Journal of Neurology (2021) 268:2482–2492 https://doi.org/10.1007/s00415-021-10409-9 ORIGINAL COMMUNICATION STIG study: real‑world data of long‑term outcomes of adults with Pompe disease under enzyme replacement therapy with alglucosidase alfa Kristina Gutschmidt1 · Olimpia Musumeci2 · Jordi Díaz‑Manera3,4,5 · Yin‑Hsiu Chien6 · Karl Christian Knop7 · Stephan Wenninger1 · Federica Montagnese1 · Alessia Pugliese2 · Graziana Tavilla2 · Jorge Alonso‑Pérez3 · Paul Wuh‑Liang Hwu6 · Antonio Toscano2 · Benedikt Schoser1 Received: 10 September 2020 / Revised: 11 December 2020 / Accepted: 13 January 2021 / Published online: 5 February 2021 © The Author(s) 2021 Abstract Background Pompe disease is one of the few neuromuscular diseases with an approved drug therapy, which has been avail- able since 2006. Our study aimed to determine the real-world long-term efficacy and safety of alglucosidase alfa. Methods This multicenter retrospective study (NCT02824068) collected data from adult Pompe disease patients receiving ERT for at least 3 years. Demographics and baseline characteristics, muscle strength, lung function (FVC), walking capability (6MWT), and safety were assessed once a year. Evaluation was done on the group and individual levels, using quantitative linear models (t test) and general univariate linear models (ANOVA). Findings Sixty-eight adult Pompe disease patients from four countries (Spain, Taiwan, Italy, Germany (STIG)) participated. The mean follow-up was 7.03 years ± 2.98. At group level in all outcome measures, an initial improvement followed by a secondary decline was observed. After 10 years, the 6­ MWT%pred showed the most sustained positive effect (p = 0.304). The ­MRC%max remained stable with a mild decline (p = 0.131), however, ­FVC%pred deteriorated significantly (p < 0.001) by 14.93% over 10 years of ERT. The progression rate of F ­ VC%pred under ERT could be explained in most of the patients (83.5%) by the disease severity at baseline. Furthermore, our study shows a decline in the FVC combined with an increase in non-invasive and invasive ventilation requirements in adult Pompe disease patients over time. Conclusions The STIG real-world study confirms an initial efficacy of ERT in the first years with a secondary sustained decline in multiple outcome measures. Further efforts are required to establish a more valid long-term monitoring and improved therapies. Keywords Alglucosidase alpha · Efficacy · Enzyme replacement therapy · Glycogen storage disease type 2 · Long term follow-up · Pompe disease 4 * Benedikt Schoser Centro de Investigación Biomédica en Red en Enfermedades benedikt.schoser@med.uni‑muenchen.de Raras (CIBERER), Valencia, Spain 5 1 John Walton Muscular Dystrophy Research Center, Department of Neurology, Friedrich-Baur-Institute, University of Newcastle, Newcastle, UK Ludwig-Maximilians University Munich, Ziemssenstr. 1, 6 80336 Munich, Germany Department of Medical Genetics and Pediatrics, National 2 Taiwan University Hospital, Taipei, Taiwan Department of Clinical and Experimental Medicine, 7 University of Messina, Messina, Italy Neurologische Praxis Neuer Wall, Hamburg, Germany 3 Neurology Department, Neuromuscular Diseases Unit, Hospital de La Santa Creu I Sant Pau and Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona, Spain 13 Vol:.(1234567890) Journal of Neurology (2021) 268:2482–2492 2483 Background Methods Pompe disease, also known as glycogen storage disorder type Study centers and inclusion criteria 2, is an autosomal recessive metabolic disorder caused by an enzymatic deficiency of acid alpha-glucosidase (GAA) in lys- This investigator-driven study was approved by the Eth- osomes [1], first described by J.C. Pompe, G. Bischoff, and ics committee of the Ludwig-Maximilians-University W. Putschar-independently in the year 1932 [2]. There is a Munich, Germany (IRB vote 231-16) in 2016. continuously rising number of mutations: currently there The study was performed at five neuromuscular centers are more than 500 GAA changes causing a variable degree with longstanding experience in the diagnosis and treat- of GAA enzyme deficiency [3]. A prevalence of 2.4% for ment of Pompe disease: (1) the Ludwig-Maximilians-Uni- Pompe disease was recently identified in adults presenting with versity Munich, Germany (lead center), (2) the Neurology hyperCKemia and/or limb-girdle muscular weakness in a large center “Neuer Wall” in Hamburg, Germany, (3) the Hos- European cohort [4]. Overall, the incidence varies between 1 to pital de la Santa Creu i Sant Pau in Barcelona, Spain, (4) 40.000 and 1 to 200.000 depending on ethnicity and geography the Neurology and Neuromuscular Diseases Unit of the [5, 6], encompassing a spectrum of infantile (IOPD) to a late- University of Messina, Italy and (5) the Medical Meta- onset type of Pompe disease (LOPD). Nowadays, LOPD is bolic Center in Taipei, Taiwan. The originally planned considered as a multisystemic disorder, mainly dominated by participation of Argentina and Brazil could not be real- the characteristic proximal and axial muscle weakness with a ized (formerly termed ATBIG study; ClinicalTrials.gov prominent respiratory impairment [7, 8]. Patients with Pompe identifier: NCT02824068), so as a substitution, Spain was disease have a high risk to require a non-invasive or invasive included leading to the new study acronym STIG. Inclu- ventilatory support during the course of their disease [9]. sion