Clinical and genetic features of hereditary angioedema with and without C1-inhibitor (C1-INH) deficiency in Japan.

CONFLICT OF INTEREST

As for COI, Dr. Hashimura has nothing to disclose. Dr. Kiyohara has nothing to disclose. Dr. Fukushi has nothing to disclose. Dr. Hirose is a consultant for Takeda Pharmaceutical Company (without salary). Dr. Ohsawa reports personal fees from Takeda Pharmaceutical Company, personal fees from CSL Behring, personal fees from Torii Pharmaceutical Company, personal fees from BioCryst Pharmaceuticals, outside the submitted work. Dr. Tahira has nothing to disclose. Dr. Horiuchi reports personal fees from Takeda Pharmaceutical Company, personal fees from CSL Behring, personal fees from Torii Pharmaceutical Company, outside the submitted work.

AUTHOR CONTRIBUTIONS

CH contributed to conception, performed data acquisition and analysis, and wrote the initial draft of the manuscript. CK and TT performed data analysis and drafted the manuscript. THirose, JIF, and IO were involved in data collection. THoriuchi designed the study, performed data acquisition and analysis, and wrote the manuscript. All authors contributed to drafting the article, revised the manuscript critically for important intellectual content, and approved the final version.

To the Editor,

Hereditary angioedema with normal C1‐inhibitor (HAEnCI) is an umbrella term for several types of HAE that phenotypically resemble HAE but in which variants affecting function have been identified in genes other than the C1‐INH gene (SERPING1). In contrast to European patients with HAE, we suppose comprehensive clinical and genetic data of HAE are scarcely reported from Asian countries, including Japan. In particular, HAEnCI in Asia has not been clearly characterized. Considering ethnic differences, it would be important to clarify the features of HAE in the Asian population. Here, we report the clinical and genetic features of Japanese patients with 158 cases from 122 families with HAE‐C1‐INH and 21 cases from 21 families with HAEnCI. HAEnCI was defined as follows: normal C1‐INH activity, no variants affecting function in the SERPING1 gene, at least one relative with recurrent angioedema attacks (i.e., positive family history), no history of urticaria, and lack of efficacy by anti‐histamines or corticosteroids.

Table 1 shows clinical features of our Japanese patients with HAE‐C1‐INH and HAEnCI. When compared with HAE‐C1‐INH, HAEnCI was significantly more predominant among females (95.2% vs. 66.9%, p = .008), more frequently affected in the face (61.9% vs. 32.9%, p = .009) and pharynx/larynx (47.6% vs. 21.5%, p = .009), associated with more frequent exacerbations in the previous year (mean ± SD: 14.4 ± 27.3 vs. 3.66 ± 7.14, p = .0001), associated with higher prevalence of patients with more than six angioedema attacks in the previous year (52.4% vs. 16.5%, p = .0001), and more susceptible to triggers such as physical stress (38.1% vs. 12.7%, p = .007) and upper respiratory infections (URI) (19.0% vs. 5.1%, p = .037). The prevalence of having a positive family history experiencing angioedema attacks was significantly higher in HAEnCI (100%) than in HAE‐C1‐INH (81.6%) (p = .027). This is because the inclusion criteria of HAEnCI demand a family history of angioedema. There were no differences in most of the clinical features of male and female patients with HAE‐C1‐INH except the incidence of urological attacks and prodromal symptoms (Table S1).

