In this study, we analyzed alpha-glucosidase activity in plasma and leukocytes following an rhGAA infusion at the prescribed dose of 40 mg/kg. We focused on the Cmax, the half-life, and the remaining enzymatic activity before the next infusion, to evaluate when enzyme activity dropped below the control level and within the untreated patient range after ERT, in order to identify a potential window to evaluate endogenous alpha-glucosidase production after gene therapy in future clinical studies.

First, we performed five PK curves and analyzed the plasmatic levels of alpha-glucosidase before, during, and after the infusion. At the end of the infusion, we observed a very high plasmatic peak, with the median Cmax exceeding the upper level of normal subjects by > 5000 times and of patients by > 100,000 times. The Cmax obtained in this study was not directly comparable to that reported by the Summary of Product Characteristics, which reports protein concentration (in microgram/milliliter) [19] instead of alpha-glucosidase activity. Elimination from circulation was rapid, with a median half-life of 3.1 h, similar to the mean half-life of 2.75 h reported in the Summary of Product Characteristics for children receiving 40 mg/kg of alpha-glucosidase [19]. Of the five PK curves, two were performed in the presence of high antibody titers: patient 2 showed the lowest Cmax, while in patient 5 Cmax occurred later because of a slower infusion. To fully understand the effect of antibodies on enzyme activity, it should be measured in plasma with the addition of artificial beads of protein A, to quantify the amount of antibody-bound enzyme [15, 17].

Subsequently, it was investigated when enzymatic activity after ERT returned below the control level and within the untreated patient range. Enzyme activity declined over time to stabilize at a plateau between the control and patient level; no enzyme activity was detected within the untreated patient level, even in a single observation of a child with classic infantile Pompe disease who did not receive ERT for 20 days. The Summary of Product Characteristics of alglucosidase alpha does not report PK data of patients receiving weekly infusions. However, in patients with late-onset Pompe disease treated with 20 mg/kg every other week, Cmax remained stable at weeks 0, 12, and 52, not revealing an accumulation over time [20]. Pharmacokinetics may differ with weekly ERT administration, which is the standard for patients with classic infantile Pompe disease. Persistence of alpha-glucosidase activity, above the patient range, was also described 48 h after administration of neo-rhGAA in patients with late-onset Pompe disease treated with 5, 10, or 20 mg/kg of avalglucosidase alpha every other week [21].

The source of the remaining enzyme activity between the control and untreated patient range is still not clear. Earlier studies in mice have shown that, after ERT administration, alpha-glucosidase activity increases by > 100 times in the liver and spleen in comparison to wild-type [22,23,24]. Furthermore, a persistence of high enzymatic activity in the liver and spleen was observed up to 15 days post-ERT administration in mice (6.5 and 26.5% of endogenous activity in the liver and spleen, in comparison to 1.1% in the muscle) [22]. Enzyme activity has not yet been studied in the liver and spleen after ERT infusion in humans. However, we can speculate that the persistence of alpha-glucosidase activity described in this study, mostly between the normal and disease ranges, could be related to the large quantities of ERT taken up by liver, spleen, and the reticulo-endothelial system [25, 26]. These organs may act as reservoirs, slowly releasing the enzyme back into the circulation. Currently, the exact biological path of the remaining enzyme remains unclear. Lysosomal exocytosis is increasingly recognized as a vital cellular process and occurs in all cell types; the remaining enzyme activity we measured in plasma might be excreted from the lysosomes [27,28,29], but this requires further studies.

We then analyzed alpha-glucosidase activity in leukocytes after an rhGAA infusion. Leukocytes display mannose-6-phospate receptors on the cellular membrane, which is the gate through which rhGAA enters the cells [30]. Previous studies have demonstrated that untreated patients exhibit periodic acid-Schiff (PAS)-positive lymphocyte vacuoles, with PAS staining used to highlight glycogen accumulation [31]. Within a few weeks of starting ERT, the percentage of PAS positive vacuoles and PAS score greatly decreased in all patients [31], indicating that ERT entered the cell. Given their accessibility, leukocytes have been used as a model to study the cellular uptake and clearance of alpha-glucosidase, as well as the intra-lysosomal half-life. Similar studies have been performed for other lysosomal storage diseases [32, 33].