criteria were: aged over 8 years; willing and able to Pompe disease is one of the few neuromuscular diseases provide signed informed consent; a confirmed diagnosis of with an established drug therapy. Enzyme replacement therapy Pompe disease based on a molecular genetic examination (ERT) with alglucosidase alfa (Myozyme) received approval with two mutations in the GAA​ gene and/or the detection in Europe in 2006. Since then, it has been the only available of a decreased activity of acid alpha-glucosidase; and ERT therapy for Pompe disease and is administered every other administration for at least 3 years. Exclusion criteria were: week at a dosage of 20 mg/kg body weight. Using this ther- concurrent participation in another clinical study including apy has undisputedly led to a significant improvement in the alglucosidase alfa or other treatment; clinically significant outcomes of patients wit IOPD and LOPD [10, 11]. Studies organic disease with the exception of symptoms relating of LOPD with a follow-up of less than 5 years showed after to Pompe disease, including clinically significant cardio- an initial improvement a secondary stabilization or decline vascular, hepatic, pulmonary, neurologic, or renal disease, [12–15]. Only very few studies have investigated the long-term or other medical condition; serious intercurrent illness; or outcomes under ERT in LOPD patients [16, 17]. Those studies extenuating circumstance that precludes participation in reported that patients benefit from ERT for the first 3–5 years. the study or potentially decreases survival. For detailed However, after this first period, the majority of patients expe- demographic characteristics of our cohort, see Table 1. rienced a slow deterioration in muscle strength, walking capa- Final database lock was 31 December 2019. bility, and lung function [18]. Nevertheless, without ERT, a continuous decline in muscle strength and endurance and lung function is expected [19–21]. Study procedures and clinical assessments As new and different therapies will emerge soon, it is essen- tial to know the benefits and safety of the current standard of We conducted a retrospective, real-world, mult-center care ERT, and disease progression during treatment in patients study in patients with LOPD to evaluate efficacy and safety with LOPD. of alglucosidase alfa over a long-term period. Clinical data We, therefore, evaluated retrospective real-world data in a were collected once per year as the regulatory obligations multicenter setting, analyzing muscle strength, walking capa- differ between the four countries. Baseline was defined as bility and lung function in patients with LOPD receiving ERT the year of ERT start [baseline (BL)]. Consecutive yearly up to 14 years. clinical assessments included muscle strength, 6-min walk test, and lung function test. 13 2484 Journal of Neurology (2021) 268:2482–2492 Table 1  Study cohort characteristics at baseline For males: (7.57 × height in cm)—(1.76 × weight kg)— Characteristic at baseline Mean ± SD; median n (%) (5.02 × age)—309 m = predicted distance. For females: (2.11 × height in cm)—(2.29 × weight kg)— Follow-up duration of ERT 7.03 ± 2.98; 6.50 68 (100) (5.78 × age) + 667 m = predicted distance. (years) Patients 68 (100) Pulmonary function assessments Male 33 (48.5) Female 35 (51.5) Lung function tests included measurement of forced vital Race capacity (FVC) in a sitting and supine position. Values of Caucasian 63 (92.6) ­FVCsitting are presented in percentage of predicted normal Asian 5 (7.4) values, to adjust the effects of age, weight, height and gen- Genotype der. We received from the different centres either absolute GAA IVS1 splice site mutation 53 (78) values or %predicted. Clinical characteristics at start of ERT Individual level 6MWT, %pred 65.54 ± 17.49; 70.00 35 (51.5) MRC sum score proximal, %max 77.36 ± 13.17; 77.14 54 (79.4) For the outcome measures MRC sum score p­ roximal%max FVC sitting, %pred 69.70 ± 21.36; 75.00 57 (83.8) ­( MRC%max), ­6 MWT%pred and F ­ VC%pred, it was assessed FVC supine, l 2.22 ± 1.17; 2.20 27 (39.1) whether the individual patient improved, deteriorated or MIP 54.52 ± 30.32; 44.00 7 (10.3) remained stable during ERT. The following cut-off val- MEP 92.89 ± 57.51; 63.00 5 (7.4) ues were defined as a minimal clinically relevant change Assistive devices (MCRC). Wheelchair dependency 3 (4.4) For MRC sum score proximal ­(MRC%max): The agreement Non-invasive ventilation 20 (29.4) of three independent clinicians as an absolute change of the Invasive ventilation 3 (4.4) MRC sum score proximal of at least 2 points (> 5.70%), Comorbidities (more than 10%) was considered a relevant change. As all values of ­MRC%max Morning headache 7 (10.3) Arterial hypertension 11 (16.2) except for the first year after baseline FU1), FU7, FU10, Scoliosis 12 (17.6) FU12 were not normally distributed, the Wilcoxon-rank-test Lordosis 15 (22.1) was used comparing the mean values of ­MRC%max. For ­6MWT%pred: A review by Lachmann and Schoser [24] indicates that in 10 clinical studies on Pompe disease using the 6MWT, a relative change to baseline in the 6MWT was Muscle strength above or within the range of 5–11%, which is established for other diseases. To detect a solid clinically relevant change, The modified Medical Research Council (MRC) grading we chose a cut-off value of 10% as MCRC. Patients with a scale (0–5) was used to determine skeletal muscle strength