Efficacy and safety of cipaglucosidase alfa plus miglustat in Japanese adults with late-onset Pompe disease: a case series from clinical trials

Abstract

Background

The safety and efficacy of cipaglucosidase alfa plus miglustat (cipa+mig) versus alglucosidase alfa plus placebo (alg+pbo) in late-onset Pompe disease (LOPD) were established in the randomised, double-blind, 52-week PROPEL study (NCT03729362) conducted across 24 countries, including Japan. An open-label extension (OLE; NCT04138277) is evaluating long-term cipa+mig outcomes. In Japan, cipa+mig was also available before marketing authorisation in an expanded access study (ATB200-19). This case series summarises the efficacy and safety of cipa+mig in Japanese patients in PROPEL, PROPEL OLE and ATB200-19.

Methods

Endpoints included 6-minute walk distance (6MWD), forced vital capacity (FVC), biomarkers and safety. Cipa+mig (20 mg/kg intravenous cipa and 130–260 mg oral mig (weight-based dosing)) and alg+pbo (20 mg/kg intravenous alg and oral pbo) were given biweekly. Patients randomised to alg+pbo in PROPEL received cipa+mig in the OLE. Safety was assessed at every visit, and biomarkers at weeks 2, 4, 6, 12, 26, 38, 52 and every 26 weeks thereafter.

Results

Six patients from PROPEL/OLE (cipa+mig, n=2; alg+pbo, n=4; all enrolled in the OLE) and four from ATB200-19 were included. Two patients discontinued treatment: one in the OLE (anaphylactic reaction) and one in ATB200-19 (withdrawal of consent). For evaluable patients (n=8), 6MWD remained stable or improved from baseline to weeks 52 and 104. In PROPEL/OLE, FVC varied with most patients improving with cipa+mig treatment; in ATB200-19, FVC remained stable to week 104. In PROPEL/OLE and ATB200-19, biomarker levels improved or stabilised by week 104, including in the PROPEL switch group, where levels worsened at week 52 but improved after switching to cipa+mig. Seven patients experienced treatment-related adverse events. No deaths occurred.

Conclusions

Although the number of patients was small, the efficacy and safety of cipa+mig in Japanese patients from PROPEL/OLE and ATB200-19 were generally sustained and durable responses consistent with positive results in the overall PROPEL population.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• The global PROPEL study (NCT03729362) showed clinically meaningful improvements for cipaglucosidase alfa plus miglustat in motor function, pulmonary function and biomarkers compared with alglucosidase alfa in ambulatory patients with late-onset Pompe disease (LOPD) in the overall and enzyme replacement therapy-experienced populations.

WHAT THIS STUDY ADDS

• This study shows that the efficacy and safety of cipaglucosidase alfa plus miglustat in Japanese patients from PROPEL, its open-label extension (NCT04138277) and the open-label expanded access study ATB200-19, which also included patients with advanced disease, non-ambulatory patients and those with a low body weight, were indicative of sustained and durable responses to cipaglucosidase alfa plus miglustat treatment. No new safety concerns were identified in Japanese patients.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This study supports the use of cipaglucosidase alfa plus miglustat in Japanese patients with LOPD, including those with advanced disease, low body weight and previous enzyme replacement therapy.

Introduction

Late-onset Pompe disease (LOPD), a rare, multisystemic, heterogeneous disorder caused by acid α-glucosidase (GAA) deficiency, is characterised by progressive loss of skeletal and respiratory muscle function.^{1 2} As LOPD progresses, patients may require a wheelchair and assisted ventilation.^{2 3} The first disease-specific therapy approved for Pompe disease was enzyme replacement therapy (ERT) with the recombinant human GAA alglucosidase alfa (alg).⁴ Despite initial efficacy in the first years of use, many patients receiving alg experience a decline in outcomes over time,^5–10 possibly due to poor alg uptake into skeletal muscle.^{4 11} Cipaglucosidase alfa plus miglustat (cipa+mig) was designed for effective delivery and uptake of cipa into muscle cells and to yield the most active form of the enzyme through complete processing inside the cell.^{4 12 13}