After an rhGAA infusion, the peak of alpha-glucosidase activity observed in leukocytes was quite low. In particular, the median Cmax was 18 (Gn)/4 (MU) times higher than the upper limit of the untreated patient range and did not exceed control levels, i.e., 0.7 (Gn) and 0.9 times (MU) the upper limit of normal values. The peak concentration after ERT was quite modest, especially when compared with the extremely high peaks in plasma, which were > 5000 times higher than the upper level controls and > 100,000 times higher than untreated patient levels. Yet, previous studies have shown that this level is sufficient to reduce glycogen accumulation in lymphocytes within a few weeks of starting ERT [31]. This suggests that even a modest increase in enzyme activity can have therapeutic effects.

However, muscle tissue differs fundamentally. Unlike leukocytes, which are directly exposed to circulating alpha-glucosidase, muscle cells are shielded by many physiological barriers, such as the endothelial barrier and the endomysium [34]. This means that while leukocyte enzyme activity may provide insight into enzyme uptake dynamics, it does not fully reflect the challenges of enzyme delivery to muscle tissue. If leukocytes, despite their direct exposure to circulating enzyme, only show a modest increase in activity, this raises concerns about whether a sufficient amount of enzyme reaches muscle cells.

The modest Cmax might be explained by the low content of mannose-6-phosphate of rhGAA. To overcome this, second-generation ERT such as avalglucosidase alpha was developed with an increased number of mannose-6-phosphate groups to optimize cellular uptake [21]. We therefore speculate that the low Cmax in leukocytes, despite their direct exposure to very high plasma concentrations of rhGAA, may be due to the molecule’s insufficient phosphorylation.

The peak concentration in leukocytes was observed about 24 h from the start of the infusion, later in comparison to plasma Cmax (at the end of the infusion). Furthermore, the half-life was 2–4 days, similar to early findings in mice after the injection of bovine testes-derived alpha-glucosidase [22] and fibroblasts of patients with Pompe disease cultured with bovine testes alpha-glucosidase [35], which is also in net contrast to the short half-life in plasma. Two patients received long-term rituximab, which depletes CD20-positive cells. As alpha-glucosidase activity is measured in leukocytes (neutrophils, lymphocytes, monocytes) and overall counts remained within the normal range, we do not expect this to have altered enzyme activity.

This study aimed to identify a potential timepoint to assess a surrogate efficacy marker of gene therapy, such as alpha-glucosidase activity in plasma and leukocytes. In patients with classic infantile Pompe disease, ERT must be continued before, during, and after gene therapy, making it essential to determine a specific moment when enzyme activity can be evaluated without the influence of ERT. In the presence of enzyme activity derived from gene therapy, such findings must be integrated with age-appropriate clinical parameters to consider safe discontinuation of ERT. We found that, in plasma, at day 7, 70% of samples expressed levels above the control range, at day 9, all enzyme activities were within or below the control level, and from day 11, 70% of samples were below the lower limit of control. Enzyme activity in plasma did not return to the untreated patient range, even after 20 days from ERT in one patient. In leukocytes, at day 7, 100% of samples were within or below control levels (Gn and MU); in the glycogen assay, from day 9, all samples were below the lower limit of control and from day 14 within the patient range. In the MU assay, from day 11, 70% of samples were below the lower limit of control and from day 14 onwards within the patient range. Enzyme activity measurement in leukocytes with the glycogen assay was the most consistent, with all samples being below the control range from day 9. These findings suggest alpha-glucosidase activity in plasma and leukocytes may serve as a surrogate marker for future gene therapy studies, even for patients with classic infantile Pompe disease on ERT. In fact, after 14 days from ERT, enzyme activities in plasma and leukocytes decreased to a level that any detectable surplus of enzyme activity derived from gene therapy could be detected and distinguished from ERT. For example, in the mouse model, lentiviral gene therapy with an insulin-like growth factor 2 tag added to a codon-optimized version of GAA (LV-IGF2.GAAco) led to a > 120 times increase of alpha-glucosidase activity in leukocytes in comparison to the knock-out mouse and a > 30 times increase in comparison to wild-type [11]. Future studies will assess whether similar findings of this study hold for second-generation ERTs.

A limitation of this study is represented by the limited amount of data available after 9 days from ERT; given the variability in enzyme activity at earlier timepoints, more samples are necessary to confirm the trend we observed. In classic infantile Pompe disease, the standard of care of our center consists of weekly infusions; therefore, it is not frequently possible to obtain such data; so said, we will continue collecting data.