TABLE 1

Clinical features of the Japanese patients with HAE‐C1‐INH and HAEnCI

Variables	HAE‐C1‐INH % (no. affected/observed a )	HAEnCI % (no. affected/observed a )	p
Female	66.9 (105/157)	95.2 (20/21)	.008
Comorbid disease
AID	3.2 (5/158)	0.0 (0/21)	1.000 b
Urticaria	7.6 (12/158)	0.0 (0/21)	.365 b
Bronchial asthma	5.1 (8/157)	9.5 (2/21)	.334 b
CVD	3.2 (5/158)	9.5 (2/21)	.192 b
Arthralgia/arthritis	3.2 (5/158)	0.0 (0/21)	1.000 b
Any kinds	27.2 (43/158)	42.9 (9/21)	.138
Family history
At least another patient in family member	81.6 (129/158)	100.0 (21/21)	.027 b
Death probably by angioedema	7.6 (12/158)	4.8 (1/21)	1.000 b
Age of onset
Mean ± SD	23.0 ± 12.6	26.8 ± 22.7	.298
≤40	95.2 (120/126)	82.4 (14/17)	.075 b
≤20	51.6 (65/126)	52.9 (9/17)	.916
Age of diagnosis
≤40	69.9 (100/143)	70.0 (14/20)	.995
≤20	13.3 (19/143)	40.0 (8/20)	.007 b
Site of attacks
Extremities	39.2 (62/158)	47.6 (10/21)	.462
Tongue	0.6 (1/158)	4.8 (1/21)	.221 b
Face	32.9 (52/158)	61.9 (13/21)	.009
Pharynx/Larynx	21.5 (34/158)	47.6 (10/21)	.009
Intestine	35.4 (56/158)	38.1 (8/21)	.812
Urogenital	6.3 (10/158)	9.5 (2/21)	.636 b
Others	7.0 (11/158)	14.3 (3/21)	.216 b
Frequency of attacks in the previous 1 year
Mean ± SD	3.66 ± 7.14	14.4 ± 27.3	.0001
≧6 times	16.5 (26/158)	52.4 (11/21)	.0001
Prodromal symptom	13.3 (21/158)	28.6 (6/21)	.097 b
Trigger
Psychological	13.9 (22/158)	33.3 (7/21)	.051 b
Physical	12.7 (20/158)	38.1 (8/21)	.007 b
Menstruation/Pregnancy	19.0 (20/105)	20.0 (4/20)	1.000 b
Dental procedure	13.9 (22/158)	14.3 (3/21)	1.000 b
URI	5.1 (8/158)	19.0 (4/21)	.037 b
Medication and acute attack c
pdC1‐INH	31.0 (49/158)	19.0 (4/21)	.259
Tranexamic acid	19.0 (30/158)	38.1 (8/21)	.083 b
Anti‐histamine	6.3 (10/158)	19.0 (4/21)	.064 b
Prednisolone	3.2 (5/158)	14.3 (3/21)	.053 b
Epinephrine	1.3 (2/158)	4.8 (1/21)	.314 b
Danazol	2.5 (4/158)	0.0 (0/21)	1.000 b
FFP	0.0 (0/158)	0.0 (0/21)	–
Intubation/Tracheostomy	3.8 (6/158)	9.5 (2/21)	.238 b
Prophylaxis
Tranexamic acid	31.0 (49/158)	33.3 (7/21)	.829
Danazol	5.1 (8/158)	4.8 (1/21)	1.000 b
Others	6.3 (10/158)	9.5 (2/21)	.636 b

A total number of registered are 158 for HAE‐C1‐INH and 21 for HAEnCI.

Abbreviations: AID, autoimmune disease; CVD, cardiovascular disease; FFP, fresh frozen plasma; SD, standard deviation; URI, upper respiratory infection.

Some of the data are lacking.

Fisher's exact test.

Icatibant was approved in Japan after the termination of this registration.

% (no. affected/observed )

% (no. affected/observed )

It is of note that our patients with HAEnCI suffer from edema in the face and pharynx/larynx twice more frequently than those with HAE‐C1‐INH. In European patients with HAEnCI, facial and oropharyngeal swellings develop more predominantly than those with HAE‐C1‐INH as well. On the other hand, there are a number of differences in the clinical characteristics such as frequency of attacks and triggers between Japanese and European patients with HAEnCI. In particular, hormonal perturbations induced by menstruation/pregnancy or oral contraceptives did not seem to influence attacks in our patients with HAEnCI. This finding is in contrast with European patients with HAEnCI whose angioedema symptoms are frequently deteriorated by oral contraceptives or pregnancies.