The approval of cipa+mig in Japan, the EU and the USA^14–16 was based on the pivotal randomised, phase 3, double-blind, 52-week PROPEL study (ATB200-03; NCT03729362) assessing the efficacy and safety of cipa+mig versus alg plus placebo (pbo) in ambulatory adults with LOPD across 24 countries. Most patients in PROPEL (77%) received alg prior to the study (median treatment duration: 7.4 years).^{17 18} While the primary endpoint (change in 6-minute walk distance (6MWD) from baseline to week 52 in the overall population) did not confirm statistically significant superiority for cipa+mig versus alg+pbo (between-group difference 13.6 m (95% CI −2.8 to 29.9)), the numerical improvement and evidence from secondary efficacy endpoints suggested clinically meaningful improvements for cipa+mig in motor function, pulmonary function and biomarkers in the overall and ERT-experienced populations.¹⁷ In the overall population, the % predicted sitting forced vital capacity (FVC) between-group difference for cipa+mig versus alg+pbo was 3.0% (95% CI 0.7 to 5.3).¹⁷ In ERT-experienced patients, the mean (SE) between-group difference for cipa+mig versus alg+pbo was 16.9 (8.8) m for 6MWD (p=0.047) and 4.1 (1.2)% for sitting FVC (p=0.006).¹⁷ In an open-label extension (OLE; ATB200-07; NCT04138277),¹⁹ cipa+mig was associated with maintained improvements (6MWD, biomarkers) or stabilisation (FVC) from PROPEL baseline and was well tolerated up to 104 weeks.¹⁹

In Japan, ERT (alg) has been available for patients with Pompe disease since 2007.⁸ However, an unmet need for effective disease control remained. Hence, adults with LOPD were permitted to receive cipa+mig prior to marketing authorisation in an open-label, multicentre, expanded access study evaluating the safety and effectiveness of cipa+mig (ATB200-19; jRCT2061210056).^{16 20} Enrolment criteria allowed participation of a broader range of patients than those eligible for PROPEL (eg, no body weight limitation, no screening threshold 6MWD or % predicted FVC values).¹⁷

This case series provides an overview of the efficacy and safety of cipa+mig in Japanese adults with LOPD in PROPEL, the PROPEL OLE and ATB200-19.

Methods

Trial designs and endpoints

PROPEL and the PROPEL OLE

The design, eligibility criteria and assessments in PROPEL and the OLE have been published previously.^{17 19} Briefly, patients in PROPEL had documented LOPD based on GAA enzyme deficiency or GAA genotyping, were either ERT experienced (≥2 years’ prior treatment with alg 20 mg/kg every 2 weeks) or ERT naïve, weighed ≥40 kg, had a sitting FVC of ≥30%, and a 6MWD of ≥75 m and ≤90% of the predicted value for healthy adults. Patients using ventilation support for >6 hours/day while awake were ineligible for participation.¹⁷ The primary efficacy endpoint was change from baseline to week 52 in 6MWD. Key secondary endpoints included change from baseline to week 52 in sitting FVC (% predicted) and manual muscle test (MMT) lower extremity. Biomarker tests included serum creatine kinase (CK) and urine hexose tetrasaccharide (Hex4).¹⁷ The primary objective of the PROPEL OLE was to assess the long-term safety and tolerability of cipa+mig.¹⁹ Key assessments in the OLE were the same as in PROPEL. The cipa+mig group included patients randomised to cipa+mig in PROPEL and continued in the OLE (2 years of treatment with cipa+mig after 52 weeks of the OLE). The switch group included patients randomised to alg+pbo in PROPEL who switched to cipa+mig in the OLE (1 year of treatment with cipa+mig after 52 weeks of the OLE).¹⁹

ATB200-19

ATB200-19 included adults with documented LOPD based on GAA enzyme deficiency or GAA genotyping, regardless of body weight, ambulatory status or use of ventilation support. Ten patients were planned for enrolment. Efficacy assessments (including 6MWD, FVC and MMT lower extremity) were not mandatory but performed according to institutional practice and depending on disease severity. Safety was assessed at every visit; biomarker levels (CK and Hex4) were determined at weeks 2, 4, 6, 12, 26, 38, 52 and every 26 weeks thereafter.

Treatment

Patients in PROPEL were randomised 2:1 to intravenous cipa (20 mg/kg) plus oral mig (195 mg for patients weighing 40 to <50 kg; 260 mg for patients weighing ≥50 kg) or intravenous alg (20 mg/kg) plus oral pbo.¹⁷ All patients in the PROPEL OLE received cipa+mig (same dosing as PROPEL).¹⁹ Patients in ATB200-19 received cipa (20 mg/kg) plus mig (130 mg for patients weighing 20 to <30 kg, 195 mg for patients weighing 30 to <50 kg, 260 mg for patients weighing ≥50 kg). Treatment will continue until withdrawal criteria are met. In all studies, treatment was administered once every 2 weeks.

