Prominent Respiratory Involvement Featured in High-Risk Screening for Late-Onset Pompe Disease in China

Backgroud: Pompe disease is a rare metabolic disorder with available enzymatic replacement therapy. Contrasting with the classic infantile form, the others subtypes have a heterogeneous presentation that makes an early and accurate diagnosis diculty. A multicenter observational study was aimed to assess the prevalence of late-onset Pompe disease (LOPD) in high-risk population, using dried blood spot (DBS) as a main screening tool. Methods: 20 Chinese neuromuscular centers were involved in the early LOPD screening study. Inclusion criteria were: (1) age ≥ 1 years, (2) either one of a) persistent hyperCKemia; or b) muscle weakness of axial and/or limb-girdle muscles; or c) unexplained restrictive respiratory insuciency. Enzymatic activity of acid α-glucosidase (GAA) was measured on DBS by tandem mass spectrometry (MS/MS) assay. For the nal diagnosis genotype was assessed by next-generation sequencing. Result: In a 9-month period, we studied 492 cases: 26 positive samples (5.3%) were detected by DBS screening. Molecular studies nally conrmed LOPD diagnosis in 8 cases (1.6%). The GAA activities in individuals bearing pseudodeciency alleles were well separated from those in LOPD patients by MS/MS assay. The median interval from the onset of symptoms to diagnosis was 5 years. Besides axial/proximal muscle weakness, all patients showed respiratory insuciency with a mean forced vital capacity of predicted of 48%. The level of creatine kinase ranged from normal to no more than 5-fold of upper normal limit. LOPD with isolated hyperCKemia was not identied. Conclusions: This study conrms that DBS test is a reliable method for screening for LOPD. Respiratory insuciency is earlier and more severe among Chinese LOPD patients. LOPD presented with paucisymptomatic hyperCKemia. Therefore, a prompt diagnosis is critical to prognosis. maltase deciency, deciency lysosomal with isolated hyperCKemia[14]. Another Spainish study revealed 2.2% prevalence of LOPD in isolated hyperCKemia[26]. Caucasian patients with LOPD have a higher prevalence of c.-32–13 T > G (around 40%). all those presymptomatic patient were heterozygous for the common c.−32−13T>G mutation[14, 26], which is at a very low frequency in Chinese LOPD cohort[16]. This difference of mutational spectrum may explain the lower CK level observed in Chinese patients.

The development of more rapid diagnostic tools, such as the dried blood spot (DBS), were widely used to detect GAA activity in large-scale screening studies in order to facilitate earlier diagnosis. A number of studies have proven the value of DBS in screening for Pompe disease in patients with unclassi ed limb-girdle weakness or asymptomatic hyperCKemia, with prevalence between 1.1%-4.6% [8][9][10][11][12][13][14][15]. In Chinese patients with LOPD, GAA mutational hotspots are different from European populations [16]. Compared with patients in other parts of the world, the Chinese patients with LOPD seems to have a more aggravated phenotype, especially regarding the respiratory involvement and the requirement of mechanical ventilation [7,17]. Unfortunately, high risk screening for LOPD in Chinese population has not yet been reported.
In this study, we conducted a prospective, multicenter, observational study to identify undiagnosed LOPD in a large high-risk population, based on measuring GAA activity by DBS, in order to explore the appropriate screening criteria for Chinese LOPD patients.

Patients And Methods
The participants were prospectively identi ed through medical examination at 20 in-and outpatient neuromuscular clinics in China between 2020 August and 2021 April. Inclusion criteria were patients with age over 1 year, and at least one of the following: (1) unexplained persistent hyperCKemia; (2) axial/limb-girdle muscular weakness, (3) unexplained respiratory insu ciency. Persistent hyperCKemia was de ned as serum creatine kinase (CK) levels above 1.5-fold the upper normal limit (UNL) evidenced at least twice with a minimum interval of 1 month. Axial/limb-girdle muscular weakness was de ned as weakness of the trunk muscles and proximal muscles in the arms and legs, including paravertebral muscles, especially the neck exors, the muscles of the shoulders, upper arms, pelvic and thighs. Respiratory insu ciency was de ned as development of respiratory symptoms (dyspnea at rest, exertional dyspnea, orthopnea, hypersomnolence, headache on awakening), hypercapnia, or abnormal pulmonary tests suggestive of neuromuscular weakness. Exclusion criteria were relatives of patients with diagnosed Pompe disease.

Sampling And Tandem Mass Spectrometry Assay
A blood sample was collected from a peripheral vein and was immediately spotted onto a lter paper. All samples wer e anonymized and analyzed at Suzhou PerkinElmer Medical Laboratory, China. GAA activity was analyzed using tandem mass spectrometry (MS/MS) assay as previously reported [18,19].

Gene Sequencing
Next-generation sequencing was performed if the DBS activity of GAA was below 1.46 µmol/L/h. Genomic DNA was extracted from the peripheral blood leukocytes using a genomic DNA extraction kit (Tiangen, China). Targeted next-generation sequencing (NGS) covering 603 muscular disorder related genes and subsequent Sanger con rmation was performed in 26 patients (Shanghai Amplicon-gene Bioscience Co., Ltd.). Human Gene Mutation Database was used to address the novelty of variants. Predicted severity for each known mutation was based on information from the Pompe mutation database, which is maintained by the Erasmus Medical Center (www.pompecenter.nl). The novel variants were interpreted and classi ed according to the American College of Medical Genetics and Genomics (ACMG) recommendation [20].

