recombinant human acid -glucosidase - neurology · skeletal muscle weakness, and early death. the...

Articles

Recombinant human acid -glucosidaseMajor clinical benefits in infantile-onset Pompe disease

PS Kishnani MD D Corzo MD M Nicolino MD PhD B Byrne MD PhD H Mandel MDWL Hwu MD PhD N Leslie MD J Levine MD C Spencer MD M McDonald MD J Li MDJ Dumontier MD M Halberthal MD YH Chien MD R Hopkin MD S Vijayaraghavan MD

D Gruskin MD PhD D Bartholomew MD A van der Ploeg MD PhD JP Clancy MD R Parini MDG Morin MD M Beck MD PhD GS De la Gastine MD M Jokic MD B Thurberg MD PhD

S Richards PhD D Bali PhD M Davison MD MA Worden BS YT Chen MD PhD and JE Wraith MD

AbstractmdashBackground Pompe disease is a progressive metabolic neuromuscular disorder resulting from deficiency oflysosomal acid -glucosidase (GAA) Infantile-onset Pompe disease is characterized by cardiomyopathy respiratory andskeletal muscle weakness and early death The safety and efficacy of recombinant human (rh) GAA were evaluated in 18patients with rapidly progressing infantile-onset Pompe disease Methods Patients were diagnosed at 6 months of age andyounger and exhibited severe GAA deficiency and cardiomyopathy Patients received IV infusions of rhGAA at 20 mgkg(n 9) or 40 mgkg (n 9) every other week Analyses were performed 52 weeks after the last patient was randomized totreatment Results All patients (100) survived to 18 months of age A Cox proportional hazards analysis demonstratedthat treatment reduced the risk of death by 99 reduced the risk of death or invasive ventilation by 92 and reduced therisk of death or any type of ventilation by 88 as compared to an untreated historical control group There was no clearadvantage of the 40-mgkg dose with regard to efficacy Eleven of the 18 patients experienced 164 infusion-associatedreactions all were mild or moderate in intensity Conclusions Recombinant human acid -glucosidase is safe and effectivefor treatment of infantile-onset Pompe disease Eleven patients experienced adverse events related to treatment but nonediscontinued The young age at which these patients initiated therapy may have contributed to their improved responsecompared to previous trials with recombinant human acid -glucosidase in which patients were older

NEUROLOGY 20076899ndash109

Editorial see page 88See also page 110

This article was previously published in electronic format as an Expedited E-pub at wwwneurologycomFrom the Department of Pediatrics (PSK MM JL DB) Duke University Medical Center Durham NC Genzyme Corporation (DC BT SRMAW) Cambridge MA Division of Pediatric Endocrinology (MN JD) Diabetology and Metabolism Hopital Debrousse Lyon France Division ofPediatrics (BB CS) Shands Hospital at University of Florida Gainesville FL Metabolic Unit (HM MH) Rambam Medical Centre Haifa IsraelDepartment of Pediatrics and Medical Genetics (WLH YHC) National Taiwan University Hospital Taipei Taiwan Department of Pediatrics andMedical Genetics (NL RH) Cincinnati Childrenrsquos Hospital Medical Center Cincinnati OH Department of Cardiology (JL) Childrenrsquos Hospital BostonMA Willink Biochemical Genetics Unit (SV JEW) Royal Manchester Childrenrsquos Hospital Manchester UK Department of Genetics (DG) EmoryUniversity Medical Center Atlanta GA Department of Molecular and Human Genetics (DB) Columbus Childrenrsquos Hospital Columbus OH Division ofMetabolic Diseases and Genetics (AVDP) Department of Pediatrics Erasmus Medical CenterSofia Childrenrsquos Hospital Rotterdam The NetherlandsDepartment of Pediatrics (JPC) University of Alabama at Birmingham University of Alabama Birmingham AL Centro Fondazione Mariani per lemalattie metaboliche dellrsquoInfanzia (RP) Department of Pediatrics University Milano-Bicocca San Gerardo Hospital Monza Italy Departement dePediatrie (GM) CHU Amiens Amiens France Metabolic Unit (MB) Universitats-Kinderklinik Mainz Mainz Germany Pediatric Intensive Care Unit(GSDlG MJ) University Hospital Center Cote de Nacre Caen France Genzyme Corporation (MD) Naarden The Netherlands Institute of BiomedicalSciences (YTC) Academica Sinica Taipei TaiwanThese authors contributed equally to this studyDisclosure D Bali B Byrne YT Chen R Hopkin PS Kishnani H Mandel M McDonald M Nicolino C Spencer and A van der Ploeg have receivedresearchgrant support from Genzyme Corporation for other research or activities not reported in this article PS Kishnani M Nicolino B Byrne and Avan der Ploeg are members of the Pompe Disease Advisory Board for Genzyme Corporation YT Chen and J Levine have served as consultants forGenzyme J Levine JE Wraith C Spencer M McDonald YT Chen R Hopkin PS Kishnani M Nicolino and H Mandel have received honoraria fromGenzyme Corporation The clinical trials with rhGAA have been supported by a grant from Genzyme Corporation at the various sites at which patients weretreated rhGAA in the form of Genzymersquos product Myozyme has now been approved in the United States Canada and the European Union as therapy forPompe disease Erasmus Medical CenterSofia Childrenrsquos Hospital Duke University and inventors of the method of treatment and predecessors of the celllines used to generate the enzyme (rhGAA) used in this clinical trial may benefit financially pursuant to their respective Policy on Inventions Patents andTechnology Transfer even if those cell lines are not used in the commercialized therapyReceived May 17 2006 Accepted in final form October 10 2006Address correspondence and reprint requests to Dr Priya Sunil Kishnani Division of Medical Genetics Department of Pediatrics Box 3528 Duke UniversityMedical Center Durham NC 27710 e-mail kishn001mcdukeedu

Copyright copy 2007 by AAN Enterprises Inc 99

Pompe disease also known as glycogen storage dis-ease type II or acid maltase deficiency is a rareprogressively debilitating and often fatal lysosomalstorage disorder Patients with Pompe disease havean autosomal-recessively inherited deficiency of theenzyme acid -glucosidase (GAA) which hydrolyzesglycogen to glucose This GAA deficiency causes gly-cogen to accumulate in multiple tissues especiallyskeletal and cardiac muscle Glycogen deposits dis-rupt muscle cytoarchitecture and function causingprogressive motor respiratory and cardiacdysfunction1