Statistical analysis

The analysis of the Japan subpopulation from PROPEL and the OLE was post hoc and included 6MWD, % predicted sitting FVC, biomarkers (CK and Hex4) and treatment-emergent adverse events (TEAEs). Efficacy and biomarker data are reported as change from PROPEL baseline to PROPEL week 52 and to OLE week 52 (104 weeks after the PROPEL baseline). ATB200-19 results were analysed using descriptive statistics. Observed data are reported for all outcomes in the Japan subpopulation, while changes from baseline to week 52 in the global overall and ERT-experienced PROPEL populations were reported as last observation carried forward.

Results

Patients

The PROPEL Japan subpopulation included six ERT-experienced patients who were randomised and treated with cipa+mig (cipa+mig group; n=2) or alg+pbo (switch group; n=4); all entered the PROPEL OLE. One patient (patient 5) in the switch group discontinued the OLE before the data cut-off (DCO; 11 January 2022) due to an adverse event (anaphylactic reaction). Median (IQR) duration of cipa+mig treatment was 25.4 (25.3–25.4) months in the cipa+mig group and 13.8 (7.6–14.3) months in the switch group. At the ATB200-19 DCO (31 July 2024), four patients (all ERT experienced) were enrolled and treated; one patient (patient 10) discontinued study treatment (withdrawal of consent). Median (IQR) treatment duration was 27.0 (16.2–30.4) months. GAA genotypes for all patients are described in online supplemental table S1.

Case studies

PROPEL OLE cipa+mig group

Patient 1 weighed 40.1 kg, was aged 31–40 years and was treated with alg for 11.6 years prior to PROPEL and received cipa+mig in PROPEL. The total duration of cipa+mig treatment during PROPEL and the OLE was 25.4 months. At baseline, 6MWD was 299.5 m (39.7% predicted) and sitting FVC was 56.0% predicted. Further baseline demographics and disease characteristics for all patients are summarised in tables 1–4. Over 104 weeks of treatment with cipa+mig, 6MWD was relatively stable. Per cent predicted sitting FVC improved from week 12 and was generally maintained through to week 104. The patient’s MMT lower extremity score worsened at week 12 and was generally maintained to week 104. Biomarker (CK and Hex4) levels improved (decreased) overall from baseline to week 104 (figure 1, online supplemental figure S1). By the PROPEL OLE DCO, the patient had no TEAEs considered related to study treatment, TEAEs leading to study drug discontinuation or serious adverse events (SAEs; table 5). All reported TEAEs are summarised in online supplemental table S2.

Table 1. Demographics and baseline characteristics for the PROPEL/PROPEL OLE Japan subpopulation.

	Cipa+mig (n=2)			Switch group (n=4)
	Patient 1	Patient 2	Mean (SD)	Patient 3	Patient 4	Patient 5	Patient 6	Mean (SD)
Age at informed consent date, years*	31–40	21–30	32.5 (3.5)	21–30	61–70	51–60	21–30	43.0 (17.7)
Age at diagnosis, years	23	17	20.0 (4.2)	11	55	43	12	30.3 (22.2)
Time from diagnosis to start of study, years	11.9	13.1	–	19.8	6.3	10.9	13.5	–
Weight (kg)	40.1	64.5	52.3 (17.2)	48.7	64.0	82.1	49.5	61.1 (15.7)
ERT experienced at study entry	Yes	Yes	–	Yes	Yes	Yes	Yes	–
ERT duration prior to study, years	11.6	12.3	12.0 (0.5)	13.2	6.2	9.7	12.2	10.3 (3.1)
Use of ambulation-assistive devices at baseline	No	No	–	Yes	Yes	No	Yes	–
Invasive ventilation assistance at baseline	No	No	–	No	No	No	No	–
History of falls	No	No	–	Yes	No	No	Yes	–
History of IARs	No	Yes	–	Yes	No	Yes	Yes	–

^*Age ranges are shown for individual patients to protect their identity.

cipa+mig, cipaglucosidase alfa plus miglustat; ERT, enzyme replacement therapy; IAR, infusion-associated reaction; OLE, open-label extension.

Table 4. Baseline assessments for ATB200-19.