relative change in walking distance of > 10% are determined [22]. In the proximal MRC sum score, the following mus- to have a clinically relevant change. For comparison of the cle groups were included: neck flexor, shoulder abductors, mean values of ­6MWT%pred the paired, two-sided t test was hip flexors, and hip extensors. Results are presented as a used as all criteria for this measurement were fulfilled. percentage of the maximum possible score (35). For ­FVC%pred: According to recent studies, ­FVC%pred decreases in the natural untreated course of Pompe patients 6‑min walk test of about 1–4.6% within 1 year [24]. Therefore, we defined an absolute change of > 4% as a cut-off value for categorizing Values of the 6-min walk test (6MWT), a measurement for patients in deteriorated, stable or improved. According to functional endurance, were converted in the percentage of the Shapiro–Wilk test, all values of ­FVC%pred were normal the predicted of normal [23]. We received the absolute/raw distributed. values from the included study centres and then calculated The following additional data were recorded at baseline: the % predicted for our analysis. The following formulas age, height, weight, gender, race, genotype, age at diagnosis, were used to calculate the standard value depending on the assistive devices (status of wheelchair use, ventilatory sup- age, height and weight of the patient: port), comorbidities, cardiac diagnostics. 13 Journal of Neurology (2021) 268:2482–2492 2485 Safety and/or ­FVC%pred at start of ERT < 75%, group 2 with low disease severity: neither 6­ MWT%pred nor F ­ VC%pred < 75%). To assess the safety of ERT, the following variables were gathered during treatment: infusion associated reactions, adverse events probably related to ERT, and peak titer of Results anti-GAA antibodies. Patients and baseline characteristics Statistical analysis We received data from 112 patients with LOPD. After super- For statistical analysis, SPSS statistics version 25 was vised data analysis for data completeness and consistency, used. Descriptive and explorative analysis was carried and matching with the study inclusion and exclusion criteria, out for all demographic data and baseline characteristics we could include 68 patients for the final analysis. as well as for safety assessments. For all metric, normally A total of 68 patients (female, 51.5%; Caucasian, 92.6%; distributed outcome measures comparing mean values of Asian, 7.4%; Table 1) were included in the statistical data two groups, a quantitative linear model with paired two- analysis, 48.5% from German centers, 22.1% from Italy, sided t-test was performed. The significance level was set 20.6% from Spain, and 8.8% from Taiwan. The mean ± SD at p < 0.05. The normal distribution was tested by Sha- age at diagnosis was 41.78 years ± 15.76 (median 43.50; min piro–Wilk test, respectively, and in one case by Komol- 4.00; max 69.00) and the mean ± SD age at the start of ERT gorov–Smirnov. Metric values without normal distribu- was 45.28 years ± 14.88 (median 45.50; min 13.00; max tion were analyzed and compared by the Wilcoxon-rank 72.00). The ages at diagnosis and ERT start for study center test. Due to the small sample size, a statistical analysis is summarized in Fig. 1. The mean ± SD follow-up duration of FVC supine, maximum inspiratory pressure (MIP) and under ERT in all study centers was 7.03 yrs. ± 2.98 (median maximum expiratory pressure (MEP) data was not appro- 6.50; min 3; max 14; Fig. 2, Table 1). All had enzymatically priately powered. confirmed GAA deficiency; additionally, in 85.3%, the diag- Longitudinal analysis was performed by general uni- nosis was confirmed by GAA​ gene analysis. The common variate linear models and linear regression models, testing GAA​splice site mutation was found in 78% (Table 1). whether differences in severity of the disease and other At baseline, there were three wheelchair-dependent independent variables at baseline had an impact on the patients (4.4%), and ventilator support was established for course of the disease during ERT for the main outcome 23 patients (33.8%), of which 29.4% were non-invasive, measures: ­MRC%max, ­FVC%pred and ­6MWT%pred. Cut-off and 4.4% were invasively ventilated. Comorbidities were points for disease severity at baseline were determined for assessed for neuropsychological, skeletal/orthopaedic, car- ­FVC%pred and ­6MWT%pred, each at 75%. Patients were cat- diovascular, endocrine, and respiratory issues (Supplemental egorized in two groups of disease severity (group 1 with Table 1). At baseline, before ERT start, we noted morning high disease severity: ­6MWT%pred at start of ERT < 75% headache (10.3%), arterial hypertension (16.2%), scoliosis Fig. 1  Age at diagnosis and ERT of 68 LOPD patients from the four International study centers 13 2486 Journal of Neurology (2021) 268:2482–2492 Fig. 2  Real world data: 68 LOPD patients under ERT: lost to follow-up over 14 years Fig. 3  Long term follow-up of M ­ CR%max, ­6MWT%pred, ­FVC%pred under ERT in LOPD (17.6%), and lumbar lordosis (22.1%). Further clinical char- score proximal in the percentage of maximum ­[MRC%max] acteristics at baseline are shown in Table 1. was 77.14% (mean ± SD 77.36 ± 13.17; min 45.71; max 100.00; Table 1, Fig. 3). In the first years of ERT, the MRC Muscle strength score improved significantly (p = 0.008) by 2.66% from BL to FU1 and non-significantly (p = 0.171) by 1.78% from