NZ786386A - Formulations comprising recombinant acid alpha-glucosidase - Google Patents

Abstract

Provided are pharmaceutical formulations comprising a recombinant acid a-glucosidase, wherein the recombinant acid a-glucosidase is expressed in Chinese hamster ovary (CHO) cells and comprises an increased content of N-glycan units bearing one or two mannose-6-phosphate residues when compared to a content of N-glycan units bearing one or two mannose-6-phosphate residues of alglucosidase alfa; at least one buffer selected from the group consisting of a citrate, a phosphate and combinations thereof; and at least one excipient selected from the group consisting of mannitol, polysorbate 80, and combinations thereof, wherein the formulation has a pH of from about 5.0 to about 7.0. Also provided are methods of treating Pompe disease using these pharmaceutical formulations.

Description

FORMULATIONS COMPRISING RECOMBINANT ACID ALPHA—GLUCOSIDASE TECHNICAL FIELD Principles and embodiments of the t invention relate generally to formulations comprising recombinant acid a-Glucosidase, and particularly liquid formulations.

BACKGROUND Pompe disease, also known as acid maltase deficiency or glycogen storage disease type II, is one of several lysosomal storage disorders. Lysosomal storage disorders are a group of autosomal recessive genetic diseases characterized by the accumulation of cellular glycosphingolipids, en, or mucopolysaccharides within ellular compartments called lysosomes. duals with these diseases carry mutant genes coding for enzymes which are defective in catalyzing the hydrolysis of one or more of these substances, which then build up in the lysosomes. Other examples of lysosomal disorders include Gaucher disease, GMl-gangliosidosis, fucosidosis, mucopolysaccharidoses, -Scheie disease, Niemann- Pick A and B diseases, and Fabry disease. Pompe disease is also classified as a neuromuscular disease or a metabolic myopathy.

Pompe disease is estimated to occur in about 1 in 40,000 births, and is caused by a mutation in the GAA gene, which codes for the enzyme lysosomal a-glucosidase (EC:3.2.1.20), also commonly known as acid a-glucosidase. Acid (x-glucosidase is involved in the metabolism of glycogen, a branched polysaccharide which is the major storage form of glucose in animals, by catalyzing its ysis into e within the lysosomes. Because individuals with Pompe disease produce , defective acid a-glucosidase which is inactive or has reduced activity, glycogen own occurs slowly or not at all, and glycogen accumulates in the lysosomes of various tissues, particularly in striated muscles, leading to a broad spectrum of clinical manifestations, including progressive muscle weakness and respiratory insufficiency. Tissues such as the heart and skeletal muscles are particularly affected.

Pompe disease can vary widely in the degree of enzyme deficiency, severity and age of onset, and over 500 ent ons in the GAA gene have been identified, many of which cause disease symptoms of g severity. The disease has been classified into broad types: early onset or infantile and late onset. r onset of e and lower enzymatic activity are generally associated with a more severe clinical course. Infantile Pompe disease is the most severe, ing from te or near complete acid osidase deficiency, and presents with symptoms that include severe lack of muscle tone, weakness, enlarged liver and heart, and cardiomyopathy. The tongue may become enlarged and protrude, and wing may become difficult. Most affected children die from respiratory or cardiac complications before the age of two. Late onset Pompe disease can present at any age older than 12 months and is characterized by a lack of cardiac involvement and better short-term prognosis.

Symptoms are related to ssive skeletal muscle dysfunction, and involve generalized muscle weakness and wasting of respiratory muscles in the trunk, proximal lower limbs, and diaphragm. Some adult patients are devoid of major symptoms or motor limitations. Prognosis generally depends on the extent of respiratory muscle involvement. Most subjects with Pompe e eventually progress to physical debilitation requiring the use of a wheelchair and ed ventilation, with premature death often ing due to respiratory failure.

Recent treatment options for Pompe disease e enzyme replacement therapy (ERT) with recombinant human acid a—glucosidase (rhGAA). Conventional rhGAA products are known under the names alglucosidase alfa, Myozyme® or Lumizyme® from Genzyme, Inc. ERT is a chronic treatment required throughout the lifetime of the patient, and involves administering the replacement enzyme by intravenous infusion. The replacement enzyme is then transported in the circulation and enters lysosomes within cells, where it acts to break down the lated glycogen, compensating for the deficient activity of the endogenous defective mutant enzyme, and thus relieving the e symptoms.

The way in which replacement enzymes, such as rhGAA, are prepared, , orted and administered to patients is difficult. The enzymes used in ERT are generally relatively complex and delicate, making selection of accompanying buffers, excipients, etc. critical. If the enzyme is not preserved properly, then high ties may be required, making treatment costly and inefficient.

Some conventional rhGAA products are provided to patients as a lyophilized (freeze—dried) powder in single—use vials without preservatives. The rhGAA must then be reconstituted in the vials, then diluted and administered intravenously. While lyophilization helps to preserve the enzyme after manufacture until it is ready to be administered to a patient, this process in and of itself can damage enzyme. Thus, great care must be taken in selection of WO 73060 the components in the rhGAA formulation so that they help preserve protein concentration and activity.

Furthermore, recombinant s are often structurally different from wild- type enzymes. Even if the amino acids in the recombinant enzyme may be identical to its wild- type counterpart, there may be ences in the carbohydrate chemistry. Thus, as new recombinant enzymes are discovered, the formulations for the enzymes must be developed ic the chemistry of the newly discovered enzymes.

Accordingly, there is an ongoing need for ations to store and transport recombinant enzymes, such as rhGAA, which preserve enzyme activity and concentration.

SUMMARY One aspect of the invention pertains to a pharmaceutical formulation. In embodiment one, the formulation comprises: (a) a recombinant acid a-glucosidase, wherein the recombinant acid a-glucosidase is expressed in Chinese hamster ovary (CHO) cells and comprises an increased content of N-glycan units g one or two mannose-6—phosphate residues when ed to a content of N-glycan units bearing one or two mannose phosphate residues of alglucosidase alfa; (b) at least one buffer selected from the group consisting of a e, a phosphate and combinations thereof; and (c) at least one excipient selected from the group consisting of mannitol, polysorbate 80, and combinations f, wherein the formulation has a pH of from about 5.0 to about 7.0.

Embodiment two includes a modification to the pharmaceutical formulation of embodiment one, wherein the inant acid (x-glucosidase is present in a concentration of about 5 to about 50 mg/mL.

Embodiment three includes a modification to the pharmaceutical formulation of embodiment one or two, wherein the recombinant acid a-glucosidase is present in a concentration of about 15 mg/mL. id="p-13" id="p-13"

[0013] Embodiment four includes a modification to the pharmaceutical formulation of any of embodiments 1-3, wherein the formulation has a pH of from about 5.5 to about 7.0.

Embodiment five includes a modification to the pharmaceutical ation of any of embodiments 1-4, wherein the formulation has a pH of about 6.0. ment six includes a cation to the pharmaceutical formulation of any of embodiments 1-5, wherein the at least one buffer comprises citrate.

Embodiment seven es a modification to the pharmaceutical formulation of any of embodiments 1-6, wherein the at least one buffer comprises a potassium, sodium or ammonium salt.

Embodiment eight includes a modification to the pharmaceutical formulation of any of embodiments 1-7, wherein the at least one buffer comprises sodium citrate. id="p-18" id="p-18"

[0018] Embodiment nine includes a modification to the pharmaceutical formulation of any of embodiments 1—8, wherein the at least one buffer is present in a concentration of about to about 100 mM.

Embodiment ten includes a modification to the ceutical formulation of any of embodiments 1—9, n the at least one buffer is present in a concentration of about 25 mM.

Embodiment 11 includes a modification to the pharmaceutical formulation of any of ments 1-10, wherein trehalose, sucrose, glycine or combinations thereof is excluded.

Embodiment 12 includes a modification to the pharmaceutical formulation of any of embodiments 1-11, n the at least one excipient is mannitol t in a concentration of about 10 to about 50 mg/mL. ment 13 includes a modification to the pharmaceutical formulation of any of embodiments 1—12, wherein the at least one excipient is polysorbate 80 present in a concentration of about 0.2 to about 0.5 mg/mL. id="p-23" id="p-23"

[0023] Embodiment 14 includes a modification to the pharmaceutical formulation of any of ments 1-13, wherein the mannitol is present at a concentration of about 20 mg/mL and the polysorbate 80 is present at a concentration of about 0.5 mg/mL.

Embodiment 15 includes a modification to the pharmaceutical formulation of any of embodiments 1-14, further comprising: (d) water. id="p-25" id="p-25"

[0025] Embodiment 16 includes a modification to the pharmaceutical formulation of any of embodiments 1—15, further comprising: (d) an alkalizing agent; and/or (e) an acidifying agent, wherein the zing agent and acidifying agent are present in amounts to in the pharmaceutical formulation at a pH of from about 5.0 to about 6.0.

Embodiment 17 includes a modification to the pharmaceutical formulation of any of embodiments 1-16, wherein at least 30% of molecules of the recombinant human acid a—glucosidase comprise one or more N-glycan units bearing one or two mannose—6—phosphate residues.

Embodiment 18 includes a modification to the pharmaceutical formulation of any of embodiments l-l7, n the recombinant human acid (x-glucosidase comprises on average from 0.5 to 7.0 moles of N-glycan units bearing one or two mannosephosphate residues per mole of recombinant human acid a—glucosidase.

Embodiment 19 includes a modification to the pharmaceutical formulation of any of embodiments 1-18, wherein the recombinant human acid (x-glucosidase comprises on average at least 3 moles of mannosephosphate residues per mole of recombinant human acid a—glucosidase and at least 4 moles of sialic acid residues per mole of recombinant human acid osidase.

Embodiment 20 includes a modification to the pharmaceutical formulation of any of embodiments 1-19, wherein the recombinant human acid (x-glucosidase comprises seven ial N—glycosylation sites, at least 50% of molecules of the recombinant human acid 0t- glucosidase comprise an N-glycan unit bearing two mannose—6-phosphate residues at the first site, at least 30% of molecules of the recombinant human acid (z-glucosidase comprise an N- glycan unit bearing one e-6—phosphate e at the second site, at least 30% of les of the inant human acid a-glucosidase comprise an N-glycan unit bearing two e-é-phosphate residue at the fourth site, and at least 20% of molecules of the recombinant human acid a-glucosidase comprise an N—glycan unit bearing one mannose-6— phosphate residue at the fourth site.

Embodiment 21 includes a modification to the pharmaceutical formulation of any of embodiments 1-20, wherein the ceutical formulation consists essentially of: (a) the recombinant acid a—glucosidase; (bl) sodium citrate; (b2) citric acid monohydrate; (cl) mannitol; (c2) polysorbate 80; (d) water; (e) optionally, an acidifying agent; and (f) optionally, an alkalizing agent, n the formulation has a pH of from about 5.0 to about 6.0.

Embodiment 22 includes a modification to the pharmaceutical formulation of any of embodiments 1-2], wherein the pharmaceutical formulation consists essentially of: (a) the recombinant acid (x-glucosidase, present at a concentration of about 15 mg/mL; (b) sodium citrate , present at a concentration of about 25 mM; (cl) mannitol, present at a concentration of about 20 mg/mL; (c2) polysorbate 80, t at a concentration of about 0.5 mg/mL; and (d) water; (e) optionally, an acidifying agent; and (f) optionally, an alkalizing agent, wherein the ation has a pH of from about 5.0 to about 6.0.

Another aspect of the invention pertains to a lyophilized ceutical composition. Accordingly, embodiment 23 pertains to a ceutical composition comprising the formulation of any of embodiments 1-22 after lyophilization. Embodiment 24 pertains to a pharmaceutical composition comprising a lized mixture comprising: (a) a recombinant acid a-glucosidase, wherein the recombinant acid a-glucosidase is expressed in e hamster ovary (CHO) cells and comprises an sed content of N-glycan units bearing one or two mannosephosphate residues when compared to a content of N-glycan units bearing one or two mannose phosphate residues of alglucosidase alfa; (b) a buffer selected from the group consisting of a citrate, a phosphate and combinations thereof; and (c) at least one excipient selected from the group consisting of trehalose, mannitol, polysorbate 80, and combinations f.

Another aspect of the invention pertains to a method of treating Pompe disease.

Accordingly, embodiment 25 comprises administering to a patient in need thereof the pharmaceutical formulation of embodiments 1-22. Embodiment 26 includes a modification to the method of embodiment 25, further sing diluting the pharmaceutical formulation prior to administration to the patient. Embodiment 27 pertains to a method of treating Pompe disease comprising: reconstituting the pharmaceutical composition of embodiment 23 or 24; and administering the reconstituted pharmaceutical ition to a patient in need thereof. id="p-34" id="p-34"

[0034] Another aspect of the invention pertains to a method of ing the above pharmaceutical formulations. Accordingly, embodiment 28 pertains to a method of preparing the pharmaceutical formulation of any of embodiments 1—22, the method comprising: adding the at least one buffer, at least one ent and recombinant acid osidase to water to provide a solution; optionally adjusting the pH of the solution; and optionally adding additional water to the solution. Embodiment 29 includes a modification to the method of embodiment 28, further comprising filtering the solution. Embodiment 30 es a modification to the method of embodiment 27 or 28, further comprising storing the solution.

BRIEF DESCRIPTION OF THE DRAWINGS id="p-35" id="p-35"

[0035] Further features of the present invention will become apparent from the following written description and the accompanying figures, in which: Figure 1A shows non-phosphorylated high mannose glycan, a mono-M6P glycan, and a P glycan.

Figure 1B shows the chemical structure of the M6P group. id="p-38" id="p-38"

[0038] Figure 2A describes productive targeting of rhGAA via glycans g M6P to target tissues (e.g. muscle s of subject with Pompe Disease).

Figure 2B describes non-productive drug clearance to non-target tissues (e.g. liver and ) or by binding of non-M6P glycans to non—target tissues.

Figures 3A and 3B, respectively, are graphs showing the results of CIMPR affinity chromatography of Lumizyme® and Myozyme®. The dashed lines refer to the M6P elution gradient. Elution with M6P displaces GAA les bound via an M6Pcontaining glycan to CIMPR. As shown in Figure 2A, 78% of the GAA activity in Lumizyme® eluted prior to addition of M6P. Figure 2B shows that 73% of the GAA Myozyme® activity eluted prior to addition of M6P. Only 22% or 27% of the rhGAA in Lumizyme® or Myozyme®, respectively, was eluted with M6P. These figures show that most of the rhGAA in these two conventional rhGAA products lack glycans having M6P needed to target CIMPR in target muscle tissues.

Figure 4 shows a DNA construct for orming CHO cells with DNA ng rhGAA. CHO cells were transformed with a DNA construct encoding rhGAA.

Figures 5A and 5B, respectively, are graphs showing the results of CIMPR affinity chromatography of Myozyme® and ATB200 rhGAA. As apparent from Figure 5B, about 70% of the rhGAA in ATB200 rhGAA contained M6P.

Figure 6 is a graph showing the results of CIMPR affinity chromatography of ATB200 rhGAA with and without capture on an anion exchange (AEX) column.

Figure 7 is a graph showing Polywax elution profiles of Lumizyme® and ATB200 rhGAAs. id="p-45" id="p-45"

[0045] Figure 8 is a table showing a summary of N—glycan structures of Lumizyme® ed to three different preparations of ATB200 rhGAA, identified as BP-rhGAA, ATB200-1 and ATB200-2. s 9A—9H show the results of a site—specific N-glycosylation analysis of ATB200 rhGAA. id="p-47" id="p-47"

[0047] Figure 10A is a graph ing the CIMPR binding affinity of ATB200 rhGAA (left trace) with that of Lumizyme® (right .

Figure 10B is a table comparing the Bis-M6P content of me® and ATB200 rhGAA.

Figure 11A is a graph comparing ATB200 rhGAA activity (left trace) with Lumizyme® rhGAA activity (right trace) inside normal fibroblasts at various GAA concentrations.

