Clinical and genetic characteristics of hypophosphatasia in Chinese children

Hypophosphatasia (HPP, OMIM: 146300, 241500, 241510), which was initially reported by Rathbun [1], is a rare inherited metabolic disorder. It is caused by the loss-of-function mutations in the ALPL gene (MIM 171760) that encodes the tissue-nonspecific isozyme of alkaline phosphatase (TNSALP), which is a homodimeric phosphohydrolase and abundantly expressed in the skeleton, developing teeth, liver, and kidney [2]. There are three well-known extracellular substrates of TNSALP, including inorganic pyrophosphate (PPi), pyridoxal-5-phosphate (PLP), and phosphoethanolamine (PEA) [3]. Among them, PPi is a potent inhibitor of mineralization [3]. Thus, the reduced TNSALP activity that due to the mutations in the ALPL gene results in the extracellular accumulation of PPi, which further leads to defective mineralization of bones and teeth.

Although HPP is characterized by impaired mineralization of bones and teeth, and reduced serum alkaline phosphatase (ALP) activity, the clinical spectrum of HPP is extremely variable. Based on the age of onset and the clinical symptoms, HPP is currently classified into six phenotypes: perinatal lethal, perinatal benign, infantile, childhood, adult, and odontohypophosphatasia (odonto-HPP) [2]. Both perinatal lethal and infantile were identified as the severe forms of HPP, whereas perinatal benign, childhood, adult, and odonto-HPP were defined as the mild forms of HPP [4‐8]. It has been reported that the severe forms of HPP are usually recessively inherited, while the mild forms of HPP show both dominant and recessive inheritance [6]. At present, the diagnosis of HPP remains as before, relied on clinical indications, while the genetic sequencing of the ALPL gene is a useful technique for the precise diagnosis of HPP.

The ALPL gene is located on chromosome 1p36.1-p34 [9] and comprises 12 exons distributed over 70 kb [10]. As of May 2020, a total of 410 different ALPL mutations have been reported worldwide (http://​www.​sesep.​uvsq.​fr/​03_​hypo_​mutations.​php). Among these reported ALPL mutations, missense mutations are the most prevalent, which account for 71.2%. The remaining mutations are small deletions mutations (11.0%), splicing mutations (4.9%), nonsense mutations (4.6%), small insertions mutations (3.4%), large deletions/duplications mutations (2.9%), insertion/deletion mutations (1.5%) and regulatory mutations (0.2%). Additionally, the great variety of ALPL mutations produces numerous combinations of compound heterozygous mutations, which further promotes the diversity of clinical manifestations of HPP. Previously, Michigami et al. analyzed 98 unrelated Japanese HPP patients found that p.Leu520ArgfsX86 (c.1559delT) and p.Phe327Leu (c.979T>C) were the two most common mutations and were usually associated with the perinatal severe and perinatal benign forms of HPP, respectively [11]. The mutation p.Glu191Lys (c.571G>A) was the most frequent mutations in Caucasian HPP patients and was commonly associated with mild forms of HPP [12]. From this aspect, finding more about the correlation between genotypes and phenotypes is crucial for genetic counseling and prognostication. However, to date, studies on the relationship between genotypes and phenotypes in Chinese HPP patients are limited, especially in children.

Therefore, in the present study, to clarify the clinical characteristic and the relationships between genotypes and clinical manifestations of HPP in Chinese children, ten unrelated children diagnosed with different forms of HPP in our hospital were analyzed. Besides, we also reviewed the clinical and mutational features of the previously reported HPP cases of children in China.

HPP is a rare inherited metabolic disorder. Currently, the prevalence of severe HPP has been estimated to be between 1/100,000 and 1/300,000, whereas mild forms of HPP are more common than severe forms (estimated at 1/6370 in Europe) [7]. China has a large territory and the largest population in the world. So far, only 23 children with HPP were reported in China, and 15 cases were reported within the past 3 years. This suggested that the actual number of pediatric HPP cases in China may be underestimated. Here we described the clinical and genetic features of the 23 reported HPP cases as well as 10 HPP cases diagnosed in our hospital to improve the awareness of pediatric HPP in China.