Data on MRC sum score was available from 54 patients at BL to FU2. There was no significant decline of ­MRC%max baseline. At baseline, the median value for the MRC sum between the BL and FU4 (p = 0.282) and FU9 (p = 0.131) 13 Journal of Neurology (2021) 268:2482–2492 2487 (Supplementary Tables 2a and 2b). The longitudinal analysis decreased significantly from BL to FU4 by −3.88% revealed that disease severity at baseline has no impact on (p = 0.045) and to FU9 by 14.93% (p < 0.001) (Supple- the progression rate of ­MRC%max under ERT (p = 0.896). mental Table 4a and 4b). The progression of ­F VC %pred during ERT could be explained in most of the patients 6‑min walk test (83.5%) by the disease severity at baseline (p < 0.001). Patients with lower disease severity at baseline showed a Data on 6MWT was available from 35 patients at base- more moderate decline of F­ VC%pred. compared to patients line. At start of ERT, the median value for ­6MWT%pred with higher severity of disease at baseline. For maximal was 70.00% (435.00 m) (mean ± SD, 65.54% ± 17.49% inspiratory pressure (MIP) and maximal expiratory pres- [436.78 m ± 133.57]; min 19% (80 m); max 103% (695 m; sure (MEP), an evaluation was not appropriate due to the Table 1, Fig. 3). During ERT there was a positive effect small number of data sets. of the 6­ MWT%pred within the first 5 years. The increase in walking distance between BL and FU1 was 3.03% (Sup- plemental Table 3a). A non-significant increase of 1.93% Individual level was seen comparing the mean values of FU1 versus FU2, 0.59% for FU1 versus FU3, and 2.21% for FU1 versus FU4. Analyzing the individual progression of patients dur- In 13 patients with values for 6­ MWT%pred at BL and FU5, ing ERT, we compared the change of each patient with there was a mean values increase of 4.00% (p = 0.243). At data available for FU from year to year categorizing the 10-year treatment time-point, there was a non-significant them in three groups: improved, stabilized, deteriorated. decline compared to BL mean value (−16.5%; SD 26.69; The cut-off levels for a MCRC are explained in detail in p = 0.304) (Supplemental Table 3a and 3b). The 6­ MWT%pred “Methods”. long-term data was independent of the disease severity at For ­MRC%max the majority of patients remained stable baseline (p = 0.862). from BL to FU1 and in the following years up to FU9 (Table 2). Comparing the ­M RC %max from BL, on aver- Lung function assessments age, 1/3 of patients improved, 1/3 remained stable and 1/3 deteriorated overall during the following years up to FU9 Data on FVC was available from 57 patients at baseline. (Table 2, Fig. 3). At BL, the median value for F ­ VC %pred in sitting posi- For ­F VC %pred between BL and FU1, there was an tion was 75.00% (mean ± SD 69.70 ± 21.36; min 17.00; improvement in 30% of patients, no relevant change in max 125.00; Table 1, Fig. 3). Patients remained stable 46%, and deterioration in 24% (Table 2, Fig. 3). between BL and FU1 with a change of 2.00% (p = 0.237; For ­6 MWT %pred, 24% of patients improved and 59% Supplemental Table 4a), followed by a secondary decline were stable at FU1 related to BL. Over the first 6 years of mean values for F ­ VC %pred in all following years by of FU, the majority of patients (> 60%) had stable values −1.82 to −21.25% (compared to FU1). The mean ­FVC%pred compared to the last year (Table 2, Fig. 3). Table 2  Comaprison of outcome measures categorized in deterioration, stabilization, improvement year by year MRC%max N (%) FVC%pred N (%) 6MWT%pred N (%) N Decreased Stable Improved N Decreased Stable Improved N Decreassed Stable Improved BL-FU1 44 5 (11.4) 24 (54.5) 15 (34.1) 46 11 (23.9) 21 (45.7) 14 (30.4) 29 5 (17.2) 17 (58.6) 7 (24.1) FU1-FU2 46 9 (19.6) 25 (54.3) 12 (26.1) 43 12 (27.9) 24 (55.8) 7 (16.3) 28 3 (10.7) 21 (75.0) 4 (14.3) FU2-FU3 36 15 (41.7) 17 (47.2) 4 (11.1) 37 17 (45.9) 17 (45.9) 3 (8.1) 26 7 (26.9) 17 (65.4) 2 (7.7) FU3-FU4 23 4 (17.4) 16 (69.6) 3 (13.0) 24 10 (41.7) 8 (33.3) 6 (25.0) 18 1 (5.6) 11 (61.1) 6 (33.3) FU4-FU5 22 3 (13.6) 15 (68.2) 4 (18.2) 27 5 (18.5) 11 (40.7) 11 (40.7) 17 5 (29.4) 11 (64.7) 1 (5.9) FU5-FU6 14 2 (14.3) 11 (78.6) 1 (7.1) 21 5 (23.8) 13 (61.9) 3 (14.3) 13 6 (46.2) 5 (38.5) 2 (15.4) FU6-FU7 10 2 (20.0) 8 (80.0) 0 (0.0) 14 2 (14.3) 10 (71.4) 2 (14.3) 9 1 (11.1) 5 (55.6) 3 (33.3) FU7-FU8 12 2 (16.7) 8 (66.7) 2 (16.7) 12 3 (25.0) 7 (58.3) 2 (16.7) 8 2 (25.0) 5 (62.5) 1 (12.5) FU8-FU9 5 2 (40.0) 3 (60.0) 0 (0.0) 8 2 (25.0) 4 (50.0) 2 (25.0) 7 3 (42.9) 4 (57.1) 0 (0.0) FU9-FU10 4 0 (0.0) 3 (75.0) 1 (25.0) 7 5 (71.4) 2 (28.6) 0 (0.0) 6 1 (16.7) 5 (83.3) 0 (0.0) FU10-FU11 4 2 (50.0) 1 (25.0) 1 (25.0) 6 1 (16.7) 2 (33.3) 3 (50.0) 3 3 (100.0) 0 (0.0) 0 (0.0) FU11-FU12 1 0 (0.0) 1 (100.0) 0 (0.0) 3 2 (66.7) 0 (0.0) 1 (33.3) 0 0 (0.0) 0 (0.0) 0 (0.0) FU12-FU13 2 0 (0.0) 2 (100.0) 0 (0.0) 2 1 (50.0) 1 (50.0) 0 (0.0) 1 0 (0.0) 1 (100.0) 0 (0.0) 13 2488 Journal of Neurology (2021) 268:2482–2492 Table 3  Linear regression of demographic, clinical and genetic fac- sum score. In two patients, a decline in 6­ MWT%pred, mean tors across the main outcome measures (­MRC%max, ­6MWT%pred and values from BL to FU12 was observed. In eight patients, ­FVC%pred) ­FVC%pred declined (Fig. 4, Supplemental Table 4a). Outcome measurement Independent variable Corrected R2 p level Diff 6MWT mean Gender (N = 39) –0.026 0.852 Supportive devices MRC at BL (N = 34) –0.030 0.862 Ventilation status Mutation (N = 