Patient age at the onset of Pompe disease symp-toms and the rate of deterioration can vary consider-ably1 Patients with infantile-onset Pompe diseaseexperience symptoms within the first year of life andprogress rapidly to death In other patients symp-toms are exhibited later and the disease progressesmore slowly but it is still associated with significantmorbidity Patients with early onset of symptomstypically manifest hypotonia muscle weakness andcardiomyopathy within the first months of life Theoutlook for these patients is grim they fail to acquireor lose motor developmental milestones and cardio-respiratory failure and death usually occur before 1year of age2-4 Slonim et al5 described a subset ofinfantile-onset patients with less severe cardiac in-volvement (ie less severe cardiomyopathy and alack of cardiomegaly before the age of 6 months) whosurvived longer with ventilatory support However arecent retrospective chart review found early cardio-megaly even among first-birthday survivors6 In thishistorical cohort early symptom onset was associ-ated with early demise

In early clinical trials IV enzyme replacementwith different forms of recombinant human (rh) GAAhas improved survival cardiac and respiratory func-tion and motor development in severely affectedinfants7-11 A subset of these treated patientsachieved independent ambulation In a recent trial ofrhGAA derived from transfected Chinese hamsterovary cells eight of eight infants showed cardiac im-provement six were alive and five were free of inva-sive ventilation after 52 weeks of treatment and fiveattained new motor milestones11 Of the three pa-tients who eventually achieved the ability to walkindependently all had begun treatment prior to theage of 6 months suggesting that early treatmentmay be critical for the optimal motor response totreatment11 In this multicenter multinationalopen-label dose-ranging study we examined thesafety and efficacy of rhGAA treatment in a largercohort of severely affected patients with Pompe dis-ease who began treatment prior to 6 months of age

Methods Study design and treatment Local institutional re-view boards or independent ethics committees approved protocolsand consent forms at each of 13 primary sites (six centers in theUntied States five in Europe one in Taiwan and one in Israel)Parents or guardians gave written informed consent for patientsrsquoparticipation An independent data safety monitoring board re-viewed safety an independent allergic reaction review board was

consulted if required for issues related to serious infusion-associated reactions (IARs)

Eligible patients had documented symptoms of infantile-onsetPompe disease including skin fibroblast GAA activity 1 of thenormal mean and hypertrophic cardiomyopathy (left ventricularmass index 65 gm2 by echocardiogram) they were no older than26 weeks at enrollment Exclusion criteria included respiratoryinsufficiency (O2 saturation 90 or CO2 partial pressure 55mm Hg [venous] or 40 mm Hg [arterial] in room air or anyventilator use) a major congenital anomaly or clinically signifi-cant intercurrent illness unrelated to Pompe disease or any priorGAA treatment A total of 18 patients were studied

Parents or legal guardians were informed that the protocol-defined stopping rules related to the occurrence of serious adverseevents (AEs) and allowed for temporary or permanent patientdiscontinuation if the investigator safety review board and spon-sor determined there was a significant risk to patient safety

It was considered unethical to include a placebo group as partof the study design for the following reasons 1) infantile-onsetPompe disease is a rapidly fatal disorder and 2) early clinicaltrials have shown that treatment with various forms of rhGAAcan improve survival cardiac and respiratory function growthand motor development in severely affected infants7-11 Thus ahistorical control group of 61 severely affected infants 6 monthsold and younger was identified by applying the study inclusionand exclusion criteria to a group of 168 patients with infantile-onset Pompe disease identified through a retrospective chart re-view6 The cohort included 168 patients from nine countries 33different sites with birth dates ranging from before 1985 to 20026

Drug effects on overall survival (with or without ventilator use)and invasive ventilator-free survival were compared betweentreated patients and this historical cohort our group of 18 treatedpatients provided 98 power to detect a difference betweentreated and control groups Table 1 summarizes the screeningcriteria applied to the original historical patient population inorder for patients to be included in the historical control subgroup

Alglucosidase alfa which contains rhGAA produced in Chinesehamster ovary cells was supplied by Genzyme Corporation Asimple randomization scheme was used to assign patients to thetwo dose groups Upon confirmation that patients met all eligibil-ity criteria and completion of pretreatment screening and baselineassessments eligible patients were randomized in a 11 ratio toreceive an IV infusion of either 20 mgkg or 40 mgkg of rhGAAevery other week Clinicians and patients were not blinded to thedose of rhGAA used The total amount of rhGAA administeredcould be adjusted every 4 weeks to account for changes in bodyweight

Clinical assessments of safety and efficacy Patients were ob-served and their vital signs were monitored during rhGAA infu-sion and for 2 hours afterward Safety assessments includedevaluating blood and urine chemistry (data not shown) the occur-rence and timing of the development of anti-rhGAA IgG antibod-ies (assayed at first rhGAA infusion and every 4 weeksthereafter) and the incidence and nature of AEs including IARs

Efficacy was measured by assessing survival ventilator useleft ventricular mass by echocardiography growth (weight andlength) muscle GAA activity and glycogen levels motor develop-ment (using the Alberta Infant Motor Scale [AIMS]12) and level ofdisability (using an adaptation of the Pediatric Evaluation of Dis-ability Index [PEDI]13 developed for Pompe disease1415 Althoughnot an efficacy endpoint cognitive development (using BayleyrsquosScales of Infant Development Second Edition [BSID-II]) was alsoserially evaluated16 Echocardiograms and glycogen content inmuscle biopsy samples were centrally read by a pediatric cardiol-ogist and a pathologist who were blinded to dose patient andtime point Motor and cognitive evaluations were centrally scoredby a nonblinded central clinician

Survival and ventilation data were analyzed up to 18 monthsof age as compared to survival of the historical control group allother efficacy data were analyzed with respect to changes frombaseline after 52 weeks of treatment The primary efficacy end-point was the Kaplan-Meier17 proportion of patients alive and freeof invasive ventilation at 18 months of age Safety data wereanalyzed for the duration of treatment Because patients wereenrolled over a period of 1 year safety data ranged from 52 weeksfor the last patient randomized to treatment to 106 weeks for thefirst patient treated with rhGAA