We found that the most common form of HPP in Chinese children was odonto-HPP, followed by infantile and childhood HPP. Only two patients with perinatal lethal HPP, and no patient was documented as perinatal benign HPP, which indicated the low rate of perinatal HPP in China. In contrast to our results, Michigami et al. analyzed 98 unrelated Japanese HPP patients found that perinatal lethal was the most frequent form of HPP in Japan, followed by the perinatal benign form [11]. Diagnostic delay is common among patients with HPP [14]. Vogt et al. conducted a retrospective review of 50 pediatric HPP patients in Germany found the obvious diagnostic delay in infantile (12 months) and childhood HPP (22.5 months) [15]. Similar findings were also identified in our present study. The average age at onset was 0.69 years, while the average age at clinical diagnosis was 3.87 years, indicating a significant delay in diagnosis in China. Additionally, the diagnostic delay was noted in infantile HPP and even obvious in childhood HPP. There may be several reasons for the diagnostic delay, including low awareness, heterogeneity of clinical manifestations, and lack of routine testing of HPP.

As previously described, the clinical spectrum of HPP was highly variable, the main clinical manifestations of the pediatric Chinese patients were in line with those documented in other studies [2, 3, 14]. Failure to thrive was one of the most common manifestations of pediatric HPP patients [16], in the current study, we found that 18 (54.5%) patients showed short stature, and 15 (45.4%) patients showed low body weight, both of which were more common in patients with infantile HPP. Therefore, serum ALP levels should be part of a routine accessory examination in infants and children with problems in gaining weight, growth retardation, and short stature. Similar to other studies [17], we found a higher proportion of males (n = 24, 72.73% approximately) than females (n = 9, 27.27% approximately) diagnosed with pediatric HPP. Indeed, previous studies performed by Grimberga et al. pointed out that compared with boys, girls gained less evaluation for short stature [18]. Thus, the greater parental concern about short stature in sons versus daughters may be one of the explanations for the higher proportion of males in pediatric HPP patients. It is worth noting that contrary to the pediatric HPP, a higher proportion of females than males was observed in adult HPP [14, 19‐21]. Considering the small number of pediatric HPP patients diagnosed in our country, the sex differences observed in the current study still need to be verified in further studies.

Serum ALP levels were decreased in all HPP patients. From the literature, it is known that serum ALP levels seem to be correlated with the disease severity [15, 22, 23]. Consistent with this, our present study conducted in the Chinese pediatric HPP patients also demonstrated significantly decreased serum ALP levels in patients with perinatal lethal/infantile forms of HPP than those with the childhood/odonto forms of HPP. Interestingly, we observed that serum Ca and i-PTH levels were also correlated with disease severity. Serum Ca levels were elevated, while serum i-PTH levels were decreased in patients with perinatal lethal/infantile forms of HPP in comparison with those with childhood/odonto forms of HPP. In support of our results, previous studies have shown that severe hypercalcemia was common, and circulating PTH levels were physiologically suppressed in infantile HPP [2, 24]. However, in childhood HPP, hypercalcemia was less common and serum PTH levels were usually within the normal range [2]. Our findings together with others [25] supported that serum Ca and PTH could provide important complementary diagnostic information, especially for patients with severe forms of HPP.