37) –0.028 0.869 Diff FVC mean Gender (N = 48) –0.010 0.461 At BL, 20 patients were non-invasively ventilated. During MRC at BL (N = 41) –0.013 0.482 the observation period of up to 14 years, 18 patients started Mutation (N = 47) –0.017 0.637 non-invasive ventilation, while invasive ventilation was initi- Diff MRC mean Gender (N = 51) –0.019 0.791 ated in three patients. MRC at BL (N = 46) –0.000 0.896 Mutation (N = 49) –0.017 0.644 Wheelchair use Linear regression (independent variable: gender male–female, ­MRC%max at baseline (BL), GAA IVS1 splice site mutation vs. other While at BL three patients were wheelchair dependent, nine mutations; dependent variable: ­MRC%max, ­6MWT%pred and ­FVC%pred); only patients with at least two times with 2 consecutive years were additional patients became non-ambulatory in the course of considered in the analysis the study. Long‑term side effects In the linear regression model, we did not find any sig- nificant association between the tested clinical, genetic and Serial anti-GAA antibody titers were tested in 62.3% of demographic factors across the main outcome measures patients (analyses performed at Sanofi Genzyme laboratory ­(MRC%max, ­6MWT%pred and ­FVC%pred) (Table 3). services) and varied between 0 and 204,800, with the major- ity of patients (27.5%) ranging from 1:1,000 to 1:10,000, Long‑term data longer than 10 years under ERT 10.1% below 1:1,000, and 8.7% above 1:10,000. There were 12 infusion-associated reactions in 9 patients, two of which Due to our small number of patients with a follow-up on led to the termination of ERT without a restart (bronchial ERT of more than 10 years, a statistical analysis was of spasm with flushing—antibody titer in the following year of limited value. The muscle strength of five patients for 1:204,800; pulmonary embolism during infusion—antibody up to 12 years showed no significant change in the MRC titer in the following year of 1:51,200). In all other cases of Fig. 4   Individual ­MCR%max scores of five LOPD patients under ERT upto 14 years 13 Journal of Neurology (2021) 268:2482–2492 2489 Table 4  Infusion related events (IRE) Year of FU Type of IRE Treatment/premedication before infusion Study centre (pat no.) 1 Infusion related reaction (anamnestically) n.i Germany (053) Anaphylactic reaction Antihistaminics + corticosteroids Taiwan (022) 2 Bronchial spasms Antihistaminics + corticosteroids Germany (002) Bronchospasm with flushing; low exacerbation of an erythema Antihistaminics + corticosteroids Germany (058) with flushing (antibody titer 1:204,800 in FU2) Stop of ERT Nausea and vomiting Antihistaminics Germany (052) 3 Possibly anaphylactic skin reaction 05/2008 n.i Germany (056) Flushing; dyspnoe, increase in blood pressure Antihistaminics + corticosteroids Germany (058) Mild anaphylactic skin reaction (pruritus at hands and feet) Antihistaminics + corticosteroids Germany (065) 4 Probably mild anaphylactic reaction n.i Germany (055) Proteinuria < 1 g (deposition of immune complexes due to ERT?) n.i Germany (043) 5 Shivering and fever during infusion Antihistaminics + corticosteroids Germany (055) 6 Pulmonary embolism during infusion* (antibody titer 1:51,200 Anticoagulants Germany (055) in FU6) Stop of ERT infusion-related events, ERT was continued—in six patients The results of this study are in line with outcomes using appropriate antihistamine and/or steroidal premedica- reported from other studies with similar or shorter follow- tion and one patient (Germany ID 043) without any known up periods as well as with the results of MRI studies show- premedication. Additionally, in another six patients, side ing a progressive increase in a fat fraction in the muscles of effects were classified as unrelated to ERT (Table 4). patients receiving ERT [25]. Positive effects are reported within the first ERT period with a secondary decline seen in the majority of patients, i.e. in the first two months [12], in Discussion the first 2–3 years [13, 14] or in the first 3–5 years, respec- tively [15, 18]. In our cohort, after a stable first treatment In this real world, retrospective, long-term follow-up of 68 period between baseline and FU1 (2.0%, p = 0.237), there patients with LOPD receiving ERT, we analysed the clinical was a secondary decline of mean values in ­FVC%pred in course of muscle strength, forced respiratory lung capacity, all the following years by −1.82 to −21.25%, which is in and 6-min walking capacity. The aim of the STIG Study contrast to the results of the Late-Onset-Treatment-Study was to evaluate the efficacy of alglucosidase alfa in a long- (LOTS) extension study [15, 26]. Our results are compara- term period, as to date there is only limited knowledge on ble with the published French Pompe registry data, describ- long-term courses over more than 5 years. Only two ERT ing an initial improvement in 6MWT with a cut-off point at drop-outs due to side effects of ERT were noticed in the 2.2 years and a significant improvement in motor function total of 112 original screened patients with LOPD for this in the first 3 years. In the French Pompe registry study, FVC analysis. So, our analysis of outcomes is not biased regard- showed the weakest effect, similar to our data, by showing a ing this issue. A beneficial aspect of this real-world study decline of 0.9% per year during ERT. Another international is the long-term ERT follow-up of 3 to up to 14 years in Pompe Registry study