Figure 113 is a table comparing ATB200 rhGAA activity (left trace) with Lumizyme® rhGAA activity (right trace) inside fibroblasts from a subject having Pompe Disease at various GAA concentrations. id="p-51" id="p-51"

[0051] Figure 11C is a table comparing Kuptake of fibroblasts from normal subjects and subjects with Pompe Disease.

Figure 12A is a graph showing the amount of glycogen relative to dose of inant human acid a-glucosidase in mouse heart muscle after t with vehicle (negative control), with 20 mg/ml alglucosidase alfa (Lumizyme®), or with 5, 10 or 20 mg/kg ATB200.

Figure 123 is a graph showing the amount of glycogen relative to dose of recombinant human acid a—glucosidase in mouse quadriceps muscle after contact with vehicle (negative control), with 20 mg/ml alglucosidase alfa (Lumizyme®), or with 5, 10 or 20 mg/kg ATBZOO.

Figure 12C is a graph showing the amount of glycogen relative to dose of recombinant human acid a-glucosidase in mouse triceps muscle after contact with vehicle (negative control), with 20 mg/ml alglucosidase alfa (Lumizyme®), or with 5, 10 or 20 rug/kg ATBZOO.

Figure 13 is a table showing that the combination of ATBZOO rhGAA and chaperone miglustat ed significantly better glycogen clearance in GAA knock-out mice than treatments with either Lumizyme® or ATB200 rhGAAs without the miglustat chaperone. id="p-56" id="p-56"

[0056] Figure 14 is a series of electron micrographs of heart, diaphragm and soleus muscle from wild-type and Gaa-knockout mice treated with vehicle, alglucosidase alfa and ATB200 in the presence and absence of miglustat, showing levels of lysosome associated membrane protein (LAMP-1).

Figure 15 is a series of on micrographs of heart and soleus muscle from wild-type and Gaa-knockout mice treated with vehicle, osidase alfa and ATB200 in the presence and absence of tat, showing glycogen levels by staining with periodic acid — Schiff reagent (PAS).

Figure 16 is a series of electron raphs ) of quadriceps muscle from wild-type and ockout mice treated with vehicle, alglucosidase alfa and ATB200 in the presence and absence of miglustat, stained with methylene blue to show vacuoles ated by arrows).

Figure 17 is a series of electron micrographs (40X) of quadriceps muscle from wild-type and Gaa—knockout mice treated with vehicle, alglucosidase alfa and ATB200 in the presence and absence of miglustat, showing levels of the autophagy markers microtubule- associated protein lA/lB-light chain 3 phosphatidylethanolamine ate (LC3A II) and p62, the insulin—dependent glucose transporter GLUT4 and the insulin-independent glucose transporter GLUT1.

Figures 18A and 18B are graphs showing wire hand and grip strength muscle data for ype and Gaa-knockout mice treated with vehicle, alglucosidase alfa and ATBZOO in the presence of miglustat.

Figures 19A-19G are graphs showing glycogen levels in ceps, triceps and heart cells from ype and Gaa—knockout mice treated with vehicle, alglucosidase alfa and ATB200 in the presence and e of miglustat.

Figure 20 is a series of photomicrographs (100x and 200x) of muscle fibers of vastus lateralis (VL) from wild—type and Gaza-knockout mice treated with vehicle, alglucosidase alfa and ATB2OO in the presence and absence of miglustat, showing dystrophin signals.

Figure 21 shows the study design of an open-label, fixed-sequence, ing— dose, first-in-human, phase 1/2 study to assess the safety, tolerability, PK, PD, and efficacy of intravenous ons of ATB200 co-administered with oral miglustat in adults with Pompe disease. id="p-64" id="p-64"

[0064] Figures 22A—22B are graphs showing the concentration—time profiles of GAA total protein in plasma in human subjects after dosing of 5, 10 or 20 mg/kg ATB200, 20 mg/kg ATB200 and 130 mg tat, or 20 mg/kg ATB2OO and 260 mg miglustat.

Figure 22C is a graph showing the AUC of GAA total n in plasma in human subjects after dosing of 20 mg/kg ATBZOO, 20 mg/kg ATB200 and 130 mg miglustat, or 20 mg/kg ATB200 and 260 mg miglustat.

Figure 22D is a graph showing the concentration-time profiles of GAA total protein in plasma in two individual human subjects after dosing of 20 mg/kg ATB200 and 260 mg miglustat.

Figure 23 is a graph showing the concentration-time profiles of miglustat in plasma in human subjects after dosing of 130 mg or 260 mg of tat.

Figures 24A-24D are graphs showing changes in alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatine phosphokinase (CPK) and hexose tetrasaccharide (Hex4) levels in human patients after administration of ascending doses of ATB200 (5, 10 and 20 mg/kg) followed by co-administration of ATBZOO (20 mg/kg) and miglustat (130 and 260 mg).

DETAILED DESCRIPTION Before describing several exemplary embodiments of the invention, it is to be understood that the invention is not limited to the details of construction or process steps set forth in the following description. The invention is capable of other embodiments and of being practiced or being carried out in various ways. It will be tood that the embodiments listed below may be combined not only as listed below, but in other suitable combinations in accordance with the scope of the ion.

It has been singly discovered that by careful selection of buffers and excipients, a formulation for the recombinant GAA n ATB200 can be provided that exhibits superior stability and can undergo the processes ated with formulation preparation, storage, transportation, reconstitution and administration while ining enzyme activity and effectiveness, but minimizing precipitation of the enzyme. Accordingly, one aspect of the invention pertains to a formulation comprising rhGAA, a , and at least one excipient. In one or more embodiments, the rhGAA comprises ATBZOO. In some embodiments, the formulation is a liquid formulation. Details and various embodiments regarding the various ients of the ation follow below.

Definitions The terms used in this specification generally have their ordinary meanings in the art, within the context of this invention and in the specific context where each term is used.

Certain terms are discussed below, or ere in the specification, to provide additional guidance to the practitioner in describing the compositions and methods of the invention and how to make and use them.

In the present specification, except where the context requires ise due to express language or necessary implication, the word "comprises", or variations such as "comprises" or "comprising" is used in an ive sense i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

As used herein, the term "Pompe disease," also referred to as acid maltase deficiency, glycogen storage disease type II (GSDII), and glycogenosis type II, is ed to refer to a genetic lysosomal storage disorder characterized by mutations in the GAA gene, which codes for the human acid a-glucosidase enzyme. The term includes but is not limited to WO 73060 early and late onset forms of the disease, including but not limited to infantile, juvenile and adult-onset Pompe disease.

As used herein, the term "acid a-glucosidase" is intended to refer to a lysosomal enzyme which hydrolyzes oc-l,4 linkages n the D-glucose units of glycogen, maltose, and isomaltose. Alternative names include but are not limited to lysosomal (Jr-glucosidase (EC:3.2.1.20); glucoamylase; 1,4—a-D—glucan glucohydrolase; amyloglucosidase; gamma- amylase and exo-l,4-a-glucosidase. Human acid a-glucosidase is encoded by the GAA gene (National Centre for Biotechnology Information (NCBI) Gene ID 2548), which has been mapped to the long arm of chromosome 17 (location l7q25.2-q25.3). The full wild—type GAA amino acid sequence is set forth in SEQ ID NO: 1, as described in US Patent No. 8,592,362 and has GenB ank accession number AHE24104.1 (GI:568760974).

More than 500 mutations have currently been identified in the human GAA gene, many of which are associated with Pompe e. Mutations resulting in misfolding or misprocessing of the acid a—glucosidase enzyme include T1064C (Leu355Pro) and C2104T (Arg702Cys). In addition, GAA mutations which affect maturation and processing of the enzyme include Leu405Pro and Met519Thr. The conserved hexapeptide WIDMNE at amino acid es 516-521 is required for activity of the acid (x-glucosidase protein. As used herein, the abbreviation "GAA" is intended to refer to the acid cosidase enzyme, while the ized abbreviation "GAA" is intended to refer to the human gene coding for the human acid a—glucosidase enzyme. Thus, the iation "rhGAA" is intended to refer to the recombinant human acid (x—glucosidase enzyme.

As used herein, the term "alglucosidase alfa" is intended to refer to a recombinant human acid a—glucosidase identified as [199-arginine,223-histidine]prepro-a- glucosidase (human); Chemical Abstracts Registry Number 420794—05-0. Alglucosidase alfa is approved for marketing in the United States by e, as of January 2016, as the products me® and Myozyme®.

As used herein, the term "ATB200" is intended to refer to a recombinant human acid (x—glucosidase described in co—pending patent application , the disclosure of which is herein incorporated by reference. id="p-78" id="p-78"

[0078] As used herein, the term n" is intended to refer to a polysaccharide chain covalently bound to an amino acid e on a protein or polypeptide. As used , the term "N-glycan" or "N-linked glycan" is intended to refer to a polysaccharide chain attached to an amino acid residue on a protein or polypeptide through covalent binding to a nitrogen atom of the amino acid residue. For example, an N-glycan can be covalently bound to the side chain nitrogen atom of an asparagine residue. Glycans can contain one or several monosaccharide units, and the monosaccharide units can be covalently linked to form a straight chain or a branched chain. In at least one embodiment, N-glycan units attached to ATB200 can comprise one or more ccharide units each independently selected from ylglucosamine, mannose, galactose or sialic acid. The N-glycan units on the protein can be determined by any appropriate analytical technique, such as mass spectrometry. In some embodiments, the N— glycan units can be determined by liquid chromatography-tandem mass spectrometry (LC- MS/MS) utilizing an instrument such as the Thermo Scientific ap Velos PicTM Mass Spectrometer, Thermo Scientific Orbitrap Fusion Lumos TribidTM Mass Spectrometer or Waters Xevo® G2-XS QTof Mass Spectrometer.

As used herein, the term "high mannose N-glycan" is intended to refer to an N- glycan having one to six or more mannose units. In at least one embodiment, a high mannose an unit can contain a bis(N—acetylglucosamine) chain bonded to an asparagine residue and further bonded to a branched polymannose chain. As used herein interchangeably, the term "M6P" or "mannosephosphate" is intended to refer to a mannose unit phosphorylated at the 6 position; i.e. having a phosphate group bonded to the hydroxyl group at the 6 on. In at least one embodiment, one or more mannose units of one or more an units are phosphorylated at the 6 position to form mannosephosphate units.

As used herein, the term "complex N-glycan" is intended to refer to an N-glycan containing one or more galactose and/or sialic acid units. In at least one ment, a complex N-glycan can be a high mannose N-glycan in which one or mannose units are further bonded to one or more monosaccharide units each independently selected from N-acetylglucosamine, galactose and sialic acid.

As used herein, the "therapeutically effective dose" and tive amount" are intended to refer to an amount of acid a-glucosidase, which is ient to result in a therapeutic se in a subject. A therapeutic response may be any response that a user (for example, a clinician) will ize as an effective response to the y, including any surrogate al markers or symptoms described herein and known in the art. Thus, in at least one embodiment, a therapeutic response can be an amelioration or inhibition of one or more symptoms or markers of Pompe disease such as those known in the art. Symptoms or markers of Pompe disease include but are not limited to decreased acid osidase tissue activity; cardiomyopathy; cardiomegaly; progressive muscle weakness, especially in the trunk or lower limbs; profound hypotonia; macroglossia (and in some cases, protrusion of the tongue); difficulty swallowing, sucking, and/or feeding; atory insufficiency; hepatomegaly (moderate); laxity of facial s; areflexia; exercise rance; exertional dyspnea; orthopnea; sleep apnea; g headaches; somnolence; lordosis and/or sis; decreased deep tendon reflexes; lower back pain; and failure to meet pmental motor milestones.

As used herein, the term "enzyme replacement y" or "ERT" is intended to refer to the introduction of a non-native, purified enzyme into an individual having a deficiency in such enzyme. The administered protein can be ed from l sources or by recombinant expression. The term also refers to the introduction of a ed enzyme in an individual otherwise requiring or benefiting from administration of a ed enzyme. In at least one embodiment, such an individual suffers from enzyme insufficiency. The introduced enzyme may be a purified, recombinant enzyme produced in vitro, or a protein purified from isolated tissue or fluid, such as, for example, placenta or animal milk, or from plants.

As used herein, the term "pharmaceutically acceptable" is intended to refer to molecular entities and compositions that are physiologically tolerable and do not typically produce untoward reactions when administered to a human. ably, as used , the term "pharmaceutically acceptable" means approved by a regulatory agency of the federal or a state government or listed in the US. Pharmacopeia or other generally recognized pharmacopeia for use in animals, and more particularly in humans.

As used herein, the term "excipient" refers to a substance other than the active ingredients included in a formulation, and are generally an inactive material. The excipient may help transport the active drug to the site where the drug is intended to take effect, control the release of the active drug, or help with solubilization, or can serve a variety of other functions. Examples of excipients include, but are not limited to, buffering agents, surfactants, antimicrobial agents, antioxidants, bulking agents, stabilizer, tonicity modifiers etc.

As used herein, the term "buffer" is intended to refer to a solution containing both a weak acid and its conjugate weak base, the pH of which s only slightly with the on of an alkali or acid. As will be explained further below, in some embodiments, the buffer used in the pharmaceutical formulation is a citrate and/or phosphate buffer.

As used herein, the terms "subject" or "patient" are ed to refer to a human or non-human animal. In at least one embodiment, the subject is a mammal. In at least one embodiment, the subject is a human.

As used herein, the terms "about" and "approximately" are ed to refer to an acceptable degree of error for the quantity measured given the nature or precision of the measurements. For example, the degree of error can be indicated by the number of significant figures provided for the measurement, as is understood in the art, and includes but is not limited to a variation of il in the most precise significant figure reported for the measurement.

Typical exemplary degrees of error are within 20 percent (%), preferably within 10%, and more preferably within 5% of a given value or range of values. Alternatively, and particularly in biological systems, the terms " and "approximately" can mean values that are within an order of magnitude, ably within 5-fold and more preferably within 2-fold of a given value. Numerical quantities given herein are approximate unless stated otherwise, meaning that the term "about" or "approximately" can be inferred when not expressly stated. id="p-88" id="p-88"

[0088] Reference throughout this specification to "one embodiment," "certain embodiments, various embodiments, one or more embodiments" or "an embodiment" means that a ular feature, ure, material, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Thus, the appearances of the phrases such as "in one or more embodiments," "in certain embodiments," "in various embodiments," "in one ment" or "in an embodiment" in various places throughout this specification are not necessarily ing to the same embodiment of the invention. Furthermore, the particular features, structures, als, or characteristics may be ed in any suitable manner in one or more embodiments.

ATB200 rhGAA The formulation comprises ATBZOO, which is a recombinant GAA ) protein suitable for use in enzyme replacement therapy. Details regarding the structure and production of ATBZOO, as well as variants, are provided below.

In at least one embodiment, the ATB200 is expressed in Chinese hamster ovary (CHO) cells and ses an increased content of N—glycan units bearing one or more mannoseph0sphate residues when compared to a t of N-glycan units bearing one or more mannosephosphate residues of alglucosidase alfa. There are seven potential N-linked glycosylation sites on rhGAA. Since each glycosylation site is heterogeneous in the type of N— linked oligosaccharides (N-glycans) present, rhGAA consist of a complex mixture of proteins with N-glycans having varying binding affinities for M6P receptor and other carbohydrate receptors. rhGAA that contains a high e N-glycans having one M6P group (mono- M6P) binds to CIMPR with low (~6,000 nM) affinity while rhGAA that contains two M6P groups on same N-glycan (bis-M6P) bind with high (~2 nM) affinity. Representative ures for non-phosphorylated, mono-M6P, and bis-M6P glycans are shown by Figure 1A. The mannose—6-P group is shown by Figure 1B. Once inside the lysosome, rhGAA can enzymatically degrade accumulated glycogen. r, conventional rhGAAs have low total levels of M6P- and bis-M6P bearing glycans and, thus, target muscle cells poorly resulting in inferior delivery of rhGAA to the lysosomes. Productive drug targeting of rhGAA is shown in Figure 2A. The majority of rhGAA molecules in these conventional products do not have phosphorylated N-glycans, thereby lacking affinity for the CIMPR. Non-phosphorylated high mannose glycans can also be cleared by the mannose receptor which results in oductive clearance of the ERT (Figure 2B).