100 NEUROLOGY 68 January 9 2007

GAA activity assay Cultured skin fibroblasts and muscle bi-opsy tissue obtained from patients were assayed for GAA activityagainst 4-methylumbelliferyl -d-glucoside in an assay similar tothe method of Reuser et al18 and described in detail in Kishnani etal11

Biochemical glycogen determination Open quadriceps musclebiopsies were performed at baseline and weeks and week 52 Bi-opsy samples taken at weeks 12 and 52 were obtained 48 hoursafter the regularly scheduled rhGAA infusion for that visit A 05

05 3-cm sample was taken from alternating sides at eachtime point under general regional or local anesthesia Methodsfor glycogen quantification were performed as described inKishnani et al11

Briefly for biochemical analysis muscle tissues were homoge-nized centrifuged boiled and treated with amyloglucosidase todigest the glycogen The glucose produced was quantified with aGlucose Trinder Kit (Sigma St Louis MO)

Biochemical measurements of glycogen levels in skeletal mus-cle of normal children are not available Therefore changes inglycogen content were compared between baseline and week 52within individual patients

Anti-rhGAA antibody testing Serum samples were takenprior to each infusion from initiation of therapy through week 24and subsequently at weeks 38 52 64 78 90 and 104 As dis-cussed above different patients received different durations ofrhGAA treatment ranging from 52 to 106 weeks depending onwhen they enrolled in the study The presence of IgG antibodies torhGAA was assessed using enzyme-linked immunosorbent assaysand confirmed using radioimmunoprecipitation as described inKishnani et al11

Genotyping DNA was isolated from the peripheral blood ofpatients and sequenced GAA mutation analysis was determinedby Genzyme Corporation as described previously11

Cross-reacting immunologic material (CRIM) status CRIMstatus was determined as described previously11 Briefly cell ly-sates derived from patientsrsquo fibroblasts were subjected to Westernblot analysis with a pool of monoclonal antibodies that recognizeboth native and recombinant GAA A patient was considered to beCRIM positive if the presence of any of these forms of GAA wasdetected in samples prepared from patient fibroblasts in the West-ern blot assay

Results Disposition of patients A total of 20 patientswere evaluated for eligibility to participate in this study(figure 1) Two patients were excluded from the study onewas determined not to have Pompe disease based on GAAactivity levels The other required ventilation during thebaseline period before receiving any treatment This pa-tient was discontinued from the study because the use ofany ventilation at the time of enrollment was an exclusioncriterion but went on to receive rhGAA under an expandedaccess program

As discussed previously it was considered unethical toadminister placebo treatment to patients in this studyThe trial was designed to show a positive drug effect onsurvival as compared to a historical control population Aretrospective review was performed on the medical recordsof an international cohort of 168 patients with infantile-onset Pompe disease who did not receive treatment6 Asubset of 62 patients (referred to as the historical controlgroup) from within this cohort was selected based on thescreening criteria for the clinical trial including age atfirst symptoms age at diagnosis presence of cardiomyopa-thy by 6 months of age GAA activity and congenital ab-normalities as summarized in table 1 One of the 62patients was excluded from the survival analyses becausethe date of death was unavailable This historical controlgroup of 61 untreated patients was used as a comparatorpopulation for the severely ill infantile-onset rhGAA-treated patients in this study

Demographics Demographic data are summarized intable 2 For rhGAA-treated patients the median age atpresentation was 10 month 43 months at postnatal diag-nosis (one patient was diagnosed prenatally) and 53months at first rhGAA infusion At enrollment ages ofpatients born before 40 weeks were corrected for gestationPatients in the clinical trial and historical control subjects

Table 1 Summary of study inclusion and exclusion criteriaapplied to patients in the historical control group

Category No () of patients

Inclusion criteria

Documented cardiomyopathy firstidentified by 26 wk of age

Yes 115 (685)

No 53 (315)

Confirmed diagnosis by 26 wk of age

Yes 112 (667)

No 56 (333)

First symptoms by 26 wk of age

Yes 153 (911)

No 15 (89)

Exclusion criteria

GAA activity level 1 of themean of the normal range

Yes 0

No 168 (1000)

LVMI 65 gmsup2 at the last visitprior to 26 wk of age

Yes 1 (06)

No 167 (994)

Documented major congenital abnormality

Yes 12 (71)

No 156 (929)

Clinically significant disease(s)(with the exception of symptomsrelating to Pompe disease)

Yes 37 (220)

No 131 (780)

Any known ventilator use between0 and 6 mo of age due to reasons otherthan procedure-related use

Yes 25 (149)

No 143 (851)

All eligibility criteria met

Yes 62dagger (369)

No 106 (631)

All age calculations were corrected for gestation

ldquoNordquo may include patients for whom data were unavailabledagger Although 62 patients met the eligibility criteria for the histori-

cal control group one patient was excluded because the date ofdeath was unavailable

LVMI left ventricular myocardial index

January 9 2007 NEUROLOGY 68 101

had similar demographic profiles in terms of age genderand ethnicity as shown in table 2

Survival and ventilator use Survival Fifteen of the18 treated patients reached the age of 18 months by theend of the study three patients were right-censored fromthis analysis because they had not reached the age of 18months by the end of the study although they were aliveat that time (at ages 159 months 179 months and 144months) This is in striking contrast to what was observedin the untreated historical control group in which only oneof 61 patients survived to the age of 18 months (19 95CI 0 to 55) A Kaplan-Meier analysis of overall 18-month survival is shown in figure 2A

Invasive ventilator use Of the six patients who re-quired ventilatory assistance three patients (including onewho had not reached the age of 18 months by the end ofthe study) required invasive ventilation beginning at 9192 and 150 months of age Invasive ventilation was de-fined as requiring passive ventilation through an endotra-cheal tube or tracheostomy The other two patients whohad not reached the age of 18 months by the end of thestudy were right-censored from this analysis althoughthey were free of invasive ventilatory support at that timeThe remaining 13 patients were free of invasive ventila-tory support at the age of 18 months Thus the Kaplan-Meier invasive ventilationndashfree survival rate (the primaryefficacy endpoint) was 889 (95 CI 744 to 100) far

greater than the survival rate for untreated patients (fig-ure 2B)