	Cipa+mig (n=4)
	Patient 7	Patient 8	Patient 9	Patient 10	Mean (SD)
Baseline 6MWD*
Metres	289	566	–	–	–
% predicted	–	–	–	–	–
Sitting FVC, % predicted*	14.5	84.5	–	–	–
MMT lower extremity total score*,†	24	38	15	–	–
Baseline serum CK (U/L)	455	347	180	52	258.5 (178.2)
Baseline urine Hex4 (mmol/mol creatinine)	45.3	10.1	21.0	26.5	25.7 (14.7)

^*Efficacy assessments were not possible for Patients 9 and 10 due to advanced disease status.

^†Total score for MMT lower extremity includes the following eight movements: right/left hip flexion, right/left hip abduction, right/left knee flexion and right/left knee extension. Total score ranges from 0 to 40, with lower scores indicating weaker muscle strength.

cipa+mig, cipaglucosidase alfa plus miglustat; CK, creatine kinase; FVC, forced vital capacity; Hex4, hexose tetrasaccharide; MMT, manual muscle test; 6MWD, 6-minute walk distance.

Figure 1. Change from PROPEL baseline in 6MWD, sitting FVC, MMT lower extremity, CK and Hex4 in the PROPEL Japan subpopulation. Total score for MMT lower extremity includes the following eight movements: right/left hip flexion, right/left hip abduction, right/left knee flexion and right/left knee extension. Score ranges from 0 to 40, with lower scores indicating weaker muscle strength. 6MWD, 6-minute walk distance; cipa+mig, cipaglucosidase alfa plus miglustat; CK, creatine kinase; FVC, forced vital capacity; Hex4, hexose tetrasaccharide; MMT, manual muscle test; OLE, open-label extension.

Table 5. Safety summary for the PROPEL/PROPEL OLE Japan subpopulation and ATB200-19.

Patients, n	PROPEL (weeks 0–52)		PROPEL+OLE (weeks 0–104)		ATB200-19
	Cipa+mig group (n=2)	Alg+pbo group (n=4)	Cipa+mig group (n=2)	Switch group (n=4)	Cipa+mig (n=4)
Any TEAE	2	4	2	4	4
TEAE leading to study drug discontinuation	0	0	0	1*	0
Treatment-related TEAE	0	2†	0	3‡	2§
Treatment-related TEAE leading to study drug discontinuation	0	0	0	1*	0
SAE	0	0	0	1*	4¶
SAE leading to study drug discontinuation	0	0	0	1*	0
Treatment-related SAE	0	0	0	1*	0
Treatment-related SAE leading to study drug discontinuation	0	0	0	1*	0
TEAE leading to death	0	0	0	0	0
IAR	0	1**	0	2††	1‡‡

Treatment-related included ‘definite’, ‘probable’ and ‘possibly’ related; not related included ‘unlikely’ and ‘unrelated’ as assessed by the investigator. All reported TEAEs are summarised in online supplemental table S2.

PROPEL DCO: 20 January 2021; PROPEL OLE DCO: 11 January 2022; ATB200-19 DCO: 31 July 2024.

^*Anaphylactic reaction in one patient.

^†Abdominal pain, constipation, nausea and upper abdominal pain in one patient, and dyspepsia, nausea and upper abdominal pain in one patient.

^‡Headache, urticaria and decreased oxygen saturation in one patient, and anaphylactic shock, dyspepsia and upper abdominal pain in one patient (including all events from PROPEL plus the PROPEL OLE).

^§Back pain, myalgia, musculoskeletal chest pain, pyrexia and urticaria in one patient, and diarrhoea in one patient.

^¶COVID-19 in three patients, and COVID-19 and pneumonia in one patient.

^**Dyspepsia, nausea and upper abdominal pain in one patient.

^††Urticaria in one patient, and anaphylactic reaction, dyspepsia and upper abdominal pain in one patient (including all events from PROPEL plus the PROPEL OLE).

^‡‡Back pain, myalgia, musculoskeletal chest pain, pyrexia and urticaria in one patient.

alg+pbo, alglucosidase alfa plus placebo; cipa+mig, cipaglucosidase alfa plus miglustat; DCO, data cut-off; IAR, infusion-associated reaction; OLE, open-label extension; SAE, serious adverse event; TEAE, treatment-emergent adverse event.

Table 2. Demographics and baseline characteristics for ATB200-19.