Genetic detection is a crucial final step in the diagnosis of HPP. Consistent with international research findings, missense mutations were the most common mutations identified in this study, and exon 5 was the predominantly affected exon. The variety of missense mutations resulted in highly variable clinical expressivity and a large number of compound heterozygous genotypes. Also consistent with other results [15], compound heterozygous mutations were far more common than heterozygous and homozygous mutations. Five novel mutations, including p.Arg138GlyfsTer27, p.Leu511Profs*272, p.Ala176Val, p.Phe268Leu, and c.297+5G>A were found in this study. Although we did not conduct functional in vitro studies of the effects of these mutations, evidence from bioinformatics analysis supported the hypothesis that these mutations were harmful. Three of the identified mutations (p.Arg138GlyfsTer27, p.Leu511 Profs*272, c.297+5G>A) could disrupt the protein structure, result in forming a truncated protein or causing a frame-shift, or loss of protein function by changing splice sites, respectively. For the other two missense mutations we identified (p.Ala176Val, p.Phe268Leu): p.Ala176Val was located in the active site valley, and p.Phe268Leu was located in the calcium-binding site. There are five crucial regions identified in TNSALP, including the active site, the active site valley, the homodimer interface, the crown domain, and the calcium-binding site, and mutations that alter residues at these sites may induce dysfunction of the protein and thus cause the HPP [6, 26, 27]. It is worth noting that our study found obvious differences in the spectrum of ALPL mutations between Chinese individuals and other countries. For example, p.Leu520ArgfsX86 and p.Asp378Val were the most common mutations in Japan [11] and the USA [17], respectively, while no report has been found in Chinese HPP patients. p.Glu191Lys was known to occur with a high frequency (up to 55%) in HPP patients with European ancestry [28]. However, it has been reported only twice in the Chinese population [5]. These results indicated that a Chinese-specific screening panel may be warranted for the diagnose of Chinese HPP patients. To date, the relationship between genotypes and phenotypes in Chinese pediatric HPP patients with ALPL mutations remains unclear. Based on the Tissue Nonspecific Alkaline Phosphatase Gene Mutations Database, we found that nonsense mutations were associated with a severe phenotype in comparison with missense mutations. However, no nonsense mutation was found in pediatric HPP patients in China. Previous studies have shown that patients with a single heterozygous mutation usually presented with mild forms of HPP [14]. The same was seen in our present study, subjects (patients 16, 17, 22, 24) carrying single missense mutation or frameshift mutation showed milder phenotypes of HPP. The number of HPP patients in this study was relatively small considering the large population of China. As the number of HPP patients increases, the genotype–phenotype correlation may be clear in the future.

Our study also showed high mortality in patients with severe forms of HPP. Except for three cases lost to follow-up, all patients with severe forms of HPP (perinatal lethal/infantile HPP) have died. Previous studies have shown that approximately 50% of cases with infantile HPP were predicted to die [2], and if patients manifested with chest deformity, respiratory difficulties, or vitamin B6-dependent seizures before 6 months of age, the mortality was significantly higher [29]. Until recently, the management of HPP in China has been symptomatic and supportive only. In 2015, bone-targeted enzyme-replacement therapy (asfotase alfa) was approved in Japan, and then in Canada, in the European Union, and in the United States to treat pediatric-onset HPP [2]. This enzyme has produced beneficial effects not only in bones but also in other organs, including the lungs and muscle [30]. While asfotase alfa seems very promising, it is not approved in China yet, and many questions regarding its long-term effects and the potential secondary adverse effects remain to be solved in the future.

There are still some limitations in the present study. First, due to its retrospective design, not all clinical or biochemical data were available from all patients. Further follow-up study will be conducted in order to explore the long-term prognosis of patients with mild forms of HPP. Second, although serum ALP, Ca and i-PTH levels were associated with the disease severity, the best cutoffs were unclear. Additionally, PLP and PEA are useful parameters for the diagnostic of HPP and serum PLP levels have been reported to be correlated with the disease severity [12]. However, due to the limited medical equipment in our hospital, PLP and PEA levels were not detected in the present study. Third, in vitro functional experiments will be needed to characterize the function of the novel mutants. Fourth, the number of patients was not big enough due to low awareness. Future research is required to reveal the phenotype-genotype correlations.