Statistical analysis
For descriptive statistical analyses, the statistics program SPSS for Windows version 23.0 (IBM Corp., Armonk, NY) was used.

Standard Protocol Approvals, Registrations, And Patient Consents
Oral and written informed consent was obtained from all participants before blood sampling. The study was conducted in accordance with ethical principles deriving from the Declaration of Helsinki. It was approved by the Ethics Committees of Huashan Hospital and participant centers.

Results
The study included 492 patients aged between 1 and 86 years with a median of 42 years of participation (male: female≈1.2:1). The demographic and clinical presentation of the cohort is summarized in Table 1. The 8 patients with nally con rmed LOPD showed an average GAA activity of 0.33 µmol/L/h. Among the rest with decreased DBS activity of GAA, 11 patients (42.3%) carried at least one pseudode ciency allele. Individual GAA activities, genotypes, and creatine kinase levels are provided in Supplementary Table 1.
The set of pseudode ciency DBS were well separated from the LOPD samples when measured with MS/MS ( Figure 2). Seven patients didn't have any GAA mutation. Since other enzymes (including acid β-glucocerebrosidase for Gaucher, acid α-galactosidase A for Fabry, and acid α-L-iduronidase for MPS-I) tested in the same multiplex assay were generally moderately low in these patients, we presumed that poor sampling or shipment might contribute to the false positivity. A second DBS test was not performed.

Discussion
This is the rst prospective cohort screening for late-onset Pompe disease in a high-risk population in China. In our study, the prevalence of adult Pompe disease is 1.6% in patients with unclassi ed axial or limb-girdle muscle weakness/ hyperCKemia/ respiratory insu ciency. This nding is slightly lower compared with previous published studies in other populations, with a combined prevalence of 2.4% (Table 3) [8 -15]. Although it was reported that there is a higher prevalence of Pompe disease in Taiwan, the patients with LOPD in Taiwan often showed onset of symptoms in their second decade of life with rapid disease progression [17,21]. In this study, since most centers involved are adult neurology departments, the prevalence of younger LOPD patients might largely be overlooked. 2065G>A]) has been identi ed at high frequency in the Japanese and Chinese populations [24]. Since newborn screening study in Taiwan indicated that mass spectrometry distinguished affected and pseudode ciency patients well [21], we chose GAA sequencing as the second test for con rmation of the diagnosis. In this study 11 of the 18 false positive samples carried at least one pseudode ciency allele. In previous Asian studies, pseudode ciency heterozygote/homozygote was reported to complicate Pompe disease screening [10,21,22]. However, mass spectrometry could help distinguish between patients with LOPD and carriers of pseudode ciency.
In our cohort, axial/proximal limb muscle weakness remained as the core feature of LOPD. However, the serum CK levels in the newly diagnosed LOPD patients were generally low. Five out of 8 patients presented with a CK level above 1.5-fold the UNL, and only 2 patients above 2-fold the UNL. None of the patients with "asymptomatic hyperCKemia" proved to have LOPD. In contrast, in the Spanish cohort, CK was elevated, in a variable manner, in 13 of the 16 patients, with values between 2 and 8 times the upper limit of normal (UNL) [25]. In the Italian study, 5/17 (29.4%) were identi ed as presymptomatic LOPD patients with isolated hyperCKemia [14]. Another Spainish study revealed 2.2% prevalence of LOPD in isolated hyperCKemia [26]. Caucasian patients with LOPD have a higher prevalence of c.-32-13 T > G (around 40%). Moreover, all those presymptomatic patient were heterozygous for the common c.−32−13T>G mutation [14,26], which is at a very low frequency in Chinese LOPD cohort [16]. This difference of mutational spectrum may explain the lower CK level observed in Chinese patients.
In LOPD, as respiratory dysfunction progresses patients develop sleep disordered breathing which progresses to nocturnal hypoventilation, and eventually development of diurnal respiratory failure. The addition of ventilation use is a signi cant event in disease progression, negatively affecting quality of life. The respiratory involvement in the 8 newly diagnosed patients is earlier and more severe compared to previous high risk screening studies[8, 12,14]. One patient (Pt.3) initially manifested with exertional dyspnea with preserved limb-girdle muscle strength and biceps muscle biopsy showed no vacuolar pathology. This was in accordance with a recent report that, compared with patients in other parts of the world, the lung function of Chinese patients with LOPD were worse, manifested by lower forced vital capacity and maximum expiratory and inspiratory pressure [7]. Generally, the predictive threshold of FVC-U for nighttime ventilation was 39% of predicted [27]. However, hypercapnia could happen in patients with larger ventilatory reserve, as Pt 2 and 6 in our study. It may involve both the presence of sleep-disordered breathing and blunted ventilatory responsiveness [27,28]. Monitoring of respiration and/or gas exchange during sleep is especially of importance for the onset of sleep disordered breathing can be easily missed.
In conclusion, we recommend that DBS and tandem mass spectrometry test be undertaken early in the diagnosis workup of high-risk patients to exclude LOPD. Extension to patients with sleep disordered breathing rather than isolated hyperCKemia is recommended in Chinese population. Such approach will facilitate early diagnosis and allows to start the treatment at less advanced stages of the disease.
Abbreviations LOPD, Late-onset Pompe Disease; UNL, upper limit of normal; DBS: dried blood spot; GAA acid α-1,4-glucosidase MS/MS tandem mass spectrometry Declarations Figure 1 A ow diagram of the study design and analysis process.