	Cipa+mig (n=4)
	Patient 7	Patient 8	Patient 9	Patient 10	Mean (SD)
Age at informed consent date, years*	21–30	21–30	31–40	41–50	33.0 (12.5)
Age at diagnosis, years	–	–	9	15	–
Time from diagnosis to start of study, years	22.1	21.4	31.9	32.3	–
Weight (kg)	28.5	48.9	55.2	42.9	43.9 (11.4)
ERT experienced at study entry	Yes	Yes	Yes	Yes	–
ERT duration prior to study, years	14.5	14.5	16.6	16.6	15.6 (1.2)
Use of ambulation-assistive devices at baseline	No	No	Yes	Yes	–
Invasive ventilation assistance at baseline	No	No	Yes	Yes	–
History of falls	–	–	–	–	–
History of IARs	Yes	Yes	Yes	No	–

^*Age ranges are shown for individual patients to protect their identity.

cipa+mig, cipaglucosidase alfa plus miglustat; ERT, enzyme replacement therapy; IAR, infusion-associated reaction.

Table 3. Baseline assessments for the PROPEL/PROPEL OLE Japan subpopulation.

	Cipa+mig group (n=2)			Switch group (n=4)
	Patient 1	Patient 2	Mean (SD)	Patient 3	Patient 4	Patient 5	Patient 6	Mean (SD)
Baseline 6MWD*
Metres	299.5	384.5	342.0 (60.1)	123.0	260.0	312.7	345.5	260.3 (98.1)
% predicted	39.7	48.7	44.2 (6.3)	17.1	50.2	58.0	46.6	43.0 (17.9)
Sitting FVC, % predicted*	56.0	79.5	67.8 (16.6)	80.5	38.0	53.0	58.5	57.5 (17.6)
MMT lower extremity total score†	30	24	27.0 (4.2)	18	27	40	17	25.5 (10.7)
Baseline serum CK (U/L)	1228	556	892.0 (475.2)	623	266	366	982	559.3 (319.4)
Baseline urine Hex4 (mmol/mol creatinine)	15.6	13.3	14.5 (1.6)	11.1	4.7	3.5	33.8	13.3 (14.1)

^*In PROPEL, baseline is the average of the last two values obtained on or prior to the first dose date.

^†Total score for MMT lower extremity includes the following eight movements: right/left hip flexion, right/left hip abduction, right/left knee flexion and right/left knee extension. Total score ranges from 0 to 40, with lower scores indicating weaker muscle strength.

6MWD, 6-minute walk distance; cipa+mig, cipaglucosidase alfa plus miglustat; CK, creatine kinase; FVC, forced vital capacity; Hex4, hexose tetrasaccharide; MMT, manual muscle test; OLE, open-label extension.

Patient 2 was aged 21–30 years and weighed 64.5 kg. The patient received alg for 12.3 years prior to PROPEL and received cipa+mig in PROPEL. In PROPEL and the OLE, the patient had a total of 25.3 months of treatment. 6MWD was 384.5 m (48.7% predicted) and sitting FVC was 79.5% predicted at baseline. 6MWD improved at week 38 (change from baseline: 167.5 m) and was maintained through to week 104, whereas % predicted sitting FVC initially worsened slightly to week 26 and then improved through to week 104. MMT lower extremity score was stable with some fluctuations around baseline, and the patient’s biomarker levels improved (decreased) from baseline to week 104. By the PROPEL OLE DCO, the patient had no TEAEs considered related to study treatment, TEAEs leading to study drug discontinuation or SAEs.

PROPEL OLE switch group

Patients 3–6 were randomised to alg+pbo in PROPEL and switched to cipa+mig in the OLE. Their weight ranged from 48.7 to 82.1 kg (mean (SD) 61.1 (15.7) kg), and their mean (SD) age was 43.0 (17.7) years. ERT duration prior to PROPEL ranged from 6.2 to 13.2 years (mean (SD) 10.3 (3.1) years). Cipa+mig treatment duration in the OLE ranged from 1.9 to 14.3 months (mean (SD) 11.0 (6.1) months; median 13.8).