The other type of N-glycans, complex carbohydrates, which contain galactose and sialic acids, are also present on rhGAA. Since x N-glycans are not phosphorylated they have no affinity for CIMPR. However, complex-type ans with exposed galactose es have te to high affinity for the asialoglycoprotein receptor on liver hepatocytes which leads to rapid non-productive clearance of rhGAA e 2B).

In at least one embodiment, the acid (x-glucosidase is a recombinant human acid a-glucosidase referred to herein as ATB200, as described in co-pending international patent application PCT/U82015/053252. ATB200 has been shown to bind cation-independent mannose—6-phosphate receptors (CIMPR) with high affinity (KD ~ 2-4 nM) and to be efficiently internalized by Pompe lasts and skeletal muscle myoblasts (meke ~ 7-14 nM). ATB200 was characterized in vivo and shown to have a shorter apparent plasma half-life "1/2 ~ 45 min) than alglucosidase alfa (tug ~ 60 min).

In one or more embodiments, the inant human acid a—glucosidase has a ype GAA amino acid sequence as set forth in SEQ ID NO: 2, which corresponds to amino acid residues 57-952 and is cal to wild-type (WT) human GAA after natural intracellular proteolytic processing that removes the initial 56 residues comprising the signal peptide and precursor peptide. The ATB200 amino acid sequence has been ed by tryptic digestion followed by liquid tography/mass spectroscopy as well as by amino acid cing. By contrast, the current standard of care rhGAA enzyme ement therapy (ERT) (commercially available ts containing alglucosidase alfa are Myozyme® in most countries and Lumizyme® in the US, e, a Sanofi Company) differs from WT GAA and contains 3 amino acid residues substitutions: histidine changed to arginine at position 199, ne changed to histidine at 223, and valine changed to isoleucine at 780.

In at least one embodiment, the recombinant human acid (x-glucosidase undergoes post-translational and/or chemical modifications at one or more amino acid residues in the n. For example, methionine and tryptophan residues can undergo oxidation. As another example, asparagine residues can undergo deamidation to aspartic acid. As yet another example, aspartic acid can undergo isomerization to iso—aspartic acid. Accordingly, in some embodiments the enzyme is initially expressed as having an amino acid ce as set forth in SEQ ID NO: 1, or SEQ ID NO: 2, and the enzyme undergoes one or more of these post- translational and/or chemical modifications. Such modifications are also within the scope of the present disclosure.

In at least one embodiment, the recombinant human acid a-glucosidase has a wild-type GAA amino acid sequence as set forth in SEQ ID NO: 1, as described in US Patent No. 8,592,362 and has GenBank accession number AHE24104.1 (GI:568760974). In at least one embodiment, the recombinant human acid a-glucosidase is glucosidase alfa, the human acid a-glucosidase enzyme encoded by the most predominant of nine observed haplotypes of the GAA gene.

In at least one embodiment, the recombinant human acid a-glucosidase is initially sed as having the full-length 952 amino acid ce of wild-type GAA as set forth in SEQ ID NO: 1, and the recombinant human acid (x-glucosidase undergoes intracellular processing that removes a portion of the amino acids, e.g. the first 56 amino acids.

Accordingly, the recombinant human acid a—glucosidase that is secreted by the host cell can have a shorter amino acid sequence than the recombinant human acid a-glucosidase that is initially expressed within the cell. In at least one ment, the shorter protein can have the amino acid sequence set forth in SEQ ID NO: 2, which only differs from SEQ ID NO: 1 in that the first 56 amino acids comprising the signal peptide and sor peptide have been removed, thus resulting in a protein having 896 amino acids. Other variations in the number of amino acids is also possible, such as having 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more deletions, substitutions and/or insertions relative to the amino acid sequence described by SEQ ID NO: 1 or SEQ ID NO: 2. In some ments, the rhGAA product includes a mixture of recombinant human acid a—glucosidase molecules having different amino acid lengths.

In at least one embodiment, the recombinant human acid a-glucosidase undergoes post—translational and/or chemical modifications at one or more amino acid residues in the protein. For example, methionine and tryptophan residues can undergo oxidation. As another example, the N—terminal glutamine can form pyro-glutamate. As another example, gine residues can undergo deamidation to aspartic acid. As yet r example, aspartic acid residues can undergo isomerization to iso-aspartic acid. As yet another example, unpaired cysteine residues in the protein can form disulfide bonds with free glutathione and/or cysteine.

Accordingly, in some embodiments the enzyme is initially expressed as having an amino acid ce as set forth in SEQ ID NO: 1 or SEQ ID NO: 2, and the enzyme undergoes one or more of these post-translational and/or chemical modifications. Such cations are also within the scope of the present disclosure.

Preferably, no more than 70, 65, 60, 55, 45, 40, 35, 30, 25, 20, 15, 10, or 5% of the total inant human acid a—glucosidase molecules lack an N-glycan unit bearing one or more mannose—6—phosphate residues or lacks a capacity to bind to the cation ndent mannosephosphate receptor (CIMPR). Alternatively, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 99%, <100% or more of the recombinant human acid a-glucosidase molecules comprise at least one N-glycan unit bearing one or more mannose-6—phosphate residues or has the capacity to bind to CIMPR.

The recombinant human acid a—glucosidase molecules may have 1, 2, 3 or 4 e—6-phosphate (M6P) groups on their glycans. For example, only one N-glycan on a recombinant human acid a-glucosidase molecule may bear M6P phosphorylated), a single N-glycan may bear two M6P groups (bis—phosphorylated), or two different N-glycans on the same recombinant human acid (x-glucosidase le may each bear single M6P .

Recombinant human acid a—glucosidase molecules may also have N—glycans bearing no M6P groups. In r embodiment, on average the N-glycans contain greater than 3 mol/mol of M6P and greater than 4 mol/mol sialic acid, such that the recombinant human acid a- glucosidase comprises on average at least 3 moles of mannosephosphate residues per mole of recombinant human acid (x-glucosidase and at least 4 moles of sialic acid per mole of WO 73060 recombinant human acid (x-glucosidase. On average at least about 3, 4, 5, 6, 7, 8, 9, or 10% of the total glycans on the recombinant human acid a-glucosidase may be in the form of a mono- M6P glycan, for e, about 6.25% of the total s may carry a single M6P group and on average, at least about 0.5, 1, 1.5, 2.0, 2.5, 3.0% of the total glycans on the recombinant human acid (x—glucosidase are in the form of a bis—M6P glycan and on average less than 25% of total recombinant human acid osidase contains no phosphorylated glycan binding to CIMPR.

The recombinant human acid a-glucosidase may have an average content of N-glycans carrying M6P ranging from 0.5 to 7.0 mol/mol recombinant human acid (1- glucosidase or any intermediate value of subrange including 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, or 7.0 mol/mol recombinant human acid a-glucosidase. The recombinant human acid (x-glucosidase can be fractionated to provide inant human acid on- glucosidase preparations with different average numbers of M6P-bearing or bis—M6P-bearing glycans thus permitting r customization of recombinant human acid a-glucosidase targeting to the lysosomes in target tissues by selecting a particular fraction or by selectively combining different fractions.

In some ments, the recombinant human acid (x-glucosidase will bear, on average, 2.0 to 8.0 moles of M6P per mole of recombinant human acid a—glucosidase. This range includes all intermediate values and subranges including 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5 and 8.0 mol M6P/mol recombinant human acid a—glucosidase.

Up to 60% of the N-glycans on the recombinant human acid (x—glucosidase may be fully sialylated, for e, up to 10%, 20%, 30%, 40%, 50% or 60% of the N—glycans may be fully sialyated. In some embodiments from 4 to 20% of the total N-glycans are fully sialylated. In other embodiments no more than 5%, 10%, 20% or 30% of N-glycans on the inant human acid a-glucosidase carry sialic acid and a terminal galactose residue (Gal).

This range includes all intermediate values and subranges, for example, 7 to 30% of the total N-glycans on the inant human acid (x—glucosidase can carry sialic acid and terminal galactose. In yet other embodiments, no more than 5, 10, 15, 16, 17, 18, 19 or 20% of the N— glycans on the recombinant human acid a-glucosidase have a terminal ose only and do not contain sialic acid. This range includes all intermediate values and subranges, for example, from 8 to 19% of the total N—glycans on the inant human acid a—glucosidase in the composition may have terminal galactose only and do not contain sialic acid.

In other embodiments of the invention, 40, 45, 50, 55 to 60% of the total N-glycans on the recombinant human acid a-glucosidase are x type N-glycans; or no more than 1, 2, 3, 4, 5, 6, 7% of total N—glycans on the recombinant human acid a-glucosidase are hybrid-type ans; no more than 5, 10, or 15% of the high mannose—type N-glycans on the recombinant human acid a—glucosidase are non-phosphorylated; at least 5% or 10% of the high mannose—type N-glycans on the recombinant human acid a-glucosidase are mono—M6P phosphorylated; and/or at least 1 or 2% of the high mannose—type N-glycans on the recombinant human acid (x—glucosidase are P phosphorylated. These values include all intermediate values and subranges. A inant human acid a-glucosidase may meet one or more of the content ranges described above.

In some ments, the recombinant human acid (x-glucosidase will bear, on e, 2.0 to 8.0 moles of sialic acid residues per mole of recombinant human acid (x- glucosidase. This range includes all intermediate values and subranges ing 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5 and 8.0 mol residues/mol recombinant human acid (x— glucosidase. Without being bound by theory, it is believed that the presence of an units bearing sialic acid residues may prevent non-productive clearance of the recombinant human acid oc—glucosidase by asialoglycoprotein receptors.

In one or more embodiments, the rhGAA has M6P and/or sialic acid units at certain N-glycosylation sites of the recombinant human lysosomal protein. For example, there are seven potential N-linked glycosylation sites on rhGAA. These potential ylation sites are at the following positions of SEQ ID NO: 2: N84, N177, N334, N414, N596, N826 and N869. Similarly, for the full-length amino acid sequence of SEQ ID NO: 1, these potential glycosylation sites are at the following positions: N140, N233, N390, N470, N652, N882 and N925. Other variants of rhGAA can have similar glycosylation sites, depending on the location of asparagine residues. Generally, sequences of ASN-X-SER or ASN-X-THR in the protein amino acid ce indicate potential glycosylation sites, with the exception that X cannot be HIS or PRO.

In various embodiments, the rhGAA has a certain N—glycosylation profile. In one or more embodiments, at least 20% of the rhGAA is phosphorylated at the first N- glycosylation site (e.g. N84 for SEQ ID NO: 2 and N140 for SEQ ID NO: 1). For example, at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA can be phosphorylated at the first N-glycosylation site. This orylation can be the result of mono-M6P and/or bis—M6P units. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA bears a mono-M6P unit at the first N—glycosylation site. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA bears a bis-M6P unit at the first N— glycosylation site.

In one or more embodiments, at least 20% of the rhGAA is phosphorylated at the second N—glycosylation site (e.g. N177 for SEQ ID NO: 2 and N223 for SEQ ID NO: 1).

For example, at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA can be phosphorylated at the second N-glycosylation site. This phosphorylation can be the result of mono-M6P and/or P units. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA bears a mono-M6P unit at the second N- glycosylation site. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA bears a bis— M6P unit at the second N-glycosylation site. In one or more embodiments, at least 5% of the rhGAA is phosphorylated at the third N-glycosylation site (e. g. N334 for SEQ ID NO: 2 and N390 for SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20% or 25% of the rhGAA is phosphorylated at the third osylation site. For example, the third N- glycosylation site can have a mixture of non-phosphorylated high mannose glycans, di-, tri-, and tetra-antennary complex glycans, and hybrid glycans as the major species. In some embodiments, at least 3%, 5%, 8%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45% or 50% of the rhGAA is sialylated at the third N-glycosylation site.

In one or more ments, at least 20% of the rhGAA is phosphorylated at the fourth osylation site (e.g. N414 for SEQ ID NO: 2 and N470 for SEQ ID NO: 1).

For example, at least 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA can be orylated at the fourth N-glycosylation site. This phosphorylation can be the result of mono-M6P and/or bis—M6P units. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA bears a mono-M6P unit at the fourth N- glycosylation site. In some embodiments, at least 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA bears a bis- M6P unit at the fourth N-glycosylation site. In some embodiments, at least 3%, 5%, 8%, 10%, %, 20% or 25% of the rhGAA is sialylated at the fourth N-glycosylation site.

In one or more embodiments, at least 5% of the rhGAA is phosphorylated at the fifth N-glycosylation site (e.g. N596 for SEQ ID NO: 2 and N692 for SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20% or 25% of the rhGAA is phosphorylated at the fifth N—glycosylation site. For example, the fifth N-glycosylation site can have fucosylated di- antennary complex glycans as the major species. In some ments, at least 3%, 5%, 8%, %, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA is sialylated at the fifth N-glycosylation site. id="p-110" id="p-110"

[00110] In one or more embodiments, at least 5% of the rhGAA is phosphorylated at the sixth N-glycosylation site (e.g. N826 for SEQ ID NO: 2 and N882 for SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20% or 25% of the rhGAA is phosphorylated at the sixth osylation site. For example, the sixth N-glycosylation site can have a mixture of di-, tri-, and tetra-antennary x glycans as the major species. In some embodiments, at least 3%, 5%, 8%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the rhGAA is sialylated at the sixth N-glycosylation site.

In one or more embodiments, at least 5% of the rhGAA is phosphorylated at the seventh N-glycosylation site (e.g. N869 for SEQ ID NO: 2 and N925 for SEQ ID NO: 1). In other embodiments, less than 5%, 10%, 15%, 20% or 25% of the rhGAA is phosphorylated at the seventh N-glycosylation site. In some embodiments, less than 40%, 45%, 50%, 55%, 60% or 65% % of the rhGAA has any glycan at the h N-glycosylation site. In some embodiments, at least 30%, 35% or 40% of the rhGAA has a glycan at the seventh N- glycosylation site.

In various embodiments, the rhGAA has an average fucose content of 0-5 mol per mol of rhGAA, GlcNAc content of 10-30 mol per mol of rhGAA, galactose content of 5-20 mol per mol of rhGAA, mannose content of 10—40 mol per mol of rhGAA, M6P t of 2—8 mol per mol of rhGAA and sialic acid content of 2-8 mol per mol of rhGAA. In various embodiments, the rhGAA has an average fucose content of 2—3 mol per mol of rhGAA, GlcNAc content of 20-25 mol per mol of rhGAA, galactose content of 8-12 mol per mol of rhGAA, mannose content of 22-27 mol per mol of rhGAA, M6P t of 3-5 mol per mol of rhGAA and sialic acid content of 4-7 mol of rhGAA.

The inant human acid a-glucosidase is ably produced by Chinese hamster ovary (CHO) cells, such as CHO cell line GA-ATB200 or ATB200X5-14, or by a subculture or derivative of such a CHO cell culture. DNA ucts, which express allelic variants of acid (x-glucosidase or other variant acid (x-glucosidase amino acid sequences such as those that are at least 90%, 95% or 99% identical to SEQ ID NO: 1, may be ucted and expressed in CHO cells. These variant acid a—glucosidase amino acid sequences may contain deletions, substitutions and/or insertions relative to SEQ ID NO: 1, such as having 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more deletions, substitutions and/or insertions to the amino acid sequence described by SEQ ID NO: 1. Those of skill in the art can select alternative vectors suitable for transforming CHO cells for production of such DNA constructs.

Various alignment algorithms and/or programs may be used to calculate the identity between two ces, including FASTA, or BLAST which are available as a part of the GCG sequence analysis package (University of Wisconsin, Madison, Wis), and can be used with, e.g., default setting. For example, polypeptides having at least 90%, 95% or 99% identity to specific polypeptides described herein and preferably exhibiting substantially the same functions, as well as polynucleotide encoding such polypeptides, are contemplated.