Any ventilator use Six patients including one whohad not reached the age of 18 months by the end of thestudy required ventilatory support (invasive or noninva-sive beginning at ages ranging from 78 to 148 months)The other two patients who had not reached the age of 18months by the end of the study were right-censored fromthis analysis although they were free of any ventilatorysupport at that time The remaining 10 patients were ven-tilator independent for a Kaplan-Meier ventilator-freesurvival rate of 667 (95 CI 449 to 884) (figure2C) Again this is much higher than the overall survivalrate in the untreated historical control group

Cox proportional hazards analysis A Cox proportionalhazards analysis19 with model terms of age at diagnosisage at symptom onset and treatment as a time-varyingcovariate was used to compare the risk of an event fortreated patients to the untreated historical control grouprhGAA treatment was found to reduce the risk of death by99 (hazard ratio of 001 95 CI 000 to 010 p 0001)the risk of death or invasive ventilation by 92 (hazardratio of 008 95 CI 003 to 021 p 0001) and the riskof death or any type of ventilation by 88 (hazard ratio of012 95 CI 005 to 029 p 0001)

Cardiac response All patients had echocardiographicevidence of cardiomyopathy at enrollment Additional

Figure 1 Disposition of patients in his-torical control group and recombinanthuman acid -glucosidase-treatedgroup


echocardiograms were performed at baseline and at weeks4 8 12 26 38 and 52 The mean baseline left ventricularmass index was 1934 gm2 (n 15 range 593 to 3018gm2) At week 52 mean left ventricular mass index haddeclined to 868 gm2 (n 15 range 449 to 1573 gm2)Treatment decreased left ventricular mass as early asweek 4 as summarized in figure 3 which shows that meanleft ventricular mass Z scores decreased from 71 at base-line to 33 by the end of the 52-week treatment period

Ejection fraction was also measured by echocardiogra-phy to provide a gross and noninvasive measure of ventric-ular function At baseline the mean ejection fraction was512 (n 16 range 253 to 759) and at week 52 themean ejection fraction was essentially unchanged at549 (n 18 range 244 to 756) It is important to notethat measurement of cardiac function through noninvasivemethods such as echocardiograms is imprecise in subjectswith evidence of left ventricular hypertrophy20 More inva-

Table 2 Patient demographics and baseline characteristics

Parameter

rhGAA-treatedpatients(n 18)

Historicalcontrol group

n 62)

Overall naturalhistory population

(n 168)

Gender n ()

Male 11 (611) 28 (452) 83 (494)

Female 7 (389) 34 (548) 85 (506)

Race n ()

White 7 (389) 31 (500) 80 (476)

Black 4 (222) 4 (65) 22 (131)

Hispanic 2 (111) 1 (16) 1 (06)

Asian 3 (167) 18 (290) 52 (310)

Other 2 (111) 1 (16) 1 (06)

Unknown 00 7 (113) 12 (71)

Age at first symptoms mo

n 18 62 166

Mean 16 19 27

Median 10 16 20

SD 178 179 248

Min max 00 55 00 59 00 120

Age at diagnosis modagger

n 18 62 165

Mean 37 36 60

Median 43 41 47

SD 217 194 883

Min max 02 68 44 66 51 842

Age at first infusion moDagger

n 18 NA NA

Mean 46 NA NA

Median 53 NA NA

SD 167 NA NA

Min max 12 61 NA NA

Age at death in historical reference group

n NA 61sect 163

Median (95 CI) NA 75 (67ndash86) 87 (82ndash94)

Min max NA 03 439 16 569

Kishnani et al6

dagger Negative values are indicative of prenatal diagnosesDagger Note that ages at first infusion were corrected for gestational age in patients who were born before 40 weekssect One patient was excluded from this analysis as date of death was unknown a total of 55 of 61 (887) patients died six were

right-censoredpara A total of 141 of 163 (865) patients died 22 were right-censored

NA not applicable


sive procedures to measure cardiac function were not usedbecause of the risk associated with these procedures andthe compromised state of many of these patients at studyentry

In patients with neuromuscular disorders and respira-tory insufficiency it can be difficult to specifically diagnosecardiac failure as the symptoms for both are similar Asdiscussed above echocardiographic evaluation by itselfmay not accurately reflect cardiac dysfunction in patientswith left ventricular hypertrophy The presence of clinicalsigns and symptoms of cardiac failure was evaluated atregular intervals by the investigators who were asked toreport AEs that were considered specific to or suggestive ofcardiac failure According to these criteria two of 18 pa-tients (111) had findings compatible with cardiac failureprior to rhGAA administration (from screening to day 0)assessed as described above (Patients H and M both as-signed to the 20-mgkg dose group) By week 52 no pa-tients had findings compatible with cardiac failurePatients H and M had recovered clinically

Physical growth Normal weight and length were con-sidered as values above the third percentile21 Fifteen ofthe 18 patients maintained weight above the third percen-tile during treatment The three patients who failed tomaintain normal weight had weights below the third per-centile prior to rhGAA treatment and were tube fed Incontrast only two of the patients who maintained weightabove the third percentile were tube fed Notably 53 ofuntreated infantile Pompe disease patients failed tothrive6 In terms of length 15 of 16 patients maintainednormal length-for-age percentiles during treatment

Motor and cognitive development Thirteen of 18 pa-tients made consistent motor and functional gains on theAlberta Infant Motor Scale from baseline to week 52 (datanot shown) Seven of these patients were considered to bewalkers as they were able to walk independently by week52 Three patients could pull to stand independently andwalk with hands held by week 52 these patients wereconsidered to be standers Three patients could roll and sitindependently but were unable to bear weight through thelegs and were therefore designated as functional sittersThe remaining five patients did not make substantialgains in AIMS scores from baseline to week 52

The Pompe PEDI was specifically designed to evaluatefunctional skills in patients with Pompe disease1415 andincludes three content domains mobility self-care andsocial skills The seven patients designated as walkersbased on the AIMS assessments also acquired the most