From baseline to week 52 (following 52 weeks of treatment with alg+pbo), mean 6MWD was stable (mean (SD) change from baseline: 1.7 (30.2) m) with improvements in patient 3 and worsening in patient 6. Per cent predicted sitting FVC was variable (mean (SD) change from baseline: −3.8 (4.0) %), with stability in patient 4 and decreases in patients 3, 5 and 6. MMT lower extremity score fluctuated around baseline (mean (SD) change from baseline: 0.5 (0.6)). CK and Hex4 levels worsened (increased; mean (SD) change from baseline to week 52 was 1.3 (47.8) U/L for CK, and 6.1 (6.6) mmol/mol creatinine for Hex4). From baseline to week 104 (52 weeks of treatment with alg+pbo followed by 52 weeks of treatment with cipa+mig), mean 6MWD was generally stable (mean (SD) change from baseline: 10.1 (40.3) m); mean % predicted sitting FVC was also stable (mean (SD) change from baseline: −0.7 (1.0) %; data available for three patients). MMT lower extremity score fluctuated, with generally small changes (mean (SD) change from baseline: −0.7 (2.5)). CK and Hex4 levels improved (decreased) to week 104 after switching to cipa+mig at week 52 (mean (SD) change from baseline to week 104 was −332.0 (198.9) U/L for CK and −6.2 (6.1) mmol/mol creatinine for Hex4). By the PROPEL DCO (20 January 2021), all four patients experienced TEAEs, and two patients had treatment-related TEAEs, including an infusion-associated reaction (IAR) in one patient. By the OLE DCO, treatment-related TEAEs occurred in three patients, two patients had IARs and one patient discontinued study drug treatment due to a treatment-related SAE (anaphylactic reaction).

ATB200-19

Patient 7 weighed 28.5 kg, was aged 21–30 years and was using non-invasive ventilation support at bedtime. The patient received alg for 14.5 years prior to ATB200-19, and the duration of cipa+mig treatment in ATB200-19 was 30.4 months. 6MWD was 289.0 m, and sitting FVC was 14.5% predicted at baseline. The patient’s 6MWD fluctuated below baseline over time but was similar to baseline at week 52 (latest available assessment), and % predicted sitting FVC was generally maintained through to week 52. MMT lower extremity score was stable over 52 weeks (baseline score: 24; week 52 score: 26). CK and Hex4 levels improved (decreased) from baseline to the latest available assessment at week 130 (figure 2, online supplemental figure S2). The patient experienced five TEAEs considered related to cipa, all of which were considered mild, non-serious IARs and not related to mig. No action was taken for most IARs. Two episodes of urticaria required slowing of cipa infusion rate and antihistamine and corticosteroid treatment; another urticaria episode required antihistamine treatment. These IARs resolved on the day of onset, and subsequent cipa infusion was completed without further dose modification or safety issues. The patient experienced one SAE, and no TEAEs led to study drug discontinuation.

Figure 2. Change from baseline in 6MWD, sitting FVC, MMT lower extremity, CK and Hex4 in ATB200-19. Total score for MMT lower extremity includes the following eight movements: right/left hip flexion, right/left hip abduction, right/left knee flexion and right/left knee extension. Score ranges from 0 to 40, with lower scores indicating weaker muscle strength. The 52-week data for patient 10 for CK and Hex4 are from an end-of-study/early termination visit on study day 281. 6MWD, 6-minute walk distance; CK, creatine kinase; FVC, forced vital capacity; Hex4, hexose tetrasaccharide; MMT, manual muscle test.

Patient 8 was aged 21–30 years and weighed 48.9 kg. The patient had 14.5 years of prior alg treatment before ATB200-19 and was treated with cipa+mig for 30.4 months in ATB200-19. 6MWD was 566.0 m, and sitting FVC was 84.5% predicted at baseline. 6MWD fluctuated below baseline over time but was similar to baseline at the latest available assessment at week 52, while % predicted sitting FVC remained stable through to week 104. MMT lower extremity score was also stable over 52 weeks. CK and Hex4 levels improved (decreased) from baseline to week 104. The patient experienced no treatment-related TEAEs or TEAEs leading to study drug discontinuation. The patient had one SAE.

Patient 9 was aged 31–40 years, weighed 55.2 kg, had a tracheostomy, required 24-hour controlled ventilation and was wheelchair dependent. The patient received alg for 16.6 years prior to ATB200-19. Cipa+mig treatment duration in ATB200-19 lasted 23.5 months. Efficacy assessments in this patient were not feasible due to advanced disease status. Their CK and Hex4 levels improved (decreased) from baseline to week 104. The patient experienced no treatment-related TEAEs or TEAEs leading to study drug discontinuation; two SAEs were recorded.