Unless otherwise ted a similarity score will be based on use of BLOSUM62. When BLASTP is used, the percent similarity is based on the BLASTP positives score and the percent sequence ty is based on the BLASTP identities score. BLASTP "Identities" shows the number and fraction of total residues in the high g sequence pairs which are identical; and BLASTP "Positives" shows the number and fraction of es for which the alignment scores have positive values and which are similar to each other. Amino acid ces having these degrees of identity or similarity or any intermediate degree of identity of similarity to the amino acid sequences disclosed herein are contemplated and encompassed by this sure. The polynucleotide sequences of similar polypeptides are deduced using the c code and may be obtained by conventional means, in particular by reverse translating its amino acid sequence using the genetic code.

In some embodiments, the recombinant human acid a-glucosidase having superior ability to target cation-independent mannosephosphate receptors (CIMPR) and cellular lysosomes as well as glycosylation patterns that reduce its non-productive clearance in vivo can be produced using e hamster ovary (CHO) cells. These cells can be induced to express recombinant human acid a—glucosidase with icantly higher levels of N—glycan units bearing one or more mannose-6—phosphate es than conventional recombinant human acid (x-glucosidase products such as alglucosidase alfa. The recombinant human acid (1- glucosidase ed by these cells, for example, as exemplified by , has significantly more muscle cell-targeting mannosephosphate (M6P) and bis-mannosephosphate N- glycan residues than conventional acid (x—glucosidase, such as me®. Without being bound by theory, it is believed that this ive glycosylation allows the ATB200 enzyme to be taken up more ively into target cells, and therefore to be cleared from the circulation more efficiently than other recombinant human acid (x-glucosidases, such as for example, alglucosidase alfa, which has a much lower M6P and bis-M6P content. ATB200 has been shown to efficiently bind to CIMPR and be efficiently taken up by skeletal muscle and cardiac muscle and to have a ylation pattern that provides a favorable cokinetic profile and reduces non-productive clearance in vivo.

It is also plated that the extensive glycosylation of ATB200 can contribute to a reduction of the immunogenicity of ATB200 ed to, for example, alglucosidase alfa. As will be appreciated by those skilled in the art, glycosylation of proteins with conserved mammalian sugars generally enhances product solubility and diminishes product aggregation and immunogenicity. Glycosylation indirectly alters protein immunogenicity by minimizing n aggregation as well as by shielding immunogenic protein epitopes from the immune system nce for Industry — Immunogenicity Assessment for Therapeutic Protein Products, US Department of Health and Human Services, Food and Drug stration, Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research, August 2014). Therefore, in at least one embodiment, administration of the recombinant human acid a-glucosidase does not induce anti-drug antibodies. In at least one embodiment, administration of the recombinant human acid a-glucosidase induces a lower incidence of anti-drug dies in a subject than the level of anti-drug antibodies induced by administration of alglucosidase alfa.

As described in co-pending international patent application , cells such as CHO cells can be used to produce the rhGAA described therein, and this rhGAA can be used in the present invention. Examples of such a CHO cell line are GA-ATB200 or ATB200X5-14, or a subculture thereof that produces a rhGAA composition as described therein. Such CHO cell lines may contain multiple copies of a gene, such as 5, 10, 15, or 20 or more copies, of a polynucleotide encoding GAA.

] The high M6P and bis—M6P rhGAA, such as ATB200 rhGAA, can be produced by transforming CHO cells with a DNA construct that s GAA. While CHO cells have been previously used to make rhGAA, it was not appreciated that ormed CHO cells could be cultured and selected in a way that would produce rhGAA having a high t of M6P and bis—M6P glycans which target the CIMPR.

Surprisingly, it was found that it was possible to transform CHO cell lines, select transformants that produce rhGAA containing a high t of glycans bearing M6P or bis-M6P that target the CIMPR, and to stably express this high—M6P rhGAA. Thus, methods for making these CHO cell lines are also described in co—pending international patent application . This method involves transforming 3 CH0 cell with DNA encoding GAA or a GAA variant, selecting a CHO cell that stably integrates the DNA encoding GAA into its some(s) and that stably expresses GAA, and selecting a CHO cell that expresses GAA having a high content of glycans bearing M6P or bis-M6P, and, optionally, ing a CHO cell having ans with high sialic acid content and/or having N—glycans with a low non—phosphorylated high mannose content. These CHO cell lines may be used to produce rhGAA and rhGAA compositions by culturing the CH0 cell line and recovering said composition from the culture of CHO cells.

In one or more ments, the ATB200 is present in an amount ranging from about 5 to about 50 mg/mL. In further embodiments, the ATB200 is present in an amount ranging from about 5, 8, 10, or 12 to about 18, 20, 25, 30, 35, 40, 45 or 50 mg/mL. In some embodiments, the ATB200 is present in an amount of about 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30 mg/mL. In further embodiments, the ATB200 is present in an amount of about 15 mg/mL. pH and Buffer id="p-121" id="p-121"

[00121] The pH of the formulation may range from about 5.0 to about 7.0 or about 5.0 to about 6.0. In one or more embodiments, the pH ranges from about 5.5 to about 6.0. In some embodiments, the formulation has a pH of about 5.0, 5.1, 5.2, 5.3, 5.4, 5.5, 5.6, 5.7, 5.8, 5.9 or 6.0, 6.1, 6.2, 6.3, 6.4 or 6.5. Generally, the amounts of components as recited will yield a pH in the range of from about 5.0 to about 6.0. However, the pH may be adjusted to a target pH by using pH adjusters (i.e., alkalizing agents and acidifying agents), such as sodium hydroxide and/or hydrochloric acid.

The formulation also comprises a buffer that is selected from the group consisting of a citrate, a phosphate and combinations thereof. As used herein, "buffer" refers to a buffer solution containing a weak acid and its conjugate base that helps to prevent s in pH. The citrate and/or phosphate may be a sodium citrate or sodium phosphate. Other salts include potassium and ammonium salts. In one or more embodiments, the buffer comprises a citrate. In further embodiments, the buffer comprises sodium citrate (e.g., a mixture of sodium citrate dehydrate and citric acid monohydrate). In one or more embodiments, buffer solutions comprising a citrate may comprise sodium citrate and citric acid. In some embodiments, both a citrate and ate buffer are t.

The formulation also comprises at least one excipient. Some embodiments of the invention comprise ents that aid with tonicity, act as a bulking agent or act as stabilizers. In one or more embodiments, the at least one excipient is selected from the group consisting of mannitol, polysorbate and combinations f. id="p-124" id="p-124"

[00124] In one or more embodiments, the excipient comprises an ingredient that can act as a tonicity er and/or bulking agent, particularly mannitol. Tonicity agents are components which help to ensure the formulation has an osmotic pressure r to or the same as human blood. Bulking agents are ingredients which add mass to the formulations (e.g. lyophilized) and provide an adequate structure to the cake. id="p-125" id="p-125"

[00125] In one or more embodiments, the total amount of tonicity and/or bulking agent ranges in an amount of from about 10 to about 50 mg/mL. In further ments, the total amount of tonicity and/or bulking agent ranges in an amount of from about 10, 11, 12, 13, 14 or 15 to about 16, 20, 25, 30, 35, 40, 45 or 50 mg/mL. In some embodiments, the tonicity and/or bulking agent comprises mannitol. In one or more embodiments, the mannitol is present in an amount of from about 10 to about 50 mg/mL. In r embodiments, the ol is present in an amount of from about 10, 11, 12, 13, 14 or 15 to about 16, 20, 25, 30, , 40, 45 or 50 mg/mL. In yet further embodiments, the mannitol is present in an amount of about 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 mg/mL. In some embodiments, mannitol is the only tonicity and/or bulking agent. Examples of other tonicity and/or bulking agents include, sodium chloride, sucrose, and ose.

In some embodiments, the excipient comprises an ingredient that can act a stabilizer, such as polysorbate 80. izers are compounds that can prevent or minimize the aggregate formation at the hydrophobic air-water interfacial surfaces. In some embodiments, the stabilizer is a surfactant. In one or more embodiments, the total amount of izer ranges from about 0.1 to about 1.0 mg/mL. In further embodiments, the total amount of stabilizer ranges from about 0.1, 0.2, 0.3, 0.4 or 0.5 to about 0.5, 0.6, 0.7, 0.8 0.9 or 1.0 mg/mL. In yet further embodiments, the total amount of izer is about 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8 0.9 or 1.0 mg/mL. In some embodiments, polysorbate 80 is the only stabilizer. Thus, in one or more ments, the polysorbate 80 ranges from about 0.1 to about 1.0 mg/mL. In further embodiments, the polysorbate 80 ranges from about 0.1, 0.2, 0.3, 0.4 or 0.5 to about 0.5, 0.6, 0.7, 0.8 0.9 or 1.0 mg/mL. In yet further embodiments, the polysorbate 80 is about 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8 0.9 or 1.0 Ing/mL.

In one or more embodiments, it may be ble to exclude certain compounds from the formulation. For example, in preparing a formulation for the treatment of given disease, it would be desirable to exclude certain compounds which would aggravate the underlying disease. As discussed above, individuals with Pompe disease have a decreased or absent ability to breakdown glycogen. Several commonly used ents, such as sucrose, trehalose and glycine, may be converted into glucose in the body, which would in turn aggravate Pompe disease. Thus, in some embodiments, sucrose, trehalose and/0r glycine is excluded from the formulation. Similarly, n components may be excluded which are not best suited for the formulation given certain contexts. For example, in some embodiments, poloxamers may be excluded.

Exemplary Embodiments In one or more embodiments, the formulation comprises or consists essentially (a) ATB200 (e.g., a recombinant acid (x-glucosidase, wherein the recombinant acid (x— glucosidase is expressed in Chinese hamster ovary (CHO) cells and comprises an sed content of an units g one or two mannose-6—phosphate residues when compared to a content of N—glycan units bearing one or two e-6—phosphate residues of alglucosidase alfa); (b) at least one buffer selected from the group consisting of a citrate, a phosphate and combinations thereof; (c) at least one excipient ed from the group consisting of mannitol, polysorbate 80, and combinations thereof; and wherein the formulation has a pH of from about 5.0 to about 6.0 or 7.0.

In some embodiments, the formulation ses or consists essentially of: (a) ATB200 (e.g., a recombinant acid a—glucosidase, wherein the recombinant acid a-glucosidase is expressed in Chinese hamster ovary (CHO) cells and comprises an sed content of N-glycan units bearing one or two mannosephosphate residues when compared to a content of N-glycan units bearing one or two mannose—6—phosphate residues of alglucosidase alfa); (b1) sodium citrate; (b2) citric acid monohydrate; (cl) mannitol; (c2) rbate 80; (d) water; (e) optionally, an ying agent; and (f) optionally, an alkalizing agent, wherein the formulation has a pH of from about 5.0 to about 6.0 or 7.0. In further ments, the pH ranges from about 5.5 to about 6.0. In yet further embodiments, the pH is about 6.0.

In a specific embodiment, the formulation comprises (a) ATB200 (e.g., a recombinant acid a-glucosidase, wherein the recombinant acid a—glucosidase is expressed in Chinese hamster ovary (CHO) cells and comprises an increased content of N-glycan units bearing one or two mannose-6—phosphate residues when compared to a content of an units bearing one or two ephosphate residues of alglucosidase alfa), present at a concentration of about 5-30 mg/mL or about 15 mg/mL; (b) sodium citrate buffer, present at a concentration of about 10-100 mM or about mM; (c1) mannitol, present at a concentration of about 10-50 mg/mL, or about 20 mg/mL; (c2) polysorbate 80, present at a concentration of about 01—] mg/mL, or about 0.5 mg/mL; and (d) water; (e) optionally, an acidifying agent; and (f) optionally, an alkalizing agent, wherein the formulation has a pH of from about 5.0 to about 6.0 or 7.0. In further ments, the pH ranges from about 5.5 to about 6.0. In yet further embodiments, the pH is about 6.0.

Preparation of Formulation id="p-131" id="p-131"

[00131] Another aspect of the invention pertains to methods of preparing the formulations described herein. In one or more embodiments, the formulation can be prepared from the enzyme solution. This on may be trated and buffer exchanged to the targeted concentration and buffers as necessary using methods known in the art. Additional ents (6. g. excipients and pH adjusters) may then be added. The formulation can then be filtered and placed into a e container and stored.

In one or more embodiments, the method of preparing any of the pharmaceutical formulations described herein comprises: adding the at least one buffer, at least one excipient and recombinant acid a-glucosidase to water to provide a solution; optionally adjusting the pH of the on; and optionally adding onal water to the solution. In further embodiments, the method further comprises ing the solution. In yet further embodiments, the method further comprises g the solution.

An exemplary, non—limiting process for the production of a formulation as described herein follows below: 1. In a suitable manufacturing vessel, add approximately 85-90% of the batch volume of Water for Injection. 2. Add and dissolve the desired excipients and buffers (e.g., sodium e dihydrate, citric acid monohydrate, polysorbate 80, mannitol) and mix until dissolved. 3. Add the ATBZOO drug substance and mix. 4. Adjust the pH to the targeted pH (e.g., 6 i 0.1) as necesary with adjusters (e. g. ,) hydrochloric acid or sodium hydroxide solution. >19." Add sufficient Water for Injection to final volume and mix.

Filter the on through a e filter into a sterile receiver.

Aseptically fill the drug product solution into vials, and insert stoppers. 8. Cap all vials and store at 2°—8°C.

In one or more embodiments, the formulation pared will be in liquid form.

That is, the formulation comprises water. This liquid formulation may undergo lyophilization (freeze—drying) process to provide a cake or powder. ingly, another aspect of the invention pertains to a pharmaceutical composition comprising anyone of the formulations described above after lyophilization. The lyophilized mixture may comprise the ATB200, buffer selected from the group consisting of a citrate, a phosphate and combinations thereof, and at least one excipient selected from the group consisting of trehalose, mannitol, polysorbate 80, and combinations thereof. In some embodiments, other ingredients (e. g., other excipients) may be added to the lyophilized mixture. The pharmaceutical composition comprising the lyophilized formulation may be provided vial, which then can be stored, transported, tituted and/or administered to a patient.

Method of Treatment and Administration to t Another aspect of the invention pertains to a method of treatment of Pompe disease and/or use of the formulations described herein for the treatment of Pompe disease.

The formulation may be administered pared, or after lyophilization and reconstitution.

Thus, in one or more ments, the method comprises administering to a patient in need thereof any of the pharmaceutical formulations described above. In other embodiments, the method comprises reconstituting the lyophilized pharmaceutical composition and administering the reconstituted pharmaceutical composition to a patient in need thereof. In some embodiments, the reconstituted pharmaceutical composition has similar or the same makeup as the pharmaceutical formulation prior to lyophilization and/or as-prepared. In either case, the pharmaceutical formulation or tituted pharmaceutical composition may be diluted prior to administration to the t. In further embodiments, the ceutical ation or tituted pharmaceutical composition is administered intravenously. id="p-136" id="p-136"

[00136] In one or more embodiments, a composition for intravenous stration is a solution in sterile isotonic aqueous . Where necessary, the composition may also include a solubilizing agent and a local anesthetic to ease pain at the site of the injection. Generally, the ingredients are supplied either separately or mixed together in unit dosage form, for e, as a dry lyophilized powder or water free concentrate in a hermetically sealed container such as an ampule or sachet indicating the quantity of active agent. Where the composition is to be administered by infusion, it can be dispensed with an infusion bottle containing sterile pharmaceutical grade water, saline or dextrose/water. Where the composition is administered by injection, an ampule of sterile water for injection or saline can be provided so that the ingredients may be mixed prior to administration.

In other embodiments, pharmaceutical formulation or reconstituted pharmaceutical composition is administered by direct administration to a target tissue, such as to heart or skeletal muscle (e.g. intramuscular), or nervous system (e.g. direct injection into the brain; intraventricularly; intrathecally). More than one route can be used concurrently, if desired.