showed a stabilization of lung func- patients with LOPD. The use of %pred for the 6MWT and tion with a slope of 0.17% per year over a 5-years follow-up the FVC allowed a reliable comparison of values independ- period [16]. ent of gender and age. On the individual level in our cohort, the M ­ RC%max com- Overall, the best therapeutic ERT effect at the group level pared to baseline, on average, improved in 1/3 of patients. was shown for the 6MWT. There was a sustained positive For the ­6MWT%pred within the first 6 years during ERT, effect until FU7. Muscle strength remained stable over the the majority of patients (≥ 60%) had stable or improved years at the beginning followed by a steady decline. FVC values compared to baseline. Long-term data beyond year showed the most evident decline with significant deteriora- 7 remain uncertain, as less than 10 patients could be ana- tion after ERT start over the course of 5 years and up to lysed. Nevertheless, there was a deterioration of at least nine 10 years. Analogous to this pulmonary function deteriora- patients becoming non-ambulant. In the future, studies on tion, ventilatory support was initiated in 21 patients during the informative value would be improved and strengthened the observation period. if a patient-reported outcome measure (PRO) is recorded 13 2490 Journal of Neurology (2021) 268:2482–2492 in parallel to the objective test to determine the clinical rel- data were collected from multiple centers and each center evance for the patient. has a different algorithm for clinical follow-up, and at study For ­FVC%pred, an overall more or less stable period up initiation, a disease-specific PRO did not exist. Furthermore, to year 6 was seen. The majority of patients relevantly we did not have an untreated patient cohort for statistical decreased in their F ­ VC%pred after 6 years on ERT. As the comparison. As this is a retrospective real-world study, there decline of Pompe patients in ­FVC%pred is assumed to be may be some variation in the performance of the tests and about 1–4.6% per year in a natural course [24], and as a outcome measures, as this STIG study is not based on a deterioration in our study is defined as a change of more standardized and controlled protocol defining in advance the than 4%, the majority of our patients had still a limited ben- methodology. However, this probable inaccuracy is unlikely efit under ERT as without ERT therapy. Nevertheless, we to have had a major impact on data evaluation, particularly found an increase of 33% from initially 1/3 non-invasive in the 6MWT or FVC, as all participating STIG centers were or invasive ventilated patients to 2/3 of all LOPD patients. trained Pompe experts, used to performing standardized out- Eighteen patients needed to start non-invasive ventilation. comes according to clinical trial regulatory obligations. Whether this additional effective pulmonary therapy contrib- utes substantially to the stabilization of LOPD patients in the late disease phase during ERT remains open. Conclusion Confounders that may also have influenced the effect dur- ing ERT compared to placebo effects, could be, in a positive The STIG study demonstrated that alglucosidase alfa is jus- sense, the motivation for physical activity and adapted nutri- tifiable to be used as the first-line therapy for Pompe disease. tion, initialized by more regular monitoring or start of venti- There is an initial positive effect on the most important out- lation [27, 28]; in a negative sense, more or less pronounced come measures, however, a more limited long-term benefit comorbidities. Our results show that the disease severity at of stabilization of the clinical course under ERT for many ERT start seems not to have a major impact on the overall patients over a long period. Though, according to our data, clinical long-term course of patients under ERT, which is this long-term therapeutic efficacy is weakest for the lung consistent with the latest results of a long-term study with capacity and consecutively the need of additional ventilatory Pompe patients under ERT from the Erasmus study group support over time. As respiratory insufficiency is the most [18]. However, our STIG study reveals one important dif- frequent cause of death in Pompe disease [21], it is impor- ference to former studies—our results on the FVC decline tant to further improve this organ function in particular. The and the increase of additionally needed non-invasive and fast approval of novel therapeutic options is a great unmet invasive ventilation requirements in the LOPD study cohort. need for Pompe patients. According to initial interim evalu- Overall, based on the data received, ERT was well toler- ations, enzyme replacement therapy with the new modified ated among this long-term period of observation. Only two drug avalglucosidase alfa appears to have a positive effect of 12 observed infusion associated reactions led to the stop on lung capacity [30]. The clinical effect of recombinant of ERT. In addition, antibodies titers did not seem to influ- alpha-glucosidase in the presence of a chaperone leading to ence the ERT response. In all other patients, ERT could be improved stability of the ERT is in parallel under investiga- continued with appropriate antihistamine and/or steroidal tion [31]. For upcoming gene therapy trials, the safety profile premedication. Nevertheless, there should be a clear aware- of these drugs