that the genotypes in both investigated groups showed a good fit with Hardy Weinberg equilibrium, as shown in Table 4.

Table 4. Hardy Weinberg comparison of HPA1, 2 and 5 observed frequencies versus expected frequencies.

Group	Gene	Observed genotype frequency			Total	Expected genotype frequency			P-value
		a/a	a/b	b/b		a/a	a/b	b/b
Patients	HPA1	79	36	5	120	78.4	37.2	4.4	0.73
	HPA2	67	45	8	120	66.8	45.5	7.8	0.91
	HPA5	88	27	5	120	85.9	31.3	2.9	0.13
Controls	HPA1	75	39	6	120	74.4	40.2	5.4	0.75
	HPA2	90	26	3	119	89.2	27.7	2.2	0.50
	HPA5	92	24	4	120	90.1	27.7	2.1	0.14

Discussion

Identifying allele and genotype frequencies of HPA is significant not only for population-related genetics but also for clinical transfusion practice. Establishment of donors HPA genotyping registry assists in finding matched HPA types for patients with rare alleles, guides the management of platelet alloimmunization with clinical varieties: (neonatal alloimmune thrombocytopenia, post-transfusion purpura, post-transfusion refractoriness, post-transfusion passive alloimmune thrombocytopenia and transplantation-associated alloimmune thrombocytopenia) and provides a base for designing clinical research related to platelet disorders.

To the best of our knowledge, this is the first study to assess HPA in the Egyptian population in context with ITP. Different HPA systems polymorphisms have been assessed in the Egyptian population by Salem et al. [11], Husebekk et al. [12] and Salem et al. [13]; however, none of these works has studied these polymorphism in association with ITP.

The current work showed that HPA2b allele phenotype frequency was significantly higher in ITP patients compared to controls, also the homozygous and the heterozygous HPA2b genotype frequencies were significantly higher in ITP patients compared to controls.

Accordingly, Pavkovic et al. [1] reported that the HPA-2b allele showed a significant higher frequency in Macedonian ITP patients compared to normal subjects and concluded that it may be involved in the formation of a specific autoepitope.

The present work showed no statistically significant difference between controls and patients groups as regards HPA5 allele and genotype frequencies. In contrast, Kim and Song reported a significant difference in the allele frequencies of the HPA 5 systems between Korean ITP patients and controls [14].

The current study revealed no statistically significant association between any of the studied alleles with the increased risk of developing chronic ITP. Accordingly, Castro et al. [15] found no association between HPA-5 alleles with the increased risk of acute ITP in Brazilian population. Thude et al. confirmed our findings when reported that HPA 1 and 5 allele frequencies were identical in chronic ITP patients and in controls, as for HPA-2a allele, they reported an association with chronic ITP [16]. It should be emphasized here that Thude et al.’s study was performed on Germans, and on selected chronic refractory patients.

In the present work, HPA1 allele and genotype frequencies appear to be in the same range reported by the Egyptian studies done by Salem et al. [11] with no statistically significant differences.

Although Husebekk et al. [12] study was confined to Egyptian pregnant females, which is not entirely representative to the Egyptian population, their reported frequencies for HPA 1, 2 and 5 systems showed no significant difference compared to our results, confirming our findings that HPA 1, 2 and 5 polymorphism frequency was not gender related.

In 2014, Salem et al. [13] reported a significantly higher HPA- 2b and 5b allele frequencies compared to our results. This might be referred to the study sample, as their participated subjects were Egyptians from Ismailia (northern Egypt), while the current study enrolled Egyptian from Cairo and down to the southern of Egypt (Giza, Beni Suef, Al Fayoum, Al Menia, Asuit, Suhag, Luxor and Aswan). It is to be noted here that Salem et al. [13] reported that gene frequencies of HPA-2 and -5 in their study were statistically different from the previously reported results by Husebekk et al. [12] and were the highest reported among Middle Eastern and Arabic populations.