The pharmaceutical formulation or reconstituted composition is administered in a therapeutically effective amount (e.g. a dosage amount that, when administered at regular als, is sufficient to treat the disease, such as by ameliorating symptoms associated with the disease, preventing or delaying the onset of the disease, and/or ing the severity or frequency of symptoms of the disease). The amount which will be therapeutically effective in the treatment of the disease will depend on the nature and extent of the disease's effects, and can be determined by standard clinical techniques. In addition, in vitro or in viva assays may optionally be employed to help identify l dosage ranges. In at least one embodiment, the inant human acid a-glucosidase is administered by intravenous infusion at a dose of about about 1 mg/kg to about 100 mg/kg, such as about 5 mg/kg to about 30 mg/kg, lly about 5 mg/kg to about 20 mg/kg. In at least one embodiment, the recombinant human acid a- glucosidase is stered by intravenous infusion at a dose of about 5 mg/kg, about 10 mg/kg, about 15 mg/kg, about 20 mg/kg, about 25 mg/kg, about 30 mg/kg, about 35 mg/kg, about 40 mg/kg, about 50 mg/kg, about 50 mg/kg, about 60 mg/kg, about 70 mg/kg, about 80 mg/kg, about 90 mg/kg or about 100 mg/kg. In at least one ment, the recombinant human acid a-glucosidase is administered by intravenous infusion at a dose of about 20 mg/kg.

The effective dose for a particular individual can be varied (e.g. increased or decreased) over time, depending on the needs of the dual. For e, in times of physical illness or stress, or if anti-acid a-glucosidase antibodies become present or increase, or if disease ms worsen, the amount can be increased.

The therapeutically effective amount of recombinant human acid a-glucosidase (or composition or medicament containing inant human acid a-glucosidase) is administered at regular intervals, depending on the nature and extent of the disease's effects, and on an ongoing basis. Administration at a "regular interval," as used herein, indicates that the therapeutically effective amount is administered periodically (as distinguished from a one— time dose). The interval can be determined by standard clinical techniques. In preferred embodiments, inant human acid osidase is administered monthly, bimonthly; weekly; twice weekly; or daily. The administration al for a single individual need not be a fixed interval, but can be varied over time, depending on the needs of the individual. For example, in times of physical illness or stress, if anti-recombinant human acid a-glucosidase dies become present or increase, or if e symptoms worsen, the al between doses can be decreased.

] In one or more embodiments, the pharmaceutical formulation or reconstituted composition is co-administered with a pharmacological one, such as oral administration of the chaperone and intravenous administration of the pharmaceutical formulation or reconstituted composition. In various embodiments, the pharmacological one is miglustat. In at least one embodiment, the miglustat is administered at an oral dose of about 200 mg to about 400 mg, or at an oral dose of about 200 mg, about 250 mg, about 300 mg, about 350 mg or about 400 mg. In at least one embodiment, the miglustat is administered at an oral dose of about 233 mg to about 400 mg. In at least one embodiment, the miglustat is administered at an oral dose of about 250 to about 270 mg, or at an oral dose of about 250 mg, about 255 mg, about 260 mg, about 265 mg or about 270 mg. In at least one embodiment, the miglustat is administered as an oral dose of about 260 mg.

In at least one embodiment, the miglustat and the recombinant human acid a-glucosidase are administered aneously. In at least one embodiment, the miglustat and the recombinant human acid (x-glucosidase are administered sequentially. In at least one embodiment, the miglustat is administered prior to administration of the recombinant human acid a-glucosidase. In at least one embodiment, the miglustat is administered less than three hours prior to administration of the recombinant human acid a-glucosidase. In at least one ment, the miglustat is administered about two hours prior to administration of the recombinant human acid a—glucosidase. In at least one embodiment, the tat is administered less than two hours prior to administration of the recombinant human acid a- glucosidase. In at least one embodiment, the miglustat is administered about 1.5 hours prior to administration of the recombinant human acid a-glucosidase. In at least one ment, the miglustat is administered about one hour prior to administration of the recombinant human acid a-glucosidase. In at least one embodiment, the miglustat is administered from about 50 minutes to about 70 s prior to administration of the inant human acid (x— glucosidase. In at least one embodiment, the miglustat is administered from about 55 minutes to about 65 minutes prior to administration of the recombinant human acid a—glucosidase. In at least one ment, the miglustat is administered about 30 minutes prior to administration of the recombinant human acid osidase. In at least one embodiment, the miglustat is administered from about 25 minutes to about 35 minutes prior to administration of the recombinant human acid a—glucosidase. In at least one embodiment, the miglustat is administered from about 27 minutes to about 33 minutes prior to administration of the recombinant human acid (z-glucosidase.

] Another aspect of the invention pertains to kits comprising the ceutical formulations described herein (including after lization). In one or more embodiments, the kit comprises a container (e.g., vial, tube, bag, etc.) comprising the pharmaceutical formulations r before or after lyophilization) and instructions for reconstitution, dilution and administration.

EXAMPLES ENZYME EXAMPLE 1: LIMITATIONS OF EXISTING MYOZYME® AND ME® rhGAA PRODUCTS id="p-143" id="p-143"

[00143] To evaluate the ability of the rhGAA in Myozyme® and Lumizyme®, the only currently approved treatments for Pompe disease, these rhGAA preparations were injected onto a CIMPR column (which binds rhGAA having M6P groups) and subsequently eluted with a free M6 nt. Fractions were collected in 96-well plate and GAA activity assayed by 4MU- (x—glucose substrate. The relative amounts of bound and unbound rhGAA were determined based on GAA activity and reported as the fraction of total enzyme.

Figures 3A-B show the problems associated with conventional ERTs (Myozyme® and Lumizyme®): 73% of the rhGAA in Myozyme® (Figure 3B) and 78% of the rhGAA in Lumizyme® (Figure 3A) did not bind to the CIMPR, see the left-most peaks in each figure. Only 27% of the rhGAA in Myozyme® and 22% of the rhGAA in Lumizyme® contained M6P that can productive to target it to the CIMPR on muscle cells.

] An ive dose of Myozyme® and me® corresponds to the amount of rhGAA containing M6P which targets the CIMPR on muscle cells. r, most of the rhGAA in these two conventional products does not target the CIMPR or on target muscle cells. The administration of a conventional rhGAA where most of the rhGAA is not targeted to muscle cells increases the risk of allergic reaction or induction of immunity to the non-targeted rhGAA.

ENZYME EXAMPLE 2: ATION OF CHO CELLS PRODUCING ATB200 rhGAA HAVING A HIGH CONTENT OF MONO- OR BIS—M6P-BEARING N-GLYCANS id="p-146" id="p-146"

[00146] CHO cells were transfected with DNA that expresses rhGAA followed by selection of ormants producing rhGAA. A DNA construct for transforming CHO cells with DNA encoding rhGAA is shown in Figure 4. CHO cells were transfected with DNA that expresses rhGAA followed by selection of transformants producing rhGAA.

After transfection, DG44 CHO (DHFR-) cells containing a stably integrated GAA gene were selected with hypoxanthine/thymidine deficient (-HT) medium).

Amplification of GAA expression in these cells was induced by rexate treatment (MTX, 500 nM). Cell pools that expressed high amounts of GAA were identified by GAA enzyme ty assays and were used to establish individual clones producing rhGAA. Individual clones were generated on semisolid media plates, picked by ClonePix system, and were transferred to 24-deep well plates. The individual clones were assayed for GAA enzyme activity to identify clones expressing a high level of GAA. Conditioned media for determining GAA activity used a -Glucosidase substrate. Clones producing higher levels of GAA as measured by GAA enzyme assays were further evaluated for viability, ability to grow, GAA productivity, N-glycan structure and stable protein expression. CHO cell lines, including CHO WO 73060 cell line GA-ATB-ZOO, expressing rhGAA with enhanced mono—M6P or bis-M6P N-glycans were isolated using this ure.

ENZYME EXAMPLE 3: CAPTURING AND PURIFICATION OF ATBZOO rhGAA Multiple batches of the rhGAA according to the invention were produced in shake flasks and in perfusion bioreactors using CHO cell line GA-ATB-200 and CIMPR binding was measured. r CIMPR receptor binding (~70%) to that shown in Figure 5B and Figure 6 was observed for purified ATB200 rhGAA from different production batches indicating that ATBZOO rhGAA can be tently produced. As shown by Figures 3A-B and 5A-B, Myozyme® and Lumizyme® rhGAAs ted significantly less CIMPR binding than ATB200 rhGAA.

ENZYME EXAMPLE 4: ANALYTICAL COMPARISON OF ATB200 TO LUMIZYME® Weak anion exchange ("WAX") liquid chromatography was used to fractionate ATBZOO rhGAA according to terminal phosphate. n profiles were generated by eluting the ERT with increasing amount of salt. The es were monitored by UV (A280nm).

ATB200 rhGAA was obtained from CHO cells and purified. Lumizyme® was obtained from a commercial source. Lumizyme® exhibited a high peak on the left of its elution profile.

ATB200 rhGAA exhibited four prominent peaks eluting to the right of Lumizyme® e 7).

This confirms that ATB200 rhGAA was phosphorylated to a greater extent than Lumizyme® since this evaluation is by terminal charge rather than CIMPR affinity.

ENZYME EXAMPLE 5: ACCHARIDE CHARACTERIZATION OF ATBZOO rhGAA Purified ATB200 rhGAA and Lumizyme® glycans were evaluated by MALDI- TOF to determine the individual glycan structures found on each ERT (Figure 8). ATB200 samples were found to contain lower amounts of non-phosphorylated high-mannose type N- glycans than Lumizyme®. The higher content of M6P glycans in ATB200 than in Lumizyme®, targets ATB2OO rhGAA to muscle cells more effectively. The high tage of mono-phosphorylated and bis-phosphorylated structures determined by MALDI agree with the CIMPR es which illustrated significantly greater binding of ATBZOO to the CIMPR receptor. N—glycan analysis via MALDI-TOF mass spectrometry confirmed that on average each ATB200 le contains at least one natural bis-M6P N-glycan structure. This higher bis-M6P N-glycan content on ATB200 rhGAA directly correlated with high-affinity binding to CIMPR in M6P receptor plate binding assays (KD about 2—4 nM) Figure 10A.

ATB200 rhGAA was also ed for pecific N-glycan profiles using two different LC-MS/MS analytical techniques. In the first analysis, the protein was denatured, reduced, alkylated and digested prior to LC-MS/MS analysis. During protein denaturation and reduction, 200 ug of protein sample, 5 [LL 1 mol/L tris-HCl (final concentration 50mM), 75 [LL 8 mol/L guanidine HCl (final concentration 6 M), 1 HL 0.5 mol/L EDTA (final concentration 5 mM), 2 [1L 1 mol/L DTT (final concentration 20 mM) and Milli-Q® water were added to a 1.5 mL tube to provide a total volume of 100 uL. The sample was mixed and incubated at 56°C for minutes in a dry bath. During alkylation, the denatured and d protein sample was mixed with 5 uL 1 mol/L iodoacetamide (IAM, final concentration 50 mM), then incubated at C in the dark for 30 minutes. After alkylation, 400 uL of precooled acetone was added to the sample and the e was frozen at —80°C refrigeration for 4 hours. The sample was then centrifuged for 5 min at 13000 rpm at 4°C and the supernatant was removed. 400 uL of precooled acetone was added to the pellets, which was then fuged for 5 min at 13000 rpm at 4°C and the supernatant was removed. The sample was then air dried on ice in the dark to remove acetone residue. 40 [LL of 8M urea and 160 "L of 100 mM 3 were added to the sample to dissolve the protein. During trypsin digestion, 50 ug of the protein was then added with trypsin digestion buffer to a final volume of 100 uL, and 5 [LL 0.5 mg/mL trypsin (protein to enzyme ratio of 20/1 w/w) was added. The solution was mixed well and incubated overnight (16 i 2 hours) at 37°C. 2.5 "L 20% TFA (final concentration 0.5%) was added to quench the reaction. The sample was then ed using the Thermo Scientific Orbitrap Velos ProTM Mass Spectrometer.

In the second LC-MS/MS analysis, the ATB200 sample was prepared according to a similar denaturation, reduction, alkylation and digestion procedure, except that iodoacetic acid (1AA) was used as the alkylation reagent d of IAM, and then analyzed using the Thermo Scientific Orbitrap Fusion Lumos TribidTM Mass Spectrometer.

] The results of the first and second analyses are shown in Figures 9A-9H. In Figures 9A-9H, the s of the first analysis are represented by left bar (dark grey) and the s from the second analysis are represented by the right bar (light grey). In Figures 9B-9G, the symbol lature for glycan representation is in accordance with Varki, A., Cummings, R.D., Esko J.D., et al., ials of Glycobiology, 2nd edition (2009).

] As can be seen from Figures 9A-9G, the two analyses provided similar results, although there was some variation between the results. This variation can be due to a number of factors, including the instrument used and the completeness of N—glycan analysis. For example, if some species of phosphorylated glycans were not identified and/or not quantified, then the total number of phosphorylated glycans may be underrepresented, and the percentage of rhGAA g the phosphorylated glycans at that site may be underrepresented. As another example, if some species of non-phosphorylated glycans were not identified and/or not quantified, then the total number of non-phosphorylated glycans may be underrepresented, and the tage of rhGAA bearing the phosphorylated glycans at that site may be overrepresented. Figure 9A shows the N-glycosylation site occupancy of ATB200. As can be seen from Figure 9A, the first, second, third, fourth, fifth and sixth N-glycosylation sites are mostly occupied, with both analyses detecting over 90% and up to about 100% of the ATB200 enzyme having a glycan detected at each potential site. However, the seventh potential N— glycosylation site is glycosylated about half of the time.

Figure 9B shows the N-glycosylation profile of the first site, N84. As can be seen from Figure 9B, the major glycan species is bis—M6P glycans. Both the first and second analyses detected over 75% of the ATB200 had a bis—M6P glycan at the first site. id="p-157" id="p-157"

[00157] Figure 9C shows the N-glycosylation profile of the second site, N177. As can be seen from Figure 9C, the major glycan species are mono-M6P glycans and non— phosphorylated high mannose glycans. Both the first and second analyses detected over 40% of the ATB200 had a mono-M6P glycan at the second site.

Figure 9D shows the N-glycosylation profile of the third site, N334. As can be seen from Figure 9D, the major glycan species are osphorylated high e glycans, di-, tri-, and tetra—antennary complex glycans, and hybrid glycans. Both the first and second analyses detected over 20% of the ATB200 had a sialic acid residue at the third site.

Figure 9B shows the N—glycosylation profile of the fourth site, N414. As can be seen from Figure 9E, the major glycan species are bis-M6P and mono-MGP s. Both the first and second es detected over 40% of the ATB200 had a bis-M6P glycan at the fourth site. Both the first and second analyses also detected over 25% of the ATB200 had a mono— M6P glycan at the fourth site.

Figure 9F shows the N-glycosylation profile of the fifth site, N596. As can be seen from Figure 9F, the major glycan species are fucosylated di-antennary x glycans.

Both the first and second analyses detected over 70% of the ATB200 had a sialic acid residue at the fifth site.

Figure 9G shows the N-glycosylation profile of the sixth site, N826. As can be seen from Figure 9G, the major glycan species are di-, tri—, and tetra—antennary complex glycans. Both the first and second analyses detected over 80% of the ATB200 had a sialic acid residue at the sixth site.

] An analysis of the glycosylation at the seventh site, N869, showed approximately 40% glycosylation, with the most common glycans being A4S3S3GF (12%), A5S3G2F (10%), A4S2G2F (8%) and F (8%).

Figure 9H shows a summary of the phosphorylation at each of the seven potential N-glycosylation sites. As can be seen from Figure 9G, both the first and second analyses ed high phosphorylation levels at the first, second and fourth sites. Both analyses detected over 80% of the ATB200 was mono— or di—phosphorylated at the first site, over 40% of the ATB200 was mon-phosphorylated at the second site, and over 80% of the ATB200 was mono- or di-phosphorylated at the fourth site.

Another glycan analysis of ATB200 was med according to a hydrophilic ction liquid chromatography-fluorescent detection-mass spectrometery (HILIC-FLD-MS) method.