Figure 2 (A) Kaplan-Meier estimate of time from date ofbirth to death Three patients were right-censored fromthis analysis (at ages 144 159 and 179 months) becausethey had not reached the age of 18 months by the end ofthe study although they remained alive at that time (B)Kaplan-Meier estimate of time from birth to invasive ven-tilator use or death Two patients were right-censored fromthis analysis (at ages 159 and 179 months) because theyhad not reached the age of 18 months by the end of thestudy although they remained free of any ventilatory sup-port that time One patient who had not reached the age of18 months by the end of the study required invasive venti-latory support and was therefore not right-censored butwas designated as failing this endpoint (C) Kaplan-Meierestimate of time from birth to any ventilator use or deathTwo patients were right-censored from this analysis (atages 159 and 179 months) because they had not reachedthe age of 18 months by the end of the study althoughthey remained free of any ventilatory support that timeOne patient who had not reached the age of 18 months bythe end of the study required ventilatory support and wastherefore not right-censored but was designated as failingthis endpoint In each panel thick solid lines show theKaplan-Meier estimates for the treated patient group thinsolid lines show those for the historical control group with95 CIs given by the corresponding dashed lines Solidcircles indicate right-censored observations Asterisk indi-cates that one patient from the historical control groupremained alive at 18 months of age this patient died atage 44 months

Figure 3 Mean left ventricular mass Z scores of treatedpatients from baseline to week 52 vertical bars representstandard error


skills in the Pompe PEDI including (but not limited to)such skills as self-feeding playing independently climbingstairs and naming objects The three patients designatedas standers also made substantial gains in all three do-mains of the Pompe PEDI although they did not acquireas many skills as the walkers The three patients desig-nated as functional sitters demonstrated modest gains onthe Pompe PEDI These patients acquired skills like headcontrol and independent play with objects in a sitting posi-tion as well as some basic communication skills The re-maining five patients did not make substantial gains inthe AIMS or the Pompe PEDI assessments A summary ofthe gains made by the all patients in each content domainof the Pompe PEDI are shown in figure 4

All patients acquired cognitive language and personalsocial development skills during the 52-week treatmentperiod although some patients acquired these skills moreslowly than same-age peers For example by week 52 nineof the 17 patients (53) who were evaluated with theBSID-II at baseline and at week 52 had mental develop-ment index (MDI) scores within normal limits (85 to 115)whereas four patients had MDI scores indicating mildlydelayed performance (70 to 84) and four patients had MDIscores consistent with significantly delayed performance(69) (data not shown) It is important to note that themotor involvement typical of patients with Pompe diseaseinterferes with the assessment of integrated oral-motorand fine-motor skills required for the administration of theBSID-II In addition hearing loss which was recently rec-ognized as a complication of infantile-onset Pompe dis-ease22 could interfere with cognitive testing

Muscle GAA activity and glycogen content We assayedGAA activity and glycogen content in quadriceps musclebiopsy specimens taken at baseline and weeks 12 and 52At baseline mean quadriceps GAA activity was 122 mmolhg (n 18 range 00 to 88) in 15 patients activity wasbelow quantifiable limits (33 nmolhg tissue) OverallGAA activity increased markedly during treatment to amean value of 205 mmolhg (n 16 range 490 to12983) This effect was larger in the 40-mgkg dose groupthan in the 20-mgkg dose group In the 40-mgkg dosegroup mean GAA levels increased from 000 to 31005

nmolhg whereas in the 20-mgkg dose group mean GAAlevels increased from 243 to 10030 nmolhg GAA activitylevels in individual patients at baseline and week 52 areshown in table 3

Baseline muscle glycogen content (measured biochemi-cally) ranged from 12 to 113 mgg tissue (wet weight)Fifteen of the 17 patients who had glycogen levels evalu-ated at week 52 exhibited stable (20) or decreased mus-cle glycogen content (20 glycogen depletion) ascompared to baseline Twelve of these 15 patients mademotor gains during the 52-week treatment period asshown in table 3 None of the three patients whose muscleglycogen content increased 20 from baseline duringtreatment showed any meaningful motor improvements

Adverse eventsinfusion-associated reactions Eleven ofthe 18 patients in this study experienced 164 IARs whichwere defined as treatment-related AEs that occurred fol-lowing infusion onset up to and including the postinfusionobservation period All the IARs were assessed as mild ormoderate in intensity none were severe The most com-mon IARs were rash (24 events) fever (23 events) urti-caria (22 events) and decreased oxygen saturation (18events) Overall IARs occurred in the following organ sys-tems skin (urticaria rash erythema livedo reticularisand pruritus) respiratory (cough tachypnea rales de-creased oxygen saturation) gastrointestinal (vomitingretching gastroesophageal reflux) cardiac (tachycardiacyanosis decreased heart rate) vascular (flushing hyper-tension hypotension pallor) psychiatric (agitation irrita-bility restlessness) nervous system (tremor) and generaldisorders (fever and rigors) IARs were typically managedby slowing or interrupting infusions and all 11 patientswho experienced IARs recovered without sequelae No pa-tient discontinued treatment because of IARs or AEs

One patient (Patient D) died of desaturation and brady-cardia following hospitalization for treatment of respira-tory distress and pneumonia at study week 61 (at the ageof 198 months) after the patient had completed 52 weeksof treatment but before the study ended The death wasassessed as being unrelated to rhGAA treatment The mostprominent histologic findings on autopsy were the pres-ence of glycogen in skeletal muscle (various samples fromupper and lower extremities and the diaphragm) and car-diomyocytes Glycogen was also evident in the smoothmuscle cell layers of the gastrointestinal tract particularlyupper esophagus and stomach Lesser amounts of glycogenwere present in vascular smooth muscle cells of arteriesand Schwann cell cytoplasm of nerves (chordae equina)Additionally brain tissue revealed periodic acidndashSchiff(PAS)ndashpositive material in vascular smooth muscle cells ofa small arteriole and occasional PAS-positive material inSchwann cell cytoplasm