Patient 10 weighed 42.9 kg, was aged 41–50 years, had a tracheostomy, required 24-hour controlled ventilation and was wheelchair dependent. Alg was received for 16.6 years prior to ATB200-19, and the duration of cipa+mig treatment in ATB200-19 was 8.8 months. As with patient 9, their advanced disease status did not allow efficacy assessments. CK levels improved (decreased) slightly from baseline to the latest available assessment (week 52). Hex4 levels were variable but improved by week 52. The patient withdrew consent and discontinued the study after 17 treatments. The patient experienced one TEAE considered related to mig (diarrhoea), but no TEAEs led to study drug discontinuation. One SAE was documented.

Discussion

This case series provides an overview of the efficacy and safety of cipa+mig in 10 ERT-experienced Japanese patients with LOPD, including six patients who participated in the phase 3 PROPEL study and PROPEL OLE, and four patients who received cipa+mig in the ATB200-19 expanded access study.

While the number of patients in the PROPEL Japan subpopulation was small and individual cases showed variability, the efficacy and safety of cipa+mig were generally consistent with the results from the global population. Baseline characteristics of the PROPEL Japan subpopulation were generally consistent with the global study population,¹⁷ with some differences: patients in the Japan subpopulation were on average younger than those in the global overall PROPEL population (39.5 vs 46.8 years), were younger at diagnosis (26.8 vs 39.0 years) and had received prior ERT for longer (10.9 vs 7.4 years).¹⁷ In the overall PROPEL population, patients receiving cipa+mig improved from baseline to week 52 in 6MWD (mean (SE) change 20.8 (4.6) m), CK (mean (SE) change −130.5 (25.1) U/L) and Hex4 (mean (SE) change −1.9 (0.3) mmol/mol creatinine), with stability versus baseline in % predicted sitting FVC (mean (SE) change −0.9 (0.7) %).¹⁷ Similar results were seen in the global population of ERT-experienced patients receiving cipa+mig in PROPEL and were maintained in the OLE.^{17 19} Of the two ERT-experienced Japanese patients receiving cipa+mig in PROPEL and the OLE, one (patient 2) showed substantial improvement from baseline in 6MWD by week 38 (167.5 m), which was maintained to week 104, while the other (patient 1) showed stability in 6MWD to week 104. Both patients showed improvement from baseline to week 104 in sitting FVC and sustained improvement from baseline in CK and Hex4 to week 104.

In the overall global group of patients who received alg+pbo in PROPEL and switched to cipa+mig in the OLE (switch group), improvement from baseline to week 52 was seen in 6MWD (mean (SE) change 7.2 (6.6) m), and worsening in % predicted sitting FVC (mean (SE) change −4.0 (0.8) %), CK (mean (SE) change 60.2 (26.2) U/L) and Hex4 (mean (SE) change 1.2 (0.7) mmol/mol creatinine).¹⁷ For ERT-experienced patients in this group, 6MWD was stable versus baseline (mean (SE) change 0.0 (7.2) m) after 52 weeks of alg+pbo treatment, while % predicted sitting FVC (mean (SE) change −4.0 (0.9) %), CK (mean (SE) change 79.6 (26.9) U/L) and Hex4 (mean (SE) change 1.9 (0.8) mmol/mol creatinine) worsened.¹⁷ After switching to cipa+mig, 6MWD and FVC outcomes were generally maintained to week 104 in the OLE. CK and Hex4 levels markedly improved after switching to cipa+mig, and biomarker levels at week 104 were similar to those of patients who had received cipa+mig throughout PROPEL and the OLE.¹⁹ We note that biomarker levels tended to improve, while functional assessments (6MWD, FVC, MMT) largely showed stability. These differing patterns are not unexpected, as biomarkers are generally considered to reflect underlying disease activity, whereas functional outcomes typically change more slowly and can be associated with greater variability, particularly in smaller cohorts.^{21 22} Furthermore, in the context of a progressive disease, stabilisation of functional measures is often regarded as a clinically meaningful outcome for a treatment where further worsening is prevented.²³ Biomarker, motor function, pulmonary function and muscle function assessments each provide complementary but distinct information and should be considered together to provide a more holistic view of disease status and treatment response. Thus, the observed biomarker and functional results are broadly consistent with current understanding of Pompe disease monitoring. In the switch group of the Japanese subpopulation, changes in 6MWD, FVC and biomarker levels were in keeping with findings from ERT-experienced patients in the global population; however, a sustained improvement from baseline in 6MWD was seen in one patient (patient 3), both during alg+pbo treatment in PROPEL and after switching to cipa+mig in the OLE. In patients with declining sitting FVC during alg+pbo treatment and available data at week 104, sitting FVC declined in three of four patients to week 52 during alg+pbo treatment but increased to near-baseline levels by week 104 after switching to cipa+mig. One patient (patient 4) with a low baseline sitting FVC (38%) remained stable by week 104 (change from baseline of −1%), suggesting that cipa+mig can maintain respiratory function even in patients with a low baseline sitting FVC.