In the case of HAE‐C1‐INH, the incidence of abdominal attacks was different between our patients and European patients. Abdominal attacks are common clinical manifestations of European patients with HAE‐C1‐INH, whose incidence reaches over 90% in these patients. In contrast, the proportion of patients who had experienced intestinal attack was 35.4% in our 158 patients with HAE‐C1‐INH (Table 1). Other reports from Asia such as the one from Mainland China  have shown similar incidence of abdominal attacks with the present study.

Genetic analysis directed toward the entire exons for the known causative genes was performed for our Japanese HAEnCI patients (n = 21). Targeted next‐generation sequencing for the factor XII gene (F12), the plasminogen (PLG) gene, the angiopoietin‐1 (ANGPT1) gene, the kininogen 1 (KNG1) gene, and the SERPING1 gene revealed missense variant: p.Lys330Glu (c.988A>G) in the PLG gene in two patients and p.Arg466Ser (c.1396C>A) in the F12 gene in 1 patient. All of these variants were heterozygous. The p.Lys330Glu variant in the PLG gene was originally reported in Europe as causative for HAEnCI. The p.Arg466Ser variant identified in the F12 gene has been reported in a patient with factor XII deficiency. It is likely that this variant does not have any relationship with angioedema. None of the known variants affecting function in the F12 gene, the ANGPT1 gene, and the KNG1 gene responsible for HAEnCI was identified in our patients. The patients carrying variants affecting function in the F12 gene (HAE‐F12) account for about 30% of HAEnCI in European countries. Lack of the F12 gene findings in our Japanese HAEnCI patients might reflect ethnic difference, or limitation of our approach focused only on small variants of the exons. In addition, the recently reported variants affecting function of the myoferlin (MYOF) gene (p.Arg217Ser) and of the heparin sulfate 3‐O‐sulfotransferase 6 (HS3ST6) gene (p.Thr144Ser) were not identified in our patients.

Ninety‐two percent (112/122) of the studied families with HAE‐C1‐INH were positive for the SERPING1 variant affecting function (Table 2). Missense variants were the most common (45 families), followed by small insertions/deletions (30 families), large insertions/deletions (15 families), nonsense variants (13 families), and splicing defects (nine families). Nineteen variants were recurrent, at least in two different families. Twenty‐five variants affecting function had not been reported in LOVD v.3.0 Leiden Open Variation Database (https://databases.lovd.nl/shared/genes/SERPING1) as of April 2021. All the novel sequence variants except two (#17, #23) were predicted to be pathogenic by in silico analysis (Table S2). The variants #17 (c.461_465delACCACinsTCAGGGAGGCTCTTCAA) and #23 (c.554_555insTGTTGCAGGGGC) were in frame insertion/deletion and were predicted as variants of uncertain significance (VUS). There were no correlations between the types of SERPING1 variants (missense vs. other variant) and the markers of severity of disease, such as the age of onset, frequency of attacks, and C1‐INH activity (Table S3).