and their therapeutic impact on lung function ness of anaphylactic reactions occurring during regular ERT outcomes will be a focus of interest. at any time [29]. A common reported event was fall, prob- Finally, based on the real-world character of the STIG ably more as a result of the muscle weakness due to the study, it is highly recommended that the harmonized interna- disease itself in all LOPD patients instead of a side effect of tional standards for clinical follow-up need to be established. ERT. Overall, there are no novel adverse events findings or In addition to the frequently and regularly performed out- infusion associated reactions to be reported. come measures like MRC, FVC sitting and 6MWT, espe- Limitations of this study are our retrospective real-world cially FVC in the supine position, MIP and MEP as well as approach so that a considerably loss of patients during disease-specific PROs, e.g. R-PACT, and timed functional the follow-up period has happened. The rising number of outcome measures should be part of each routine follow-up patients lost to follow up is the reason why only a limited visit, at least once a year [32]. comparison of the values was possible after year 7 of ERT. Another limitation, not only of this study, but rather of any Supplementary Information The online version contains supplemen- Pompe disease data collection is that data on MIP and MEP tary material available at https:​ //doi.org/10.1007/s00415​ -021-10409-​ 9. are not available, as they were not collected in a clinical routine yet. An explanation for the lack of data, especially Acknowledgements We would like to thank all participating study centers for their commitment and time as well as the participants for for patient-reported outcomes (PROs), is the fact that our 13 Journal of Neurology (2021) 268:2482–2492 2491 their consent to analyze their data in this study. We thank Haris Babačić 2. Joannes Cassianus P (1932) Over idiopatische hypertrophie van for the initial set-up of the ATBIG/STIG database. This paper is dedi- het hart. Ned Tijdschr Geneeskd 76:304 cated to our dear colleague and friend Dr. Zoltan Janos Lukacs who 3. http://www.pompe​varia​n tdat ​a base​. nl/pompe​_ mutat​i ons_list. sadly died during the COVID19 pandemic in summer 2020. php?goto=5&order​by=aMut_ID1. 26 March-2020 4. Lukacs Z et al (2016) Prevalence of Pompe disease in 3,076 Author contributions KG: Statistical analysis and interpretation of patients with hyperCKemia and limb-girdle muscular weakness. data, discussion of results, first manuscript draft, critical revision of Neurology 87(3):295–298 the manuscript for intellectual content. JDM, OM, Y-HC, KCK, SW, 5. Meikle PJ et al (1999) Prevalence of lysosomal storage disorders. FM, PW-LH, AT: Acquisition of data, interpretation of data, criti- JAMA 281(3):249–254 cal revision of the manuscript for intellectual content. AP, GT, JA-P: 6. Poorthuis BJ et al (1999) The frequency of lysosomal storage Acquisition of data. BS: Acquisition of data, initiation, design and diseases in The Netherlands. Hum Genet 105(1–2):151–156 concept of the study, analysis and interpretation of all data, discussion 7. van der Beek NA et al (2006) Pompe disease (glycogen storage of results, critical revision of the manuscript for intellectual content, disease type II): clinical features and enzyme replacement therapy. final manuscript draft. Acta Neurol Belg 106(2):82–86 8. Toscano A, Rodolico C, Musumeci O (2019) Multisystem late onset Pompe disease (LOPD): an update on clinical aspects. Ann Funding Open Access funding enabled and organized by Projekt Transl Med 7(13):284 DEAL. This study was supported by an unrestricted research grant 9. Wenninger S et al (2019) Safety and efficacy of short- and long- from Sanofi Genzyme to BS. The sponsor of the study did not affect term inspiratory muscle training in late-onset Pompe disease any part of the design or evaluation of this study. (LOPD): a pilot study. J Neurol 266(1):133–147 10. Hahn SH et al (2018) Efficacy, safety profile, and immuno- Availability of data and material The data that support the findings genicity of alglucosidase alfa produced at the 4,000-liter of this study are available from the corresponding author, BS, upon scale in US children and adolescents with Pompe disease: reasonable request. ADVANCE, a phase IV, open-label, prospective study. Genet Med 20(10):1284–1294 Compliance with ethical standards 11. Chien YH et al (2015) Long-term prognosis of patients with infantile-onset Pompe disease diagnosed by newborn screening Conflicts of interest KG, FM, SW received travel support and speaker and treated since birth. J Paediatr 166(4):985–991 honoraria from Sanofi Genzyme. BS received funding for research, 12. Schoser B et al (2017) Survival and long-term outcomes in late- honoraria for the participation of advisory board meetings and con- onset Pompe disease following alglucosidase alfa treatment: a sulting, and speaker fees from Sanofi Genzyme, Amicus therapeutic, systematic review and meta-analysis. J Neurol 264(4):621–630 Audentes therapeutics, Nexien, and Lupin. AT and JDM received hon- 13. Kuperus E et al (2017) Long-term benefit of enzyme replacement oraria for the participation of advisory board meetings and for teach- therapy in Pompe disease: a 5-year prospective study. Neurology ing activities, KCK received lecture fees and advisory boards from 89(23):2365–2373 Sanofi-Genzyme and Lupin. All