The results of HILIC-FLD-MS analysis are provided in Table A below. In Table A, the first number in the three—digit number indicates the number of branches in the glycan, the second number indicates the number of core fucose units and the third number indicates the number of terminal sialic acid units. Using this nomenclature, "303" represents a tri-antennary glycan (the first 3) with 0 core fucose (the 2nd 0) and 3 terminal sialic acids (the last 3), "212" represents a ennary glycan with 1 core fucose and 2 terminal sialic acids, "404" represents a tetra-antennary glycan with 0 core fucose and 4 terminal sialic acids, etc.

Table A FLD MS Peak Peak RT (min) Glycan Structure Number Number 1 1 BisP_Man 8 2.83% BisP_Man 7 BisP_Man 6 M0n0P_Man 6 Mon0P_Man 5 MonoP_Man 8 MonoP_Man 7 MonoP_Man 7 _(+)GlcNAc Mon0P_hMan6_101 M0n0P_Man 6_(+)GlcNAc Man 6 hMan 5_101 hMan 5_100_(-)Gal 300_(-)Gal 301_(-)Ga1 301_(-)Gal hMan6_1 1 1 M0n0P_hMan6_1 1 1 MonoP_Man 6 _1 10_(-)Gal 201_(-)Gal 44 52.42 201 1.18% 57.45,58.34 414_(+)GlcNAcGal 56.62,56.91,57.99 56.11,57.26,57.99 413_(+)GlcNAcGal 413_(+)GlcNAcGal 63.18,64.32 413_(+)GlcNAcGa1 63.5, 64.36 3 1 1_(-)Ga1 65.73, 66.20 65.85, 66.49 310_(—)Gal 412_(+)G1cNAcGa1 412_(+)GlcNAcGal 413_(+)2(GlcNAcGal) 42 76 70.66 412_(+)GlcNAcGa1 0.73% 43 77 71.74 21 1 1.09% 78 71.23 211_(-)Gal 0.19% 44 79 72 46 212 3.66% 45 80 74.82 221_(-)Ga1(+)GalNAc 0.38% 81 74.43 ,74.96 41 1_(+)GlcNAcGa1 0.66% 46 82 75.92 410 0.42% 47 83 76.73,77.87 211_(-)Gal 1.24% 84 77.23 211 3.64% 48 85 79.05 211 1.52% 86 79.38 210_(-)2Gal 0.45% 49 87 80.11 210_(-)Gal 1.58% 50 88 81.15 210 2.41% 51 89 84.22-87.15 311 1.26% 52 90 95.35 Mono_Acetyl_NANA_2l2 0.99% 53 91 96.23 Mono_Acetyl_NANA_21 1 0.76% 54 92 97.37 Bis_Acetyl_NANA_212 0.42% Based on this HILIC-FLD-MS analysis, the ATE200 tested is expected to have an average fucose content of 2-3 mol per mol of ATB200, GlcNAc content of 20-25 mol per mol of ATB200, galactose content of 8—12 mol per mol of ATB200, mannose content of 22-27 mol per mol of , M6P content of 3-5 mol per mol of ATB200 and sialic acid content of 4-7 mol of ATB200.

ENZYME E 6: CHARACTERIZATION OF CIMPR AFFINITY OF ATB200 ] In addition to having a r percentage of rhGAA that can bind to the CIMPR, it is important to understand the quality of that interaction. Lumizyme® and ATB200 rhGAA receptor binding was determined using a CIMPR plate binding assay. Briefly, CIMPR- coated plates were used to capture GAA. Varying concentrations of rhGAA were applied to the immobilized receptor and unbound rhGAA was washed off. The amount of remaining rhGAA was determined by GAA activity. As shown by Figure 10A, ATB200 rhGAA bound to CIMPR significantly better than Lumizyme®.

Figure 10B shows the relative content of bis-M6P glycans in Lumizyme®, a conventional rhGAA, and ATB200 according to the ion. For Lumizyme® there is on average only 10% of molecules have a bis—phosphorylated . Contrast this with ATB200 where on average every rhGAA molecule has at least one bis-phosphorylated glycan.

ENZYME EXAMPLE 7: ATB200 rhGAA WAS MORE EFFICIENTLY ALIZED BY FIBROBLAST THAN LUMIZYME® The relative cellular uptake of ATB200 and Lumizyme® rhGAA were compared using normal and Pompe fibroblast cell lines. Comparisons involved 5—100 nM of ATB200 rhGAA according to the invention with 10-500 nM tional rhGAA Lumizyme®. After 16-hr incubation, external rhGAA was inactivated with TRIS base and cells were washed 3-times with PBS prior to harvest. Internalized GAA measured by 4MU-a- Glucoside hydrolysis and was graphed ve to total cellular protein and the results appear in Figures 11A—B.

ATBZOO rhGAA was also shown to be ently internalized into cells (Figure 11A and 11B), respectively, show that ATB200 rhGAA is internalized into both normal and Pompe fibroblast cells and that it is internalized to a greater degree than tional Lumizyme® rhGAA. ATBZOO rhGAA saturates cellular receptors at about 20 nM, while about 250 nM of Lumizyme® is needed. The uptake efficiency constant (Kuplake) extrapolated from these results is 2-3 nm for ATB200 and 56 nM for Lumizyme® as shown by Figure 11C.

These results suggest that ATBZOO rhGAA is a well-targeted treatment for Pompe e.

ENZYME EXAMPLE 8: GLYCOGEN REDUCTION IN GAA-KNOCKOUT MICE ] Figures 12A to 12C show the effects of administering alglucosidase alfa (Lumizyme®) and ATB2OO on glycogen clearance in Gaa knockout mice. s were given two IV bolus administrations (every other week); tissues were harvested two weeks after the last dose and analyzed for acid a—glucosidase ty and en content.

As seen from Figures 12A to 12C, ATB200 was found to deplete tissue glycogen in acid a—glucosidase (Gad) knockout mice in a dose-dependent fashion. The 20 mg/kg dose of ATB200 consistently removed a greater proportion of stored glycogen in Gaa knockout mice than the 5 and 10 mg/kg dose levels. However, as seen in s 12A to 12C, ATB200 stered at 5 mg/kg showed a similar reduction of glycogen in mouse heart and skeletal muscles (quadriceps and s) to Lumizyme® administered at 20 mg/kg, while ATB200 dosed at 10 and 20 mg/kg showed significantly better reduction of glycogen levels in skeletal muscles than Lumizyme®.

Figure 15 shows the effects of administering alglucosidase alfa (Lumizyme®) and ATB200 on glycogen clearance in Gaa knockout mice. Twelve week old GAA KO mice treated with Lumizyme® or ATBZOO, 20 mg/kg IV every other week 4 injections. Tissues were ted 14 days after last enzyme dose for glycogen measurement. Figure 13 shows the relative reduction of glycogen in quadriceps and triceps skeletal muscle, with ATBZOO providing a greater reduction of glycogen than Lumizyme®.

ENZYME EXAMPLE 9: MUSCLE PHYSIOLOGY AND LOGY IN GAA— KNOCKOUT MICE Gaa knockout mice were given two IV bolus administrations of recombinant human acid (x-glucosidase (alglucosidase alfa or ATBZOO) at 20 mg/kg every other week.

Control mice were treated with vehicle alone. Soleus, quadriceps and diaphragm tissue is harvested two weeks after the last dose of recombinant human acid a-glucosidase. Soleus and diaphragm tissue were analyzed for glycogen levels, by staining with ic acid — Schiff reagent (PAS), and for lysosome proliferation, by ing levels of the lysosome-associated membrane protein (LAMPl) marker, which is upregulated in Pompe disease. hin sections of quadriceps muscle ed in epoxy resin (Epon) were d with methylene blue and observed by electron copy (lOOOX) to determine the extent of the presence of vacuoles. ceps muscle samples were analyzed immunohistochemically to determine levels of the autophagy markers microtubule-associated protein lA/lB-light chain 3 phosphatidylethanolamine conjugate (LC3A II) and p62, the insulin-dependent glucose transporter GLUT4 and the n-independent glucose transporter GLUTl.

In a similar study, Gaa knockout mice were given four IV bolus administrations of recombinant human acid cosidase (alglucosidase alfa or ATB200) at 20 mg/kg every other week. Control mice were treated with vehicle alone. Cardiac muscle tissue was harvested two weeks after the last dose of recombinant human acid a-glucosidase and analyzed for glycogen levels, by staining with periodic acid — Schiff reagent (PAS), and for lysosome proliferation, by measuring levels of LAMP].

As seen in Figure 14, administration of ATB200 showed a reduction in lysosome proliferation in heart, diaphragm and skeletal muscle (soleus) tissue compared to conventional treatment with alglucosidase alfa. In addition, as seen in Figure 15, administration of ATB200 showed a reduction in punctate glycogen levels in heart and skeletal muscle (soleus) tissue compared to tional treatment with alglucosidase alfa.

As well, as seen in Figure 16, ATB200 significantly reduced the number of vacuoles in muscle fiber in the quadriceps of Gaa knockout mice ed to untreated mice and mice treated with alglucosidase alfa. As seen in Figure 17, levels of both LC3 II and p62 are increased in Gaa knockout mice compared to wild type mice. In addition, levels of the insulin—dependent glucose transporter GLUT4 and the n-independent glucose transporter GLUTl are increased in Gaa knockout mice compared to wild type mice. The elevated GLUT4 and GLUTl levels associated with acid (x—glucosidase deficiency can contribute to increased e uptake into muscle fibers and increased glycogen synthesis both basally and after food intake.

ENZYME EXAMPLE 10: MUSCLE FUNCTION IN GAA-KNOCKOUT MICE In longer-term studies of 12 biweekly administrations, 20 mg/kg ATBZOO plus mg/kg miglustat progressively increased functional muscle strength in Gaa KO mice from baseline as measured by both grip strength and wire hang tests (Figures 18A-18B). osidase alfa (Lumizyme®)-treated mice ing the same ERT dose (20 mg/kg) were ed to decline under identical conditions throughout most of the study (Figures 18A— 18B). As with the shorter-term study, ATB200/miglustat had ntially better glycogen clearance after 3 months (Figures 19A-19C) and 6 months (Figures 19D—19G) of treatment than alglucosidase alfa. ATBZOO/miglustat also reduced autophagy and intracellular accumulation of LAMP] and dysferlin after 3 months of treatment (Figure 20) compared to alglucosidase alfa. In Figure 18A, * indicates statistically significant compared to Lumizyme® alone (p<0.05, 2-sided ). In Figures 19A-19G, * indicates statistically significant compared to Lumizyme® alone (p<0.05, multiple comparison using t’s method under one-way ANOVA analysis).

] Taken together, these data indicate that ATBZOO/miglustat was efficiently targeted to muscles to reverse ar dysfunction and e muscle function. Importantly, the apparent improvements in muscle architecture and reduced autophagy and intracellular accumulation of LAMPl and dysferlin may be good surrogates for improved muscle physiology that correlate with improvements in functional muscle strength. These results suggest that monitoring autophagy and these key muscle proteins may be a al, practical method to assess the effectiveness eutic treatments for Pompe disease in Gaa KO mice that may prove to be useful biomarkers from muscle biopsies in clinical studies.

Figure 20 shows that 6 months of ATB200 administration with or without miglustat lowered ellular accumulation of dystrophin in Gaa KO mice. There was a greater reduction for dystrophin accumulation for ATB200 i miglustat than with Lumizyme®.

FORMULATION EXAMPLE 1: PH AND BUFFER Analytical Methods for Formulation Examples Analyses of the examples described herein with respect to appearance, pH, n concentration, etc. were carried out according to the below methods, unless otherwise Appearance ] The appearance of samples, including clarity, color, and visible particles, were examined under black and white background using YB-2 lightbox.

Sample pH was measured with a SevenMultiTM pH Meter.

Protein Concentration id="p-184" id="p-184"

[00184] Protein concentration was determined by UV280 readings using a NanoDropTM 2000 spectrophotometer. All measurements were repeated twice with 2.5 uL sample each time and an average was taken.

Size ion High Performance Liguid Chromatography [SEC—HPLC] For the pH and buffer, freeze-thaw, and excipient examples, separation of protein monomers and its high molecular weight species and fragments was performed using a TSngl® G3000 SWXL column (Tosoh Bioscience, 0 mm, Sum, 25 °C) on an Agilent 1260 HPLC system. The mobile phase consisted of 50 mM sodium phosphate, 100 mM sodium chloride and 20 leI sodium citrate (pH 6.0i0.2). The chromatographic system employed a flow rate of 1.0 ml/min, 50-uL injection volume ally 1 mg/mL), and 20-min run time with an isocratic nt. Signals were detected by a UV detector at 280 nm (Reference: 360 nm).

In the PS80 example, separation of protein monomers and its high molecular weight species and fragments was performed using a BioSepTM-SEC-SSOOO column (Phenomenex, 7.8x300 mm, 5pm, 25°C) on an Agilent 1260 HPLC system. The mobile phase ted of 50 mM sodium phosphate and 100 mM sodium de (pH 6.2i0.2). The chromatographic system employed a flow rate of 1.15 ml/min, 50—uL ion volume (typically 1 mg/mL), and 25-min run time with a isocratic gradient. Signals were detected by a UV detector at 280 nm.

SDS-PAGE g Non-reduced 1 The able values are the purity and molecular weight of protein in non— reducing SDS-PAGE. Samples were red non-reduced in the presence of excessive SDS to attain a uniform negative charge. In an applied electric field (165V), these SDS-coated species were separated based on their apparent molecular weight through polyacrylamide gel.

The separated bands were detected by sie Blue staining. 4—MUG Enzyme Activity ] 10 ul of sample was diluted and hydrolyzed (by GAA, 37°C for 60 min) to generate fluorescent product 4—MU. 125 ul of 1 M glycine or 0.1 M NaOH was added in to stop the reaction.

] A series of 4-MU standards were analyzed with samples to generate a standard calibration curve based on cence signal. The conversion of RFU to 4-MU amount was achieved by a re mediated comparison to a standard curve, which was regressed according to 4 parameter logistic regression model. Then GAA enzyme activity (nmol 4-MU released/hr/ml GAA) in sample was calculated based on the 4-MU standard curve. 4—MUG tic Concentration 10 ul of sample and GAA reference stande were diluted and hydrolyzed (by GAA, 37°C for 60 min) to generate cent product 4-MU. 125 ul of 1 M NaOH was added in to stop the reaction. id="p-191" id="p-191"

[00191] A series of GAA reference standards were analyzed with samples to generate a standard calibration curve based on fluorescence signal. The conversion of RFU to 4—MU amount was achieved by a software mediated comparison to a standard curve, which was regressed according to 4 parameter logistic regression model. Then GAA enzyme tration (nmol 4—MU released/hr/ml GAA) in the sample was calculated based on the GAA standard curve.

Dynamic Light Scattering [DLS] A micropipette was used to transfer an aliquot of 40 [LL of undiluted sample to a 40 [LL disposable cuvette. cate measurements were performed for each sample.

Particles: HIAC id="p-193" id="p-193"

[00193] 200 ul of each sample was diluted into 2000 ul with a filtered reference buffer.

The sample was tested three times and 450 ul was used for each test. Average number of particles of each size, lum, 3 pm, 5 um, 10 um, and 25 um per ml, were reported.

MicroCal Differential Scanning Calorimetry [DSC] The capillary cell differential scanning calorimetry (DSC) is ed to measure the thermal stability of proteins by detecting the ence in the amount of heat required to increase the temperature of a sample and reference as a function of temperature. Specifically, it is used to measure the thermal transition midpoint (Tm), which is an tor of the relative stability of protein in solution.

Samples were diluted to about 1 mg/mL with reference . An aliquot of 400 [LL of reference buffer was added into each odd—numbered well of a 96-well plate while an aliquot of 400 [LL of each sample was added into the corresponding even-numbered well. The scanning temperature ranges from 10°C to 110°C.

Modulated ential Scanning Calorimetry (mDSC) The glass transition (Tg’) temperature and eutectic temperature (Te) were tested using Netzsch ential Scan Calorimeter (DSC 204 F1). 15 "L of sample was loaded into loading disc for testing. Firstly, the temperature was decreased from 20 0C to -60 °C at the rate of 10 °C /min, and the Te value was obtained from the cooling curve during this step. ly, the temperature was increased from -60 CC to 40 °C at the rate of 10 CC /min, and the Tg’ value was analyzed from the heating curve.