Antibody responses Individual patientsrsquo rhGAA titersduring the treatment period are shown in figure 5 Sixteenof the 18 patients developed IgG antibodies to rhGAA Onepatient was seropositive at baseline suggesting preexist-ing cross-reactivity and two patients remained seronega-tive throughout the study Nine of the remaining patientsseroconverted by week 4 another four patients serocon-verted by week 8 one patient seroconverted by week 12and one patient seroconverted at week 64 One patient(Patient L) who developed high titers of IgG tested posi-tive for inhibitory antibodies at weeks 52 and 64 of treat-

Figure 4 Change from baseline to week 52 in Pompe Pedi-atric Evaluation of Disability Inventory scaled scoresOpen circles self-care (predominantly upper limb skills)open squares mobility (predominantly lower limb skills)filled circles social function (interactive play with adultsplay with objects) Scaled scores compare patientsrsquo perfor-mance against themselves at baseline The graph repre-sents average data from all patients (N 18) except atweek 12 where n 17


ment exhibiting 124 inhibition at week 52 and 232inhibition at week 64

Dose effects The proportions of patients who werealive free of invasive ventilation and showed improve-ments in left ventricular mass growth and motor develop-

ment were similar in the 20- and 40-mgkg dose groupsAlthough the clinical responses were similar in the twodose groups skeletal muscle GAA activity increased morein the 40-mgkg group (median 137 nmolhg tissue) thanin the 20-mgkg group (median 701 nmolhg tissue) at

Table 3 Skeletal muscle glycogen content and motor function

Patient

Dosegroup

(mgkg)GAA activity

(nmolhg tissue)

Percentage ofincrease in glycogen

content frombaseline to wk 52

Change inglycogencontent

Motorstatus atwk 52dagger

A 40 Baseline BQL 643 Increased No gains

Wk 52 1409

B 20 Baseline 88 22 Stable Walking

Wk 52 699

C 40 Baseline BQL 264 Decreased Walking

Wk 52 750

D 20 Baseline BQL 380 Decreased No gains

Wk 52 689

E 20 Baseline 85 496 Increased No gains

Wk 52 1388

F 40 Baseline BQL 486 Decreased Walking

Wk 52 740

G 40 Baseline BQL 532 Decreased Walking

Wk 52 1448

H 20 Baseline 46 264 Decreased Walking

Wk 67Dagger 263

I 20 Baseline BQL 193 Decreased Standing

Wk 52 490

J 40 Baseline BQL 775 Decreased Sitting

Wk 52 12983

K 20 Baseline BQL 497 Decreased Sitting

Wk 52 670

L 40 Baseline BQL ND ND Standing

Wk 52 ND

M 20 Baseline BQL 283 Decreased Standing

Wk 52 1527

N 40 Baseline BQL 295 Decreased Walking

Wk 52 1333

O 20 Baseline BQL 209 Decreased Walking

Wk 52 704

P 40 Baseline BQL 893 Decreased No gains

Wk 52 5588

Q 40 Baseline BQL 45 Stable Sitting

Wk 52 554

R 20 Baseline BQL 115 Stable No gains

Wk 52 1858

A change in glycogen content was designated as increased or decreased if the percentage of change from baseline was 20 Patientswith glycogen levels within 20 of their baseline values were designated as stable

dagger Motor status was determined by assessment of motor milestonesDagger Patient H underwent a skeletal muscle biopsy at week 67 because the patient was too ill to undergo the procedure at week 52

BQL below quantifiable levels (33 nmolhg tissue) ND not determined


week 52 Similarly glycogen levels in skeletal muscle asmeasured by biochemical analysis also showed a trend forgreater reductions in the 40-mgkg group (median decrease750 from baseline) as compared to the 20-mgkg dosegroup (median decrease 452 from baseline) Similarnumbers of patients in each dose group experienced IARshowever the five patients in the 20-mgkg dose group ex-perienced a total of 41 IARs and the six patients in the40-mgkg dose group experienced a total of 123 IARs Mostof the IARs (695) were experienced by three patientswho had recurrent reactions one in the 20-mgkg dosegroup and two in the 40-mgkg dose group Of the patientswho experienced IARs those in the 40-mgkg group hadmore symptoms per infusion during which IARs occurredthan those in the 20-mgkg dose group however theseevents were generally similar in nature in the two dosegroups

Genotyping and CRIM status Mutations in the GAAgene were analyzed in 15 patients (30 alleles) Elevennovel mutations were identified Each of the 15 patientswith GAA gene mutation data had a unique combination ofGAA mutations therefore it was not possible to draw con-clusions regarding the effect of specific GAA mutations onefficacy outcomes The genotyping data will be presentedin detail in a separate study (manuscript in preparation)Three of the 18 patients (Patients C L and R) were CRIMnegative as determined by Western blot the remaining 15patients were CRIM positive

Discussion This study examined the safety andefficacy of IV rhGAA treatment at 20 or 40 mgkgevery other week for 52 weeks in 18 patients withsevere infantile-onset Pompe disease Safety datawere analyzed for the duration of treatment (rangingfrom 52 to 106 weeks depending on when the pa-tient was enrolled) Our inclusion criteria were de-signed to select the most severe end of the phenotypespectrum of Pompe disease both in degree of GAAdeficiency2 and in the presence of cardiomyopathy5The clinical endpoints survival (with and withoutinvasive or any ventilation) left ventricular massdecrease and motor development are in accord withthe known natural history of the disease26 A previ-ous clinical trial with another form of rhGAA showedthat patients who started treatment before the oc-currence of extensive muscle damage had better mo-

tor outcomes than patients who began treatment atmore advanced stages of disease progression11

Therefore we enrolled patients who were aged 6months or younger at the onset of treatment

Pompe disease is progressively debilitating andoften fatal For patients with the most aggressiveform of the disease such as those enrolled in thisstudy death from cardiac andor respiratory failuretypically occurs before 12 months of age126 rhGAAtreatment markedly extended survival and improvedrespiratory performance in these patients Fifteen ofthe 18 patients reached the age of 18 months at theend of the study period The three patients who wereright-censored from the survival analysis were allalive at the end of the study (at ages 159 179 and144 months) In contrast only one of 61 historicalcontrol patients survived to the age of 18 months Sixof the 18 patients in the study required some type ofventilator support (three invasive three noninva-sive) at 18 months of age Our comparisons ofinvasive-ventilator-free or ventilator-free survival intreated patients to overall survival in the controlgroup of untreated patients represent a conservativeapproach large treatment vs control differences inthese endpoints were confirmed through numeroussensitivity analyses (data not shown) In addition aCox proportional hazard analysis demonstrated thatrhGAA treatment reduced the risk of death by 99reduced the risk of death or invasive ventilation by92 and reduced the risk of death or any type ofventilation by 88 as compared to the untreated his-torical control group (all significant with p 0001)