Our reported HPA 1, 2 and 5 alleles and genotypes frequencies were compared to those of other ethnic groups, as detailed in Table 5.

Table 5. HPA allele/genotype frequencies in our population compared to other populations.

Population	No.	HPA 1 allele %		PV	HPA 1 genotype %			PV	No.	HPA 2 allele %		PV	HPA 2 genotype %			PV	No.	HPA 5 allele %		PV	HPA 5 genotype %			PV
		1a	1b		1a/1a	1a/1b	1b/1b			2a	2b		2a/2a	2a/2b	2b/2b			5a	5b		5a/5a	5a/5b	5b/5b
Egyptian (current study)	120	0.79	0.21	–	62.5	32.5	5.0	–	120	0.86	0.14	–	75.8	21.6	2.5	–	120	0.88	0.12	–	76.7	20.0	3.3	–
Egyptian [11]	206	0.77	0.23	0.59	59.2	35.0	5.83	0.72	206	–	–	–	–	–	–	–	206	–	–	–	–	–	–	–
Egyptian [12]	6974	0.79	0.20	0.93	62	34.2	3.8	0.76	376	0.87	0.13	0.78	75.5	22.8	1.7	0.96	376	0.84	0.16	0.19	70.6	26.9	2.6	0.31
Egyptian [13]	206	0.77	0.23	0.59	59.2	35.0	5.83	0.72	218	0.76	0.24	0.00	56.0	39.9	4.13	0.00	230	0.73	0.27	0.00	56.5	32.6	10.9	0.00
ARABS
Bahraini [17]	194	0.76	0.24	0.46	56.8	38.5	4.7	0.58	194	0.77	0.23	0.00	60.1	33.1	6.8	0.01	194	0.86	0.13	0.66	75.7	20.9	2.7	0.93
Moroccan [18]	107	0.74	0.25	0.28	57.9	33.7	8.41	0.40	107	0.82	0.18	0.15	66.36	30.80	2.80	0.31	112	0.86	0.14	0.57	73.22	25.89	0.89	0.53
Algerian [19]	485	0.83	0.17	0.09	–	–	–	–	485	0.83	0.17	027	–	–	–	–	485	0.84	0.16	0.19	–	–	–	–
Tunisian [20]	90	0.75	0.25	0.30	–	–	–	–	–	–	–	–	–	–	–	–	90	0.78	0.22	0.01	–	–	–	–
Saudi Arabian [21]	100	0.80	0.20	0.75	75	23	2	0.11	100	0.71	0.29	0.00	62	32	6	0.08	100	0.80	0.20	0.03	76.50	23.50	0	0.46
EUROPE
Southern (Mediterranean) Europe
Italian [22]	144	0.85	0.15	0.15	–	–	–	–	144	0.89	0.11	0.07	–	–	–	–	144	0.9	0.10	0.45	–	–	–	–
Macedonian [23]	122	0.87	0.13	0.01	75.40	22.10	2.50	0.12	132	0.85	0.15	0.66	72.00	26.50	1.50	0.78	126	0.91	0.09	0.29	82.5	16.70	0.80	0.51
Spanish [24]	727	0.81	0.19	0.41	–	–	–	–	462	0.9	0.1	0.09	–	–	–	–	454	0.88	0.11	0.87	–	–	–	–
Eastern Europe
Slovenian [25]	152	0.83	0.17	0.22	–	–	–	–	152	0.90	0.10	0.25	–	–	–	–	152	0.89	0.11	0.64	–	–	–	–
Polish [26]	135	0.82	0.18	0.32	75.55	23.70	0.75	0.09	211	0.90	0.10	0.14	81.04	17.54	1.42	0.57	166	0.94	0.06	0.00	87.35	12.65	0%	0.03
Northern Europe
Norwegian [27]	105	0.86	0.13	0.02	81.0	20.0	4.0	0.06	105	0.94	0.06	0.00	94.0	10.0	1.0	0.03	105	0.93	0.07	0.09	91.0	13.0	1.0	0.13
Swiss [28]	500	0.80	0.19	0.47	70.40	26.0	3.60	0.24	500	0.92	0.11	0.00	81.20	17.40	1.40	0.34	500	0.93	0.07	0.00	79.6	19.2	1.2	0.24
Western Europe
British [29]	134	0.84	0.16	0.16	–	–	–	–	134	0.93	0.08	0.85	–	–	–	–	134	0.91	0.09	0.20	–	–	–	–
German [30]	1583	0.84	0.16	0.04	–	–	–	–	1576	0.93	0.07	0.00	–	–	–	–	1643	0.92	0.08	0.01	–	–	–	–
AFRICA
Cameroonian [31]	118	0.91	0.09	0.00	82.2	16.9	0.80	0.01	118	0.76	0.24	0.00	60.20	32.20	7.60	0.02	118	0.75	0.25	0.00	56.8	35.6	7.63	0.01
Central African [31]	110	1.0	0	0.00	100	0	0	0.00	110	0.61	0.39	0.00	35.10	53.20	11.7	0.00	110	0.60	0.41	0.00	37.0	45.0	18.0	0.00
AMERICA
Argentinean [32]	192	0.88	0.12	0.00	76.6	22.4	1	0.00	192	0.86	0.13	0.65	77.10	20.8	2.1	0.95	192	0.93	0.07	0.05	86.5	12.5	1	0.06
Brazilian [33]	100	0.93	0.08	0.00	86.0	13.0	1.0	0.0	100	0.85	0.15	0.71	73	24	3	0.90	100	0.92	0.08	0.12	85.0	14.0	1.0	0.23
ASIA
Pakistan [34]	593	0.89	0.12	0.00	79.4	19.0	2.0	0.01	593	0.92	0.08	0.01	85	13.5	1.2	0.00	593	0.90	0.10	0.24	82.0	16.0	2.5	0.00
Han Chinese [35]	1000	0.99	0.01	0.00	98.9	1.2	0	0.00	1000	0.95	0.05	0.00	90.4	9.5	0.1	0.00	1000	0.99	0.01	0.00	97.3	2.6	0.1	0.00