M6P was released from sample by ysis (4 M TFA, 100 CC, 4h) and dried by centrifugal vacuum evaporator. The dried M6P and reference standard were suspended in purified water prior to analysis. A CarboPac PA10 BioLCTM ical column (4 mm x 250 mm, 3.5 pm, 100 A, 30°C) and a CarboPac PA10 BioLCGuard column (4 mm x 50 mm, 3.5 um, 100 A, 30 °C) were used. The mobile phase consisted of phase A (100 mM NaOH) and phase B (1 M NaOAc, 100 mM NaOH). The chromatographic system employed a flow rate of 1 ml/min, 25-uL injection volume, and 30-min run time with a gradient. Signals were detected by a pulsed metric detection. The M6P content in sample was calculated based on the standard curve.

Sialic Acid The sialic acids were released from drug molecules by hydrolysis (2 M HAc, 80°C,2h), and then labeling all samples and mixed standard solution with DMB (50 °C,17iO.5 h in dark) prior to be separated using a Zorbax Eclipse Plus C18 column (Agilent, 4.6 mm><100 mm, 3.5 um, 45 CC ) on Agilent 1260 HPLC system. The mobile phase consisted of phase A (9% ACN, WO 73060 7% MeOH) and phase B (100% ACN). The chromatographic system employed a flow rate of 0.5 , 20 uL injection volume, and 20 min run time with a gradient. Signals were detected by a fluorescence detector (lex=373 nm, lem=448 nm). Neu5Gc and Neu5Ac content in sample were calculated based on a standard curve.

] Ten buffer formulations were prepared having a pH ranging from 4.0 to 8.0, containing 25 leI Sodium Phosphate or 25 mM Sodium Citrate, with or without 50 mM NaCl, as shown in Table 1. The ATB200 4-MUG enzymatic concentration was 1 mg/mL.

Table 1 Formulation Compositions: Buffer NaCl lfi P40 40* P50 _ 5.0 P60 fiiflastgdwm 50 mM 6.0 P7 7.0 P80 8.0 P60 without 25 mM Sodium None 6 0 NaCl Phos hate ‘ C50 5.0 $23 25 mM Sodium Citrate 50 mM —23 C65 6.5 * HCl was used to adjust the pH to 4.0.

Sample ation ] The material used in the pH and buffer evaluation study is as bed in Table 2 below.

Table 2 Raw Material Information of H and Buffer Evaluation Stud : C0110, mg/ml Enzyme Sialic Acid, M6P, activity, SDS-PAGE, % mol/mol mol/mol U/L protein protein Conc.

* The buffer of this ATB200 enzyme solution is 50 mM sodium phosphate (pH 6.2), 50 mM NaCl, 2% mannitol ** The extinction coefficient was 1.51 AU"mL*mg'1*cm‘1 25 mM Sodium Phosphate buffer containing 50 mM NaCl at pH 4.0 (P40), 5.0 (P50), 6.0 (P60), 7.0 (P70), and 8.0 (P80), 25 mM Sodium Phosphate buffer at pH 6.0 (P60 without NaCl), and 25 mM Sodium Citrate buffer containing 50 mM NaCl at pH 5.0 (C50), 5.5 (C55), 6.0 (C60) and 6.5 (C65) were prepared. For pH 4.0, HCl was used to adjust the pH in the Sodium Phosphate buffer (pH 4.0).

The ATB200 enzyme solution was concentrated firstly using ultra-filtration centrifugal s under the condition of 15°C and 3500 rpm for 40 min. After that, concentrated enzyme on was buffer—exchanged into the three different buffers described above by two rounds of ultra-filtration at 15°C and 3500 rpm for 50 min and 55 min. Buffer- exchanged enzyme solutions were subsequently analyzed for protein concentration and 4- MUG enzyme concentration.

Finally, appropriate volume of each buffer was added to adjust the final ATB200 enzyme concentration to 1.0 . The final ATB200 concentration was ed by both UV_A280 ance and 4—MUG enzyme concentration.

The solutions were aseptically filtered with 0.22-um polyethersulfone (PES) filter.

Each formulation was aseptically filled into 2-mL glass vials in a bio-safety hood with the filling volume of 500 uL ~ 1000 [LL according to the sample amount requirement of analytical tests. Vials were stoppered and then crimp-oversealed innnediately after filling.

Sample Testing Vials of each formulation were stored at 5°C and 40°C for up to 8 weeks, and ed at 25°C at 100 rpm on an orbital shaker for up to 5 days (See Table 3). The formulations were sampled initially (T0), 5 days (5D), 2 weeks (2W), 4 weeks (4W) and 8 weeks (8W), as described in Table 3, for the tests of appearance, pH, UV concentration, 4— MUG enzymatic tration, SEC, HIAC, DLS, 4-MUG enzyme activity, and DSC.

Table 3 Stud Parameters in ATB200 H and Buffer Evaluation Stud‘ X Y _--- WWW: n..- X: Appearance, pH, UV Concentration, 4-MUG Enzymatic Concentration, SEC, HIAC*, DLS, 4-MUG Enzyme activity Y: DSC *: HIAC was only tested for T0 s and agitation samples Results ] Thermal Stability Results - MicroCal DSC The results of the thermal stability measurements are shown below in Table 4: Table 4: sam91e TM] (°c) onset (°C) P40 56.34 66.25 P50 63.49 73.42 P60 60.47 72.05 P70 47.55 64.09 P80 37.21 47.72 P60 t NaCl 61.22 72.25 C50 62.66 72.60 C55 62.89 73.39 C60 59.26 70.83 C65 53.45 67.24 Higher t indicates better thermal stability of the protein in the particular formulation. Accordingly, the ations exhibiting the highest thermal stability were P50, C55, C50, P60 without NaCl, P60 and C60. These results indicate that ATB200 enzyme has better thermal stability in weak acidic buffer than in basic condition. id="p-212" id="p-212"

[00212] Appearance - Agitation The results of the appearance of the agitation study are shown below in Table 5: Table 5: Sam le Agitation, 100 ppm, °C, 5 days ess, clear, tiny Visible particles Colorless, clear, very tiny Visible particles Colorless, clear, P60 free of Visible Colorless, clear, very tiny particles Visible particles ess, clear, tiny Visible particles Colorless, Clear, Visible particles sam 16 Agitationz 100 rpm, °C, 5 days P60 without Colorless, clear, free of NaCl Visible particles C50 Colorless, clear, free of e particles C55 Colorless, clear, free of Visible les C60 Colorless, clear, free of Vis1ble particles C65 Colorless, clear, free of Visible particles As can be seen from the table, after 5—day agitation, P60 without NaCl, C50, C55, C60, and C65 maintained colorless, clear, and free of Visible particles, but Visible particles were observed in P40, P50, P60, P70, and P80. This data shows that sodium citrate buffer ized the formulation better than sodium phosphate buffer after agitation.

The results of the pH measurements are shown below in Table 6: Table 6: Agitation, 100 rpm, 25 CC, As seen from the data, both phosphate and citrate buffers with 50 mM NaCl ranged from pH 5.0 to 6.5 were able to maintain designated pH throughout. During buffer exchange, concentration adjustment, 8-week storage at 5°C and 40°C, and ion, there was no significant change in the pH of the samples.

By contrast, pH decreased in P60 without NaCl after 8-week storage at both °C and 40°C, and pH increased in P40 after buffer exchange and 8-week storage at both temperatures. However, agitation did not lead to any change of pH in both P40 and P60 t NaCl. id="p-219" id="p-219"

[00219] Protein Concentration The results of the protein concentration measurements are shown below in Table 7: Table 7: UV Concentration, mg/mL °C 40°C 5am 1° gm.

T0 4L100 m — 2w 4w 8w 2w 8W* — — — — 4_W 8_w °C,5days P40 1.71 1.88 1.70 1.69 1.50 1.73 P50 1.81 1.81 1.91 1.99 1.98 1.56 1.80 P601.88 1.87 1.96 2.10 2.04 1.31 1.87 P70 1.73 1.83 1.75 1.72 1.66 1.74 P80 2.21 2.20 2.25 2.21 2.23 2.19 2.25 without 1.82 1.85 1.82 1.81 1.87 1.90 1.92 1.58 1.86 C50 1.81 1.81 1.80 1.76 1.72 1.71 1.80 C55 1.83 1.84 1.89 1.85 1.79 1.54 1.82 €601.97 1.97 2.08 2.21 2.14 1.19 1.96 C65 2.00 2.17 2.19 2.39 1.26 2.06 id="p-221" id="p-221"

[00221] 8—week e at 5°C and agitation for 5 days did not affect the n concentration. No significant ion was observed among all formulations.

] By contrast, during storage at 40°C, protein concentration slightly increased in P50, P60, P60 without NaCl, C60, and C65, slightly decreased in C50, and maintained in P40, P70, P80, C55. e of the formation of visible particles in 40°C samples, 40°C/ 8-week samples were re-tested after centrifugation at 12000 rpm for 1 minute. The results showed a drop in protein concentration after centrifugation in the samples containing les, which indicated that the particles present in the samples had an impact on the adsorption at 280nm. ] 4—MUG Enzymatic Concentration The results of the 4-MUG enzymatic concentration measurements are shown below in Table 8: Table 8: 4-MUG Enzymatic Concentration, mg/mL °C 40°C Agitation, Sample Name lOOrpm, 2w 4w 8w 2w 4w 8W 25°C,5 P40 0.86 0.91 0.91 0.76 0.34 0.16 0.04 1.02 P501.08 1.05 0.78 0.62 0.33 1.12 P60 _1.13 1.13 0.92 0.87 0.58 1.14 P70 0.92 0.86 0.00 0.00 0.00 1.00 P80 0.00 0.00 0.00 BQL NA 0.00 C501.06 1.05 0.89 0.68 0.33 1.14 C55 _1.11 1.09 1.03 0.96 0.63 1.13 C601.33 1.23 1.08 0.92 0.53 1.26 C65 1.02 1.23 1.29 1.28 0.15 0.01 0.00 1.29 After buffer ge, the 4-MUG enzymatic concentration of P80 dropped to 0.27 mg/InL immediately. This indicated that pH 8.0 ed the enzyme activity significantly.

After 2-week e at both 5°C and 40°C, the enzymatic concentration dropped to zero.

Except for P80, after 8-week storage at 5°C, the enzymatic trations of most formulations were stable and dropped a little in P40.

By contrast, 40°C storage affected the enzymatic concentration obviously. After 2 weeks, the enzymatic concentration was gone in P70, dramatically dropped in P40 and C65, and decreased obviously in P50. After 4 weeks, the enzymatic concentration continued to decrease. Finally, after 8 weeks, the enzymatic trations were close to zero in P40 and C65, dropped to 0.33 mg/mL in pH 5.0 buffers (P50 and C50) and to 0.5~0.7 mg/mL in pH 5.5 ~ 6.0 buffers (P60, P60 without NaCl, C55, and C60). Among them, P60 without NaCl had a highest enzymatic concentration (0.73 mg/mL) y. The results indicate that the enzyme concentration was most preserved with respect to 4—MUG enzymatic concentration in the range of pH 5.5 ~ 6.0, but there was no significant difference between sodium phosphate buffer and sodium citrate buffer.

During the agitation study, except P80, the enzymatic concentration did not significantly change. 4—MUG Enzyme Activity The results of the 4—MUG enzyme activity measurements are shown below in Table 9: Table 9: 4-MUG Enzyme Activity, U/L 1? 100 rpm, 2W 4W 8W 2W 4W 8W 25°C 5 P40 . 5092.7 4153.7 3697.2 1822.3 712.2 211.5 5439.9 P50 . 6021.5 4947.8 5101.4 4344.4 2836.9 1561.7 5953.8 P60 . 6133.7 5193.7 5487.5 5170.4 4002.9 2790.4 6087.0 P70 . 5233.1 4211.5 4195.4 0.2 0.4 0.0 5360.5 P80 . 1.8 1.6 0.7 0.2 BQL 0.0 m without . 5120.0 5662.4 5799.6 C50 . 5110.8 4964.9 6057.4 C55 . 6450.1 5091.7 5321.7 5740.9 4411.9 3054.1 6036.5 C60 . 6829.9 6061.0 5999.2 6026.4 4243.2 2563.3 6662.8 C65 5639.3 6848.2 5892.8 6238.7 761.6 65.5 3.4 6827.8 The 4-MUG enzyme activity change trend paralleled that of 4-MUG enzymatic concentration. id="p-231" id="p-231"

[00231] P80 showed worst ity in 4-MUG enzyme activity under the testing conditions. The enzyme activity of P80 had around 70% decrease after buffer exchange. Later on, the enzyme ty was almost totally lost after 2-week e at 5°C and 40°C. The basic condition significantly affected the enzyme activity.

After 8-week e at 5°C, the enzyme activity of P880 almost completely lost; 20% of decrease was found in P40; no significant change was observed in other formulations.

By contrast, 40°C storage led to obvious decrease in the enzyme activity in all formulations. After 2 weeks, the enzyme activity almost completely lost in P70, dramatically decreased in P40 and C65, and obviously dropped in P50. The enzyme activity continued to decrease in different degree from 2 weeks to 4 weeks. At the end of the study, the enzyme activity almost tely lost in C65, dropped to 211.5 U/L in P40 to ~ 1550 U/L in P50 and C50, and to 2500 ~ 3000 U/L in C60, P60, and C55. Among the formulations, P60 without NaCl kept the highest ty of 3549.7 U/L.

Based on the testing results of 4—MUG enzyme activity, the enzyme was most stabilized at a pH in the range of pH 5.5 ~ 6.0, but no distinction was seen between sodium phosphate buffer and sodium citrate .

: SEC-HPLC id="p-236" id="p-236"

[00236] The results of the SEC measurements are shown below in Table 10: Table 10: 40°C Agm. 4W 8W 2W 4W 8W 81.9 74.3 41.2 18.7 -1.6 24 2.4 ND ND ND 4.0 5.9 18.6 33.9 57.9 2.8 96.8 96.2 96.1 924 808 96.8 1.2 1.9 2.1 0.6 2.7 1.1 2.0 2.0 1.8 7.0 16.5 2.1 92.4 87.0 0.5 ND 0.1 93.5 3.3 5.4 11.0 96.9 96.4 96.7 4.1 . 2.3 1.9 2.7 3.6 3.2 2.4 34.4 26.7 ND ND 0.2 15.2 WO 73060 IU1 0 54.5 69.3 97.3 972. . m 4.0 2.7 2.8 3.3 96.6 92.4 0.5 . 0.7 2.0 3.62 7.0 98.3 93.0 96.3 62.3 32.3 92.1 0.2 0.4 0.2 2. 0. ND 0.4 0.4 4.0 6.8 1.7 37.6 67.7 7.6 8 U] 99.4 99.3 95.9 94.81 88.52 68.5 92.6 0.2 0.2 1.4 0.29 0.19 0.2 0.3 0.5 2.7 4.9 11.3 31.3 7.1 "Q 92.8 . 1.1 1.1 1.0 5 5 0.8 I" O 1.4 1.3 0.9 3.2 17.0 6.4 98.8 98.9 97.6 97.9 22.4 10.0 ND 91.4 53 4; 1.2 1.9 76.7 86.1 86.8 4.4 1.2 0.3 1.0 3.9 13.2 4.2 After buffer exchange, the SEC purity of some of the formulations sed significantly comparing to the starting material (SEC monomer: 98.9%). In P80, the SEC purity dropped to 44.9%, and 54.5% of aggregates formed; in P40, there was a 4.9% decrease on monomer percentage compared to the before—exchange DS, with more LMW fragments formed than HMW molecule (4.7% VS 1.3%); in P70, a slight decrease (1.2%) in monomer percentage was found, corresponding to an increase in aggregates; in P60 without NaCl, there was a 2.3% decrease in monomer percentage.