Echocardiographic measurements showed that rh-GAA treatment markedly improved cardiomyopathy(ie left ventricular mass reduction) in all patientswith 52-week follow-up data (n 15) This responseto treatment is similar to what has been observed inprevious clinical trials with rhGAA enzyme replace-ment therapy (ERT)7-11 and suggest that rhGAA treat-ment effectively reduces glycogen accumulation inhuman cardiac tissue In contrast cardiomyopathy andmassive cardiomegaly lead to fatal cardiorespiratoryfailure in a majority of untreated patients26 Giventhat cardiorespiratory failure is one of the primarycauses of death in patients with infantile-onset Pompedisease126 the improvement in cardiomyopathy ob-served in response to rhGAA treatment almost cer-tainly contributes to the prolonged survival

Infantile-onset Pompe disease is characterized byprofound muscle weakness which generally involvessubstantial hypotonia and flaccidity and is often re-ferred to as floppy baby syndrome1 Patients treatedwith rhGAA exhibited motor development that pro-gressed well beyond the point at which it becomesarrested in untreated patients26 In striking contrastto the nonattainment or loss of motor skills in pa-tients in the untreated historical control group 13 ofthe 18 treated patients acquired substantial motorand functional skills as assessed by the AIMS andPompe PEDI tests These skills including indepen-dent ambulation represent a much more advanced

Figure 5 Anti-rhGAA IgG antibody titers over time Notethat data are displayed only for nonzero antibody titers(Patients K and N did not display detectable antibody ti-ters at any time point)


stage of motor development than what is typicallyobserved in untreated patients26 The remaining fivepatients exhibited few motor and functional gainsover the 52-week treatment period However all fiveof these patients had very little if any residual mo-tor function at baseline and required some type ofventilation during the 52-week treatment periodwhich indicates an advanced stage of muscle involve-ment including the respiratory muscles Overall the18 patients in this study made more motor gainsthan patients in previous clinical trials with rhGAAERT7-11 possibly because ERT was initiated at suchan early age (and therefore at an earlier stage ofdisease progression) in this study

rhGAA treatment greatly increased skeletal mus-cle GAA activity in all patients but to a greaterdegree in patients receiving 40 mgkg rhGAA than inpatients receiving 20 mgkg rhGAA Fourteen of the17 patients who had glycogen levels evaluated atweek 52 exhibited stable or decreased muscle glyco-gen content as compared to baseline as determinedbiochemically Twelve of these 14 patients made mo-tor gains during the 52-week treatment period Noneof the three patients whose muscle glycogen contentincreased during treatment showed any meaningfulmotor improvements These findings suggest an as-sociation between the stabilization of or reduction inskeletal muscle glycogen content and improvements inmotor function in most but not all patients These re-sults are consistent with previous findings showingthat increases in GAA activity levels in a muscle groupdo not always correlate with glycogen clearance in thatmuscle and glycogen reduction in a given muscle can-not be generalized to other muscle groups11

No patients discontinued the rhGAA treatmentdue to unmanageable infusion-associated reactionssimilar to what has been observed in previous clini-cal trials with another form of rhGAA ERT7-11 IARsexperienced were managed with rate reductions in-terruptions of rhGAA infusions or symptomatictreatment such as antihistaminics acetaminophenor glucocorticosteroids

Similar numbers of patients in each dose groupexperienced AEs and IARs although more IARs(750) were experienced by the group of patientsassigned to the 40-mgkg dose than by those as-signed to the 20-mgkg dose of rhGAA Patients re-ceiving the 40-mgkg dose of rhGAA tended to developa more robust antibody response experienced moreIARs and when experiencing IARs had more symp-toms than the patients in the 20-mgkg dose groupThe timing of the first occurrence of IARs did notclosely coincide with the timing of seroconversion orthe magnitude of anti-rhGAA IgG titers There was noclear advantage of the higher dose on efficacy

Three of the 18 patients in this study wereCRIM negative (Patients C L and R) By week 52Patient C had learned to walk Patient L had learnedto stand and Patient R had made no motor gains In aprevious study of three patients with infantile-onsetPompe disease two patients who gained and then lost

motor skills were CRIM negative whereas the thirdCRIM-positive patient in that study steadily gainedmotor milestones7 However a study of four patientswith infantile-onset Pompe disease (one of whom wasCRIM negative) who were treated with rhGAA derivedfrom rabbit milk did not find an association betweenCRIM status and motor development23 Although thetotal number of patients studied remains small in-creasing evidence suggests that CRIM-negative pa-tients are more likely to develop a higher moresustained immunologic response against rhGAA thanCRIM-positive patients and potentially a more limitedduration of clinical benefit after rhGAA administra-tion In the current study one of the 18 patients devel-oped antibodies that inhibit in vitro rhGAA enzymeactivity which may represent a mechanism of antibodyinterference in humans The presence of inhibitory an-tibodies has previously been observed in patients whohave received ERT to treat other lysosomal storagediseases such as Gaucher and Fabry disease althoughit is an uncommon occurrence24-27 Clearly these obser-vations cannot be considered conclusive as the totalnumber of CRIM-negative patients studied to date isquite limited In addition to CRIM status and the de-velopment of inhibitory antibodies it is important tonote that other factors such as rate of disease progres-sion and residual functional status at onset of rhGAAtreatment may affect clinical outcomes in patientswith infantile-onset Pompe disease Thus long-termfollow-up is warranted in order to provide a better un-derstanding of the response to ERT

This study includes the largest cohort of patientswith Pompe disease treated with ERT to date Thesefindings show that rhGAA treatment clearly pro-longs ventilator-free survival and overall survival inpatients with infantile-onset Pompe disease as com-pared to an untreated historical control populationFurthermore this study demonstrates that initiationof ERT prior to 6 months of age which could befacilitated by newborn screening shows great prom-ise to reduce the mortality and disability associatedwith this devastating disorder