other authors declare that they have 14. Scheidegger O et al (2018) 36-Months follow-up assessment after no conflicts of interest. cessation and resuming of enzyme replacement therapy in late onset Pompe disease: data from the Swiss Pompe Registry. J Neu- Ethics approval The study was approved by the Ethics Committee of rol 265(12):2783–2788 the Ludwig Maximilian University of Munich, Germany (IRB vote 15. Nagura H, Hokugo J, Ueda K (2019) Long-Term observation 231-16) in 2016 and has therefore been performed in accordance with of the safety and effectiveness of enzyme replacement therapy the ethical standards laid down in the 1964 Declaration of Helsinki and in Japanese patients with pompe disease: results from the post- its later amendments. All patients gave an informed consent for their marketing surveillance. Neurol Ther 8(2):397–409 participation in the study and publication of the results. 16. Semplicini C et al (2020) Long-term benefit of enzyme replace- ment therapy with alglucosidase alfa in adults with Pompe dis- ease: prospective analysis from the French Pompe Registry. J Open Access This article is licensed under a Creative Commons Attri- Inherit Metab Dis 43:1219–1231 bution 4.0 International License, which permits use, sharing, adapta- 17. Stockton DW et al (2020) Respiratory function during enzyme tion, distribution and reproduction in any medium or format, as long replacement therapy in late-onset Pompe disease: longitudinal as you give appropriate credit to the original author(s) and the source, course, prognostic factors, and the impact of time from diagnosis provide a link to the Creative Commons licence, and indicate if changes to treatment start. J Neurol 267:3038–3053 were made. The images or other third party material in this article are 18. Harlaar L et al (2019) Large variation in effects during 10 years included in the article’s Creative Commons licence, unless indicated of enzyme therapy in adults with Pompe disease. Neurology otherwise in a credit line to the material. If material is not included in 93(19):e1756–e1767 the article’s Creative Commons licence and your intended use is not 19. van der Beek NA et al (2011) Rate of progression and predictive permitted by statutory regulation or exceeds the permitted use, you will factors for pulmonary outcome in children and adults with Pompe need to obtain permission directly from the copyright holder. To view a disease. Mol Genet Metab 104(1–2):129–136 copy of this licence, visit http://creat​iveco​mmons​.org/licen​ses/by/4.0/. 20. Schoser B et al (2016) Survival and long-term outcomes in late- onset Pompe disease following alglucosidase alfa treatment: a systematic review and meta-analysis. J Neurol 26:621–630 21. Gungor D et al (2011) Survival and associated factors in 268 References adults with Pompe disease prior to treatment with enzyme replace- ment therapy. Orphanet J Rare Dis 6:34 1. Hers HG (1963) alpha-Glucosidase deficiency in generalized gly- 22. Compston A (2010) Aids to the investigation of peripheral nerve cogenstorage disease (Pompeʼs disease). Biochem J 86:11–16 injuries Medical Research Council: Nerve Injuries Research Com- mittee. His Majesty’s Stationery Office: 1942; pp. 48 (iii) and 74 figures and 7 diagrams; with aids to the examination of the 13 2492 Journal of Neurology (2021) 268:2482–2492 peripheral nervous system. By Michael O’Brien for the Guaran- 29. van der Ploeg AT et al (2010) A randomized study of alglu- tors of Brain. Saunders Elsevier 2010 pp. [8] 64 and 94 Figures. cosidase alfa in late-onset Pompe’s disease. New Engl J Med Brain 133(10):2838–2844 362(15):1396–1406 23. ATS statement (2002) guidelines for the six-minute walk test. Am 30. Pena LDM et al (2019) Safety, tolerability, pharmacokinet- J Respir Crit Care Med 166(1):111–117 ics, pharmacodynamics, and exploratory efficacy of the novel 24. Lachmann R, Schoser B (2013) The clinical relevance of outcomes enzyme replacement therapy avalglucosidase alfa (neoGAA) in used in late-onset Pompe disease: can we do better? Orphanet J treatment-naive and alglucosidase alfa-treated patients with late- Rare Dis 8:160 onset Pompe disease: A phase 1, open-label, multicenter, multina- 25. Figueroa-Bonaparte S et al (2018) Quantitative muscle MRI to tional, ascending dose study. Neuromuscul Disord 29(3):167–186 follow up late onset Pompe patients: a prospective study. Sci Rep 31. Parenti G et al (2014) A chaperone enhances blood alpha-glu- 8(1):10898 cosidase activity in Pompe disease patients treated with enzyme 26. de Vries JM et al (2012) Effect of enzyme therapy and prognostic replacement therapy. Mol Ther 22(11):2004–2012 factors in 69 adults with Pompe disease: an open-label single- 32. van der Ploeg AT, Kruijshaar ME, Toscano A, Laforêt P, Ange- center study. Orphanet J Rare Dis 7:73 lini C, Lachmann RH, Pascual Pascual SI, Roberts M, Rösler 27. van den Berg LE et al (2015) Safety and efficacy of exercise train- K, Stulnig T, van Doorn PA, Van den Bergh PYK, Vissing J, ing in adults with Pompe disease: evalution of endurance, muscle Schoser B (2017) European consensus for starting and stopping strength and core stability before and after a 12 week training enzyme replacement therapy in adult patients with Pompe disease: program. Orphanet J Rare Dis 10:87 a 10-year experience. Eur J Neurol 24(6):768–e31 28. Tarnopolsky MA, Nilsson MI (2019) Nutrition and exercise in Pompe disease. Ann Transl Med 7(13):282 13