After 8-week 5°C storage, the SEC purity of formulations P60, P60 without NaCl, and C65 maintained well but the SEC purity of the other formulations significantly decreased compared to T0. In P40, the SEC purity dropped dramatically to 74.3% (T0: 94.0%), and 23.4% of LMW nts formed; in P50 and C50, there were a slight decrease (5~7%) in the monomer and an se (5~7%) in LMW nts. In P80, an 18.2% decrease was found which was mainly transferred to aggregation (14.8%). So as in P70, there was a 10.7% se in monomer and 9.2% increase in HMW fragments. In C55, there was a slight change in monomer, HMW and LMW percentage. 40°C storage led to dramatic monomer percentage change in all formulations. In P70 and P80, the SEC monomer dropped to 0~0.5% after 2 weeks and in C65 there was 22.4% left, and they mostly turned to aggregates after 8 weeks. In P40, P50 and C50, there was a significant drop (50~90%) in monomer, attributed mostly to formation of LMW fragments. In P60, P60 without NaCl, and C60, after 8 weeks the SEC monomer tage decreased to 80.8%, 83.6% and 77.5%, respectively.

The SEC results ted that pH 6.0 performed best for ATB200 stability, and no significant difference was found between sodium phosphate buffer and sodium citrate buffer. Furthermore, absence of sodium chloride in the ations did not impact ATB200 stability.

During agitation study, SEC purity decreased slightly in P60 and P60 without NaCl. In P40, P50, P70, P80, C50, C60 and C65, there was an obvious decrease in the r percent.

Polydispersity: DLS The results of the DLS measurements are shown below in Table 11: Study Condition 0.161 10.7 4352 0 0.137 10.9 4560 P80 13.2 0.100-n 0 Without 7562.0 1.000 4099.0 0 0 (.11 (0 9-144 0 (DA #0) 11.08 2891 (Tl \l P70 9.9 0.087 4442 000000 N0100 000000000 594W0 without 9.8 0.102 10.8 0 O O 2913 O \l A O J}o 03 \1 000 4:.0904:- 4042 M 000 P40 15.1 P50 14.1 0.3350.329..14.0 AA AA COO) . \IOO 00 CO P70 10.5 0.152- 4397 .—‘.—‘ O>\l .4 0 000 OOO °C’8W 11.0 4055 480') O) O 583.3 OO 34O) 01 0000 co(pow 5272 co COO P50 8330 o P80 0.952 741.5 E 1318 533 A (.30 oo _x 0000 N 33.27 OO O 40°C,2W 143 2912 . 21 4325 5.199 5.1 C55 30.84 0.721 704.9 9.81 0) 59.1 40.9 2574 1 168.9 5444 GOO C60 57.1 42.9 GOO C65 98.59 0.31 4822 96.6 P40 2118 0.249 00 2178 5238 94.9 0‘le OO 40°c,4W——n O 100 O P70 34.21 0.122 39.44 I.

P80 22.64 0.096 25.18 " Without 239.5 0.53 394.6 4499 95 C50 173.1 0.7158 9.513 1373 5098 45.1 C55 561.3 1932 187.7 9.455 69.1 14 68. 83 61.6 98.6 1.4 4.277 93.4 92.7 7.3 37.93 62.2 —-n 100 0 —-m- 100 0 40°C,8W without 3215 0.445 1103 100 0 C50 2118 0.184 2711 5.565 92.1 7.9 C55 145.3 1 801.3 153.6 10.1 59.8 23 C60 462.7 0.918 912.6 175.8 4998 54 30.9 C65 196.2 0.309 279.6 45.86 51077 92.1 6.9 P40 13.0 0.331 11.8 190.8 19.9 _8—0.185.2 _97.9 14.8 P70 10.1 0.136 11.4 0 Agitation, P80 51.6 0.657 21.1 3805 100 rpm, P60 °C, t 9.7 5D NaCI C55 9. 8 C65 9.6 DLS data reflected the hydrodynamic radius of protein les and polydispersity of particles. Opalescence was observed in some samples, so DLS was used to analyze the sible particles. The DLS result was generally consistent with the indication from appearance.

During 8—week 5°C storage, both the hydrodynamic radius of protein molecules and polydispersity index (PDI) were stable and comparable in P60, P70, P80, P60 without NaCl and C60. The hydrodynamic radius turned from around 10 nm to a few hundred nm in all other five formulations, especially in P40 after 8 weeks.

WO 73060 During 8-week 40°C storage, there was a dramatic change in the hydrodynamic radius and PDI in all ations. However, due to the complex profiles of the aggregates, it is difficult to compare those formulations based on the result.

In the agitation study, the hydrodynamic radius and PDI of P80 had dramatic se; in P40, P50 and C50, the hydrodynamic radius and PDI had slight increase; in P60, P70, P60 without NaCl, C55, C60, and C65, no significant change were observed.

According to all above DLS data, P60, P70, P60 without NaCl, C55, C60, and C65 were better than P40, P50, P80, and C50.

Particles: HIAC id="p-250" id="p-250"

[00250] The results of the HIAC measurements of the agitation study are shown below in Table 12: Table 12: Differential Counts/mL Particle Sam 1e SE1 Agltation, 100 — rpm, 25°C 5 days 2 5 1704 2593 P40 2 10 97 163 2 25 8 0 Z 5 4600 7378 P50 2 10 600 815 2 25 15 0 2 5 1052 1245 P60 2 10 112 134 Z 25 0 0 Z 5 171 2860 P70 2 10 15 341 2 25 0 8 Z 5 15 830 15104 P80 2 10 5786 4104 2 25 630 341 P60 2 5 289 149 without 2 10 52 23 NaCl 2 25 0 0 2 5 200 200 C50 2 10 23 15 Z 25 0 0 Z 5 223 319 C55 2 10 23 23 2 25 0 0 Z 5 408 497 2 10 15 23 During the agitation study, P70 had a significant increase in the particle number after agitating at 25°C for 5 days, and all other formulations had comparable particle numbers.

Summary Overall, P60 and C60 stood out as most stable ations compared to the others. Therefore, it was concluded that ATB200 was most stable in a range of 0.

However, no distinction could be made between phosphate and citrate . In addition, the absence of sodium chloride in the formulation didn’t show significant impact on ATB200 stability.

FORMULATION EXAMPLE 2: FREEZE-THAW Three formulations were evaluated for stability during the freeze—thaw process.

The three ations are summarized below in Table 13 below. The target concentration of ATB200 enzyme was 5 mg/mL.

Table 13: w25mMPhosphate 50 mM mM Phosphate- Citrate buffer Sample Preparation The ATB200 DS was buffer exchanged into the 3 formulation buffers using dialysis cassette (20000 MWCO). Dialyses were performed at 5°C with a gentle stirring, each using three buffer changes, once every 6~10 hours.

After each dialysis, appropriate volume of formulation buffer was added to adjust the final UV concentration to 5 mg/mL. The ons were then aseptically ed with 0.22-um PES filter. Each ation was then aseptically filled into 2-mL glass vials in a bio—safety hood with the filling volume of 500 uL ~ 1000 uL according to the sample amount requirement of analytical tests. Vials were stoppered and then crimp—oversealed ately after filling.

Before freeze—thaw vials were taken for each formulation and the remaining sample vials were subjected to the designed freeze-thaw cycles. After—freeze-thaw sample vials were taken at pre—defined sampling .

Sample Testing Three freeze-thaw processes were tested, as listed below: Process 1: rolled freezing and thawing. Samples were frozen in a -80°C freezer and thawed in a 25°C chamber. id="p-262" id="p-262"

[00262] Process 2: controlled freezing and rolled thawing. Samples were placed in a Frosty container and frozen in a —80°C r. The Frosty container used isopropanol to achieve controlled freezing rate at 1°C /min. The Frosty container was placed in a 25°C chamber to thaw the samples. The rate of temperature increase was approximately 1°C /min.

Process 3: controlled freezing and g. A lyophilizer was used to freeze and thaw the samples. The lowest sample temperature achieved during ng was —47°C. The sample temperature was brought up to 25°C. The rate of temperature change for both freezing and thawing was controlled at about 05°C /min. Results were confirmed by repeating the ment.

Five or three freeze-thaw cycles were med using each process. The following tests were performed for the samples before and after freeze-thaw: Appearance, concentration (UV & Enzymatic) and SEC-HPLC. A summary of the testing parameters follows below in Table 14: Table 14: Eguipment and Freezing rate Heating rate Condition and time and time Refrigerator, -80°C ND, 40~60 ND, 40~6O to 25°C min min Nalgene Mr. Frosty Cryo 1°C Freezing 1°C/min, ~1°C/min, Container, —80°C to 100~120 min 100~120 min Lyophilizer, -47°C 0.5°C/min, to 25°C ~164min X: Appearance, Concentration (UV & tic), Enzyme activity, SEC *: This experiment was repeated, 3 FT cycles for the first experiment and 5 FT cycles for the second experiment.

Results Appearance — 1St Round The s of the first round of appearance measurements are shown below in Tables 15A and 15B: Table 15A: mm erator Name 1 FT 3 FT Colorless, Colorless, Colorless, slightly slightly slightly Colorless, slightly opalescence, opalescence, opalescence, C60 opalescence, free of little amount more e large amount Visible particles of Visible particles than of Visible particles 1FT particles Table 15B: Mr. Frosty Container Sample Lyophilizer 1 FT 3FT P60 Colorless Colorless Colorless, Colorless, Colorless, slightly , slightly , slightly slightly slightly opalescen opalescence, opalescence, opalescence, C60 ce, little ce, free little amount more e large amount amount of Visible of Visible particles than of Visible of Visible particles particles 1FT particles CP60 les At T0 (before freeze—thaw), all formulations appeared to be colorless, slightly opalescent, and free of Visible particles, using corresponding formulation buffers as a reference.

After 5 cycles of quick freeze—thaw (process 1), all formulations ed to contain more Visible particles compared to their T0. CP60 contained fewest particles among the three after 5 FT.

] After 5 cycles of slow -thaw (process 2), all formulations appeared to contain more Visible particles than T0, whereas less than the ones treated by process 1. Fewer particles were observed in CP60 samples after 5 FT cycles than in P60 and C60 samples after 1 FT cycle.

After 3 cycles of slow freeze-thaw (process 3), all formulations appeared to have more visible particles than TO. There was no difference among the three formulations.

Appearance — 2nd Round The results of the second round of appearance measurements are shown below in Tables 23: Table 16: Refrigerator Colorless, opalescence, large amount of visible protein Colorless, Colorless, ess, particles slightly slightly slightly opalescence, cence, little amount of opalescent, Colorless, few. visible visible visible particles more slightly partlcles partlcles (more than T0) opalescence, little amount of e particles (compare to 3 Buffers (P60, C60, and ess, clear, free of visible particles In 2nd round of slow freeze—thaw study (process 3), more FT cycles were carried out than the first round. There were large number of e les appeared in P60 after 5 FT compared to T0, whereas small number of particles were observed both in C60 and CP60. 4-MUG Enzymatic Concentration and Activity The results of the 4-MUG enzymatic concentration and activity measurements are shown below in Table 17 (processes 1—2 and first round of process 3) and Table 18 (second round of process 3): Table 17: Mr. Frosty Refrigerator mm Container "—11333 ./ 1 ./ml -187—116 -3m0;11 -173—389 -323 -181—459 _277 00 ------ Table 18: L o hilizer ——5F_T Conc. it C_onc.. —Activity, Conc. Activity, _g_lm mg./ml % P60 3.09 172528 2. 51 —-U— 13954.9 79253 0.96 4973.7 C60 2.98 8 3.36 18816.8 16931.2 3.80 20606.3 CP60 2.97 16561.6 2.93 16375.3 15261.8 0.99 5157.5 After 5 cycles of quick freeze—thaw (process 1), a slight decrease in 4—MUG enzymatic concentration was observed in P60 when compared to the T0 sample, while no significant difference in C60 and CP60 was observed.

] After 5 cycles of slow freeze-thaw using a Frosty container (process 2), no significant difference was observed compared to T0 samples, in any of the 3 formulations.

After 3 cycles of slow freeze-thaw using a lizer (process 3), there was distinct enzymatic concentration decrease in both P60 and CP60, but no significant change in C60, when compared to T0. The freeze—thaw with process 3 was repeated with 5 cycles and same trend was observed. Enzymatic concentration of P60 samples dropped 18.8% after 1 cycle of F/T, and became worse after 3 cycles (decreased by 53.1%) and 5 cycles (decreased by 68.9%). In CP60, enzymatic concentration d to drop after 3 cycles, and finally decreased to the same level as P60 after 5 cycles. In C60, there was almost no change after 5 cycles of slow freeze-thaw.

The 4-MUG enzymatic concentration results showed that the freezing/heating rate, number of F/T cycles and buffer type might impact ATB200 stability. The citrate buffer system (C60) provided good stabilizing effect regardless of the freeze-thaw process used.

The 4-MUG enzyme activity change trend was the same as the 4-MUG enzymatic concentration.

] During quick freeze-thaw study (process 1), a slight decrease in enzyme activity happened in P60 after 5 cycles, no significant change was observed in C60 and CP60.

No distinct change was observed in any sample during slow freeze—thaw study with 1°C/min temperature change (process 2). id="p-284" id="p-284"

[00284] In the slow freeze-thaw study with a lyophilizer ss 3), the enzyme activity in P60 had an obvious decrease and in CP60 had a slight decrease. However, there was almost no change in C60. During the 2nd round of process 3 slow freeze-thaw study, the enzyme activity of P60 sed 19.1% after 1 cycle, 54.1% after 3 ) and 71.2% after 5 cycles.

In CP60, the enzyme activity started to drop after 3 cycles, dropped to a similar level to P60 after 5 cycles. There was negligible change in C60 after 5 cycles.

Purity: SEC-HPLC ] The results of the purity ements are shown below in Table 20 (processes 1—2 and first round of process 3) and Table 21 (second round of process 3): —20:Table _v_p_Lo hilizer s_p_1am e S_EC T_0 3_FT 5_FT 3_FT 5_FT n99.7 99.6 99.5 99.7 99.7 —-—m0-3 999 99.9 99.9 99.8 99.8 C60 0-1 99.7 99.7 99.7 99.8 99.8 m CP60 0-1 Table 21: Lyophilizer Monomer % HMW% . . .

LMW% ND ND <0.1 <0.1 No SEC monomer percentage change was detected in any sample after up to 5 cycles of quick F/T (process 1) or slow F/T (process 2).

In the first slow F/T study (process 3), P60 samples showed obvious SEC r percentage decline (mainly due to formation of HMW species) after 3 cycles. And no distinct change happened in C60 and CP60. In the 2nd process 3 slow F/T study (0.5°C/min), the monomer in P60 samples d to drop after 1 cycle and finally dropped 68.1% after 5 cycles. And in CP60, the monomer percentage stated to drop after 3 cycles, but to a much less extent than in P60; however, it reached the same level as in P60 after 5 .

There was no change on monomer percentage in C60 after up to 5 cycles.

The SEC purity results were consistent with the performance in enzymatic concentration and activity, confirming that freezing/heating rate, number of F/T cycles and buffer type might impact ATB200 stability during freeze—thaw process. Buffers containing sodium ate did not protect ATB200 as well during freeze—thaw cycles, based on the SEC results.

Summar ATB200 formulated in citrate buffer (C60) ood multiple freeze—thaw cycles better than the other two s (P60 and CP60). Regardless of the freeze-thaw process, ATB200 remained stable in e buffer.

FORMULATION EXAMPLE 3: ENT Eight formulations (El—8) were prepared with various excipients. The ATB200 enzyme concentration in excipient evaluation study was 5 mg/mL. Three buffers, two stabilizers and one surfactant were selected to assess the protein ity in formulations described in Table 22 below.

Table 22: Sample Buffer pg NaCl ose Polysorbate 80 E1 2% — 005% —25 mM Phosphate E2 / — 005% E3 2% — 005% _ — 25 mM Citrate E4— / — 005% 6-0 50 mM E5 25 mM Phosphate / —— E6 25 mMCitrate / _— E7 25 mM Phosphate— 2% ——0.05% Citrate combination E