All patients in the Japan subpopulation experienced TEAEs in PROPEL and the OLE, which is similar to the PROPEL global overall population, in which nearly all patients experienced TEAEs.^{17 19} All TEAEs in the Japan subpopulation were mild or moderate in intensity. One patient (patient 5) who had received alg+pbo in PROPEL discontinued treatment in the OLE due to an anaphylactic reaction. This patient had a history of IARs and of multiple allergies, including food allergies, allergic reactions to bee stings and allergic skin reactions, which were considered relevant to the anaphylaxis event. In the Japan subpopulation in PROPEL and the OLE, the most frequent adverse events were arthralgia, falls and nausea, occurring in three patients each regardless of treatment group. These were also among the most frequent TEAEs reported for the overall population (falls: 40/123 patients; arthralgia and nausea: 18/123 patients).¹⁷

Compared with PROPEL, ATB200-19 enrolled a broader range of patients; eligibility criteria permitted the inclusion of non-ambulatory patients, those requiring 24-hour ventilation support, or those with a body weight <40 kg, which are all indicative of advanced disease in LOPD. Efficacy evaluations were not mandatory to allow for patients who found it difficult to complete the assessments. In keeping with these criteria, three of the four patients in ATB200-19 would not have met the PROPEL eligibility criteria: patient 7 had a body weight of <40 kg and a sitting FVC of <30%, and patients 9 and 10 were non-ambulatory, received 24-hour controlled ventilation support, and therefore could not complete efficacy assessments. Patients in ATB200-19 also had a longer mean duration of prior ERT with alg than patients in the global PROPEL OLE ERT-experienced population (15.6 vs 7.3 years).¹⁹ Outcomes in patients who received cipa+mig in ATB200-19 support that cipa+mig was effective in patients with LOPD who had more advanced disease, lower body weight and longer ERT history than those in PROPEL. Although there was variability across assessment time points in the two evaluable patients, 6MWD, sitting FVC and MMT lower extremity score were generally maintained throughout the assessment period. The biomarkers CK and Hex4 showed improvement in all four patients. Similar to the findings from PROPEL and the OLE, the variability in biomarker and functional results in ATB200-19 was not unexpected, given that functional outcomes tend to change more slowly and can be associated with greater variability than biomarkers, especially in smaller cohorts and in the context of a progressive disease where stability is considered a meaningful treatment outcome.^21–23 There were no new safety concerns unique to patients in ATB200-19. The small sample size of the PROPEL Japan subpopulation and ATB200-19 limits the conclusions that can be drawn.

The most frequent genetic alteration of the GAA gene in patients with LOPD in Japan (c.546G>T)²⁴ was also most common in the patients presented here (4/10 patients, all in the PROPEL study), followed by c.1857C>G in three patients (one in PROPEL and two in ATB200-19). The c.546G>T alteration has only been reported in Japanese or Korean patients^{25 26} and appears to be associated with a low age of symptom onset,²⁷ which may be reflected in the younger age of diagnosis in Japanese patients in PROPEL compared with the global PROPEL population. While patient numbers were low, there was no clear indication that the response to cipa+mig was influenced by specific GAA gene variants. The c.670C>T alteration observed in two patients in ATB200-19 appears to be mainly associated with infantile- or childhood-onset Pompe disease.^{25 26} The alterations in patients in this case series differ from the alterations seen most commonly in patients with LOPD in Western^{28 29} and other Asian countries.³⁰ Overall, consistency between findings from the global PROPEL population and the Japan subpopulation suggests that, despite known differences in genetic background, cipa+mig demonstrates efficacy in Japanese people with LOPD.

In summary, although the number of patients is small, the efficacy and safety of cipa+mig in Japanese patients from PROPEL, its OLE and ATB200-19, including those with advanced disease, low body weight and prolonged prior ERT, were indicative of sustained and durable responses to cipa+mig treatment and generally consistent with findings from the overall population in the global PROPEL trial and its OLE.

Data availability statement

Data are available upon reasonable request.