Background: Pompe disease is caused by a deficiency of the enzyme acid alpha-glucosidase (GAA). People with infantile-onset disease have either a complete or a near-complete enzyme deficiency; people with late-onset Pompe disease (LOPD) retain some residual enzyme activity. GAA deficiency is treated with an intravenous infusion of recombinant human acid alglucosidase alfa, an enzyme replacement therapy (ERT). Alglucosidase alfa and avalglucosidase alfa are approved treatments, but cipaglucosidase alfa with miglustat is not yet approved., Objectives: To assess the effects of enzyme replacement therapies in people with late-onset Pompe disease., Search Methods: We searched the Cochrane Inborn Errors of Metabolism Trials Register, compiled from electronic database searches and handsearching of journals and conference abstract books. We also searched MEDLINE OvidSP, clinical trial registries, and the reference lists of relevant articles and reviews. Date of last search: 21 April 2022., Selection Criteria: We included randomised controlled trials (RCTs) of ERT in people with LOPD of any age., Data Collection and Analysis: Two review authors independently assessed trial eligibility, extracted data, assessed the risk of bias and the certainty of the evidence (using GRADE). We resolved disagreements through discussion and by consulting a third author., Main Results: We included six trials (358 randomised participants) lasting from 12 to 78 weeks. A single trial reported on each comparison listed below. None of the included trials assessed two of our secondary outcomes: need for respiratory support and use of a walking aid or wheelchair. Certainty of evidence was most commonly downgraded for selective reporting bias. Alglucosidase alfa versus placebo (90 participants) After 78 weeks, alglucosidase alfa probably improves the six-minute walk test (6MWT) distance compared to placebo (mean difference (MD) 30.95 metres, 95% confidence interval (CI) 7.98 to 53.92; moderate-certainty evidence) and probably improves respiratory function, measured as the change in per cent (%) predicted forced vital capacity (FVC) (MD 3.55, 95% CI 1.46 to 5.64; moderate-certainty evidence). There may be little or no difference between the groups in occurrence of infusion reactions (risk ratio (RR) 1.21, 95% CI 0.57 to 2.61; low-certainty evidence), quality of life physical component score (MD -1.36 points, 95% CI -5.59 to 2.87; low-certainty evidence), or adverse events (RR 0.94, 95% CI 0.64 to 1.39; low-certainty evidence). Alglucosidase alfa plus clenbuterol versus alglucosidase alfa plus placebo (13 participants) The evidence is very uncertain about the effect of alglucosidase alfa plus clenbuterol compared to alglucosidase alfa plus placebo on: change in 6MWT distance after 52 weeks (MD 34.55 metres, 95% CI-10.11 to 79.21; very low-certainty evidence) and change in % predicted FVC (MD -13.51%, 95% CI -32.44 to 5.41; very low-certainty evidence). This study did not measure infusion reactions, quality of life, and adverse events. Alglucosidase alfa plus albuterol versus alglucosidase alfa plus placebo (13 participants) The evidence is very uncertain about the effect of alglucosidase alfa plus albuterol compared to alglucosidase alfa plus placebo on: change in 6MWT distance after 52 weeks (MD 30.00 metres, 95% CI 0.55 to 59.45; very low-certainty evidence), change in % predicted FVC (MD -4.30%, 95% CI -14.87 to 6.27; very low-certainty evidence), and risk of adverse events (RR 0.67, 95% CI 0.38 to 1.18; very low-certainty evidence). This study did not measure infusion reactions and quality of life. VAL-1221 versus alglucosidase alfa (12 participants) Insufficient information was available about this trial to generate effect estimates measured at one year or later. Compared to alglucosidase alfa, VAL-1221 may increase or reduce infusion-associated reactions at three months, but the evidence is very uncertain (RR 2.80, 95% CI 0.18 to 42.80). This study did not measure quality of life and adverse events. Cipaglucosidase alfa plus miglustat versus alglucosidase alfa plus placebo (125 participants) Compared to alglucosidase alfa plus placebo, cipaglucosidase alfa plus miglustat may make little or no difference to: 6MWT distance at 52 weeks (MD 13.60 metres, 95% CI -2.26 to 29.46); infusion reactions (RR 0.94, 95% CI 0.49 to 1.80); quality of life scores for physical function (MD 1.70, 95% CI -2.13 to 5.53) and fatigue (MD -0.30, 95% CI -2.76 to 2.16); and adverse effects potentially related to treatment (RR 0.83, 95% CI 0.49 to 1.40) (all low-certainty evidence). Cipaglucosidase alfa plus miglustat probably improves % predicted FVC compared to alglucosidase alfa plus placebo (MD 3.10%, 95% CI 1.04 to 5.16; moderate-certainty evidence); however, it may make little or no change in % predicted sniff nasal inspiratory pressure (MD -0.06%, 95% CI -8.91 to 7.71; low-certainty evidence). Avalglucosidase alfa versus alglucosidase alfa (100 participants) After 49 weeks, avalglucosidase alfa probably improves 6MWT compared to alglucosidase alfa (MD 30.02 metres, 95% CI 1.84 to 58.20; moderate-certainty evidence). Avalglucosidase alfa probably makes little or no difference to % predicted FVC compared to alglucosidase alfa (MD 2.43%, 95% CI -0.08 to 4.94; moderate-certainty evidence). Avalglucosidase alfa may make little or no difference to infusion reactions (RR 0.78, 95% CI 0.42 to 1.45), quality of life (MD 0.77, 95% CI -2.09 to 3.63), or treatment-related adverse events (RR 0.92, 95% CI 0.61 to 1.40), all low-certainty evidence., Authors' Conclusions: One trial compared the effect of ERT to placebo in LOPD, showing that alglucosidase alfa probably improves 6MWT and respiratory function (both moderate-certainty evidence). Avalglucosidase alfa probably improves 6MWT compared with alglucosidase alfa (moderate-certainty evidence). Cipaglucosidase plus miglustat probably improves FVC compared to alglucosidase alfa plus placebo (moderate-certainty evidence). Other trials studied the adjunct effect of clenbuterol and albuterol along with alglucosidase alfa, with little to no evidence of benefit. No significant rise in adverse events was noted with all ERTs. The impact of ERT on some outcomes remains unclear, and longer RCTs are needed to generate relevant information due to the progressive nature of LOPD. Alternative resources, such as post-marketing registries, could capture some of this information., (Copyright © 2023 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.)