TABLE 2

SERPING1 mutations in our Japanese 112 families of HAE‐C1‐INH

	Alteration (physical location on chromosome 11) a	cDNA numbering (NM_000062.2)	Location	Effect on protein	Families	Reference b (LOVD)
1	g.57365720A>G	c.‐22‐2A>G	Intron1	Splicing defect	1	*
2	g.57365746G>A	c.3G>A	Exon2	p.Met1Ile	1	*
3	g.57365748_57365749del	c.5_6delCC	Exon2	p.Ala2Valfs*17	1
4	g.57365760_57365767dup	c.17_24dupCCCTGCTG	Exon2	p.Thr9Profs*3	1
5	g.57365795G>T	c.51+1G>T	Intron2	Splicing defect	1	*
6	g.57367351G>A	c.52‐1G>A	Intron2	Splicing defect	1	*
7	g.57367406_57367407del	c.106_107delAG	Exon3	p.Ser36Phefs*21	3	*
8	g.57367416_57367417insGGATC	c.116_117insGGATC	Exon3	p.Asp39Glufs*42	1
9	g.57367438_57367507del	c.138_207del	Exon3	p.Thr47Glnfs*9	1	*
10	g.57367447del	c.147delT	Exon3	p.Ile50Serfs*29	1
11	g.57367504del	c.204delC	Exon3	p.Asn69Thrfs*10	1
12	g.57367526del	c.226delA	Exon3	p.Thr76Profs*3	1
13	g.57367646C>T	c.346C>T	Exon3	p.Gln116*	2	*
14	g.57367700_57367704del	c.400_404delGAGAG	Exon3	p.Glu134Serfs*121	1
15	g.57367703_57367704del	c.403_404delAG	Exon3	p.His136Phefs*120	1	*
16	g.57367749C>T	c.449C>T	Exon3	p.Ser150Phe	2	*
17	g.57367761_57367765delins TCAGGGAGGCTCTTCAA	c.461_465delACCACins TCAGGGAGGCTCTTCAA	Exon3	p.Tyr154_His155delins PheArgGluAlaLeuGln	1
18	g.57367767C>A	c.467C>A	Exon3	p.Ala156Asp	2	*
19	g.57367775del	c.475delG	Exon3	p.Ala159Glnfs*2	1
20	g.57367848T>C	c.548T>C	Exon3	p.Leu183Pro	2	*
21	g.57367850G>A	c.550G>A	Exon3	p.Gly184Arg	2	*
22	g.57369507G>C	c.551‐1G>C	intron3	Splicing defect	1	*
23	g.57369500‐57369511dup	c.554_555ins TGTTGCAGGGGC	Exon4	p.Ala185_Gly186ins ValAlaGlyAla	1
24	g.57369510del	c.553delG	Exon4	p.Ala185Leufs*26	1	*
25	g.57369523C>A	c.566C>A	Exon4	p.Thr189Asn	1	*
26	g.57369586T>C	c.629T>C	Exon4	p.Leu210Pro	1	*
27	g.57369610T>A	c.653T>A	Exon4	p.Val218Asp	1	*
28	g.57369623_57369624del	c.666_667delTC	Exon4	p.Gln223Aspfs*33	1	*
29	g.57369631_57369632delinsAA	c.674_675delinsAA	Exon4	p.Phe225*	2	*
30	g.57373482G>T	c.686‐1G>T	intron4	Splicing defect	1
31	g.57373492T>A	c.695T>A	Exon5	p.Ile232Lys	1	*
32	g.57373549T>A	c.752T>A	Exon5	p.Leu251Gln	1
33	g.57373617A>G	c.820A>G	Exon5	p.Ile274Val	3	*
34	g.57373649dup	c.852dupT	Exon5	p.Thr285Tyrfs*20	1
35	g.57373687G>A	c.889+1G>A	intron5	Splicing defect	1	*
36	g.57373886T>G	c.895T>G	Exon6	p.Trp299Gly	1	*
37	g.57373928T>C	c.937T>C	Exon6	p.Phe313Leu	1	*
38	g.57373956T>G	c.965T>G	Exon6	p.Val322Gly	2
39	g.57373962T>G	c.971T>G	Exon6	p.Met324Arg	1	*
40	g.57373989C>A	c.998C>A	Exon6	p.Ala333Asp	1	*
41	g.57379188A>G	c.1030‐2A>G	intron6	Splicing defect	1
42	g.57379189G>C	c.1030‐1G>C	intron6	Splicing defect	1	*
43	g.57379189G>A	c.1030‐1G>A	intron6	Splicing defect	1	*