Note: Bold-italic represents statistically significant values.

We support the previous reports concerning the high prevalence of the HPA-1b, compared to other ethnic groups, which may predispose to higher risk of allo-immunization, as was noted previously [11].

Comparable HPA2 allele and genotype frequencies were reported by previous Egyptian study [12], as well as studies done in some Middle Eastern Arab countries such as Tunisia, Algeria and Morocca (Moroccan Berber population).

HPA polymorphism was reported as predictor of response to certain drugs used in ITP treatment. In Chinese ITP patients, those with the HPA-2 a/b and HPA-3 a/a genotypes are less likely to respond to Kami-kihi-to therapy [36]. In response to Cepharanthin treatment, a statistically higher incidence of HPA-2a/2a in the responders group of ITP and a significantly higher HPA-2a/2b and HPA-3a/3b in non-responders were reported [37].

According to our results, none of the studied HPA1, HPA2 and HPA5 systems’ alleles was associated with the clinical course/severity of ITP patients (platelet count/significant bleeding), persistency, chronicity, lack of/incomplete response to either clinical or laboratory treatment.

In conclusion, the current study shows that individuals with HPA 2b are 2.37 times more likely to develop ITP compared to normal subjects. HPA allelic profile of our population is closely linked to that of other ethnic groups within the Middle East and the neighboring Arab countries.

This study provides comprehensive information on HPA allele distribution among the Egyptian population. These results may guide for future clinical research associated with platelet disorders and may be a first step to build up a data bank on platelet antigen polymorphisms among Egyptians.

Disclosure statement

The authors declare no conflict of interest.