44	g.57379193G>A	c.1033G>A	Exon7	p.Gly345Arg	1	*
45	g.57379194G>A	c.1034G>A	Exon7	p.Gly345Glu	1	*
46	g.57379216dup	c.1056dup	Exon7	p.Leu353Serfs*16	1	*
47	g.57379241C>T	c.1081C>T	Exon7	p.Gln361*	1	*
48	g.57379279_57379302del	c.1119_1142del	Exon7	p.Leu374_Ala381del	1
49	g.57379317del	c.1157delT	Exon7	p.Leu386Argfs*11	1	*
50	g.57379317_57379318del	c.1157_1158delTG	Exon7	p.Leu386Argfs*38	2
51	g.57379344_57379345	c.1184_1185delTC	Exon7	p.Leu395Profs*29	3
52	g.57379355C>T	c.1195C>T	Exon7	p.Pro399Ser	1	*
53	g.57379379C>T	c.1219C>T	Exon7	p.Gln407*	2
54	g.57379395T>A	c.1235T>A	Exon7	p.Ile412Asn	1	*
55	g.57381820T>G	c.1269T>G	Exon8	p.Tyr423*	1	*
56	g.57381835T>A	c.1284T>A	Exon8	p.Cys428*	1	*
57	g.57381891T>C	c.1340T>C	Exon8	p.Leu447Pro	1	*
58	g.57381919_57381921del	c.1368_1370delGGC	Exon8	p.Ala457del	1
59	g.57381920G>C	c.1369G>C	Exon8	p.Ala457Pro	1	*
60	g.57381947C>T	c.1396C>T	Exon8	p.Arg466Cys	7	*
61	g.57381947del	c.1396delC	Exon8	p.Arg466Alafs*110	2
62	g.57381948G>T	c.1397G>T	Exon8	p.Arg466Leu	3	*
63	g.57381982C>G	c.1431C>G	Exon8	p.Phe477Leu	1	*
64	p.57381996G>A	c.1445G>A	Exon8	p.Trp482*	1
65	g.57382026T>A	c.1475T>A	Exon8	p.Met492Lys	1	*
66	g.57382028G>A	c.1477G>A	Exon8	p.Gly493Arg	1	*
67	g.57382029G>A	c.1478G>A	Exon8	p.Gly493Glu	1	*
68	g.57382031C>T	c.1480C>T	Exon8	p.Arg494*	3	*
69	g.57382034_57382036del	c.1483_1485delGTA	Exon8	p.Val495del	1
70	g.57382044C>A	c.1493C>A	Exon8	p.Pro498His	1	*
71	g.57382044C>T	c.1493C>T	Exon8	p.Pro498Leu	1	*
72	Deletion of exon 1 to 4				1	*
73	Deletion of exon 1 to 8				2	*
74	Deletion of exon 4				5	*
75	Deletion of exon 4 to 8				1
76	Deletion of exon 5 to 6				1	*
77	Deletion of exon 5 to 8				3	*
78	Duplication of exon 3				1
79	Duplication of exon 4				1	*

GRCh37/hg19 genomic chromosomal coordinates are shown.

The mutations previously reported in LOVD v.3.0 (https://databases.lovd.nl/shared/genes/SERPING1) as of April 2021 are shown (*).

In conclusion, we demonstrated the clinical and genetic features of the patients with HAE‐C1‐INH and HAEnCI in Japan. When compared with European patients with HAE‐C1‐INH, the clinical features of our Japanese patients with HAE‐C1‐INH were almost similar, but the frequency of abdominal pain in our patients was considerably low. The clinical features of our Japanese patients with HAEnCI were significantly different from those of HAE‐C1‐INH in a number of aspects such as female‐male ratio and sites of attacks. Lack of the F12 gene variants affecting function in our HAEnCI patients might indicate that the genetic background is different between Japanese and European patients with HAEnCI.

Table S1, S3

Click here for additional data file.

Table S2

Click here for additional data file.