AcknowledgmentThe authors thank the patients who participated in this clinicalstudy and their families They also thank the study coordinatorsnurses and technical staff at the various treatment centers Theexpertise of Katherine Cox Jennifer Hunt Edward Kaye MDNancy Mulrow MD Dominique Nijkamp Tara OrsquoMeara KhazalParadis MD and Nancy Silliman PhD (from Genzyme Corpora-tion) also supported this study

References1 Hirschhorn R Reuser AJ Glycogen storage disease type II acid

-glucosidase (acid maltase) deficiency In Scriver CR Beaudet AL SlyWS et al eds The metabolic and molecular bases of inherited diseaseNew York McGraw-Hill 20013389ndash3420

2 van den Hout HMP Hop W van Diggelen OP et al The natural courseof infantile Pompersquos disease 20 original cases compared with 133 casesfrom the literature Pediatrics 2003112332ndash340

3 Kishnani PS Steiner RD Bali D et al Pompe disease diagnosis andmanagement guideline Genet Med 20068267ndash288

4 Chen YT Amalfitano A Towards a molecular therapy for glycogenstorage disease type II (Pompe disease) Mol Med Today 20006245ndash251


5 Slonim AE Bulone L Ritz S Goldberg T Chen A Martiniuk F Identi-fication of two subtypes of infantile acid maltase deficiency J Pediatr2000137283ndash285

6 Kishnani PS Hwu W-L Mandel H et al A retrospective multina-tional multicenter study on the natural history of infantile-onsetPompe disease J Pediatr 2006148671ndash676

7 Amalfitano A Bengur AR Morse RP et al Recombinant human acid-glucosidase enzyme therapy for infantile glycogen storage disease typeII results of a phase III clinical trial Genet Med 20013132ndash138

8 Van den Hout JM Reuser AJ de Klerk JB Arts WF Smeitink JA Vander Ploeg AT Enzyme therapy for Pompe disease with recombinanthuman alpha-glucosidase from rabbit milk J Inherit Metab Dis 200124266ndash274

9 Van den Hout JMP Kamphoven JHJ Winkel LPF et al Long-termintravenous treatment of Pompersquos disease with recombinant humanalpha glucosidase from milk Pediatrics [serial online] 2004113448ndash457 Available at httppediatricsaappublicationsorg Accessed August1 2006

10 Klinge L Straub V Neudorf U et al Safety and efficacy of recombinantacid alpha-glucosidase (rhGAA) in patients with classical infantilePompe disease results of a phase II clinical trial Neuromuscul Disord20051524ndash31

11 Kishnani PS Nicolino M Voit T et al Results from a phase II trial ofChinese hamster ovary cell-derived recombinant human acid-glucosidase in infantile-onset Pompe disease J Pediatr 200614989ndash97

12 Piper MC Darrah J Motor assessment of the developing infant Phila-delphia WB Saunders 1994

13 Feldman AB Haley SM Coryell J Concurrent and construct validity ofthe Pediatric Evaluation of Disability Inventory Phys Ther 199070602ndash610

14 Haley SM Fragala MA Aseltine R Ni P Skriner AM Development ofa disease-specific disability instrument for Pompe disease Pediatr Re-habil 2003677ndash84

15 Haley SM Fragala-Pinkham MA Ni PS Skriner AM Kaye EM Pedi-atric physical functioning reference curves Pediatr Neurol 200431333ndash341

16 Bayley N Bayley Scales of Infant Development 2nd ed New YorkHarcourt Brace 1993

17 Kaplan EL Meier P Nonparametric estimation from incomplete obser-vations J Am Stat Soc 195853457ndash481

18 Reuser AJ Koster JF Hoogeveen A Galjaard H Biochemical immuno-logical and cell genetic studies in glycogenosis type II Am J HumGenet 197830132ndash143

19 Cox DR Regression models and life tables (with discussion) J R StatistSoc Ser 1972B34187ndash220

20 Wandt B Bojo L Tolagen K Wranne B Echocardiographic assessment ofejection fraction in left ventricular hypertrophy Heart 199982192ndash198

21 Kuczmarski RJ Ogden LGrummer-Strawn LM et al CDC growthcharts United States Advance data from vital and health statisticsno 314 Hyattsville MD National Center for Health Statistics2000

22 Kamphoven JH de Ruiter MM Winkel LP et al Hearing loss ininfantile Pompersquos disease and determination of underlying pathology inthe knockout mouse Neurobiol Dis 20041614ndash20

23 Van den Hout H Reuser AJ Vulto AG Loonen MC Cromme-DijkhuisA Van der Ploeg AT Recombinant human alpha-glucosidase from rab-bit milk in Pompe patients Lancet 2000356397ndash398

24 Brady RO Murray GJ Oliver KL Leitman SF Fleisher TA BartonNW Management of neutralizing antibody to ceredase in a patient withtype 3 Gaucher disease Pediatrics 1997100E11

25 Ponce E Moskovitz J Grabowski GA Enzyme therapy in Gaucherdisease type 1 effect of neutralizing antibodies to acid b-glucosidaseBlood 19979043ndash48

26 Rosenberg M Kingma W Fitzpatrick MA Richards SM Immunosur-veillance of alglucerase enzyme therapy for Gaucher patients inductionof humoral tolerance in seroconverted patients after repeat administra-tion Blood 1999932081ndash2088

27 Zhao H Bailey LA Grabowski GA Enzyme therapy of Gaucherdisease clinical and biochemical changes during production of andtolerization for neutralizing antibodies Blood Cells Mol Dis 20033090ndash96

WWWNEUROLOGYORG OFFERS IMPORTANT INFORMATIONTO PATIENTS AND THEIR FAMILIES

The Neurology Patient Page provides

mdash a critical review of ground-breaking discoveries in neurologic research that are written especially for patients andtheir families

mdash up-to-date patient information about many neurologic diseasesmdash links to additional information resources for neurologic patients

All Neurology Patient Page articles can be easily downloaded and printed and may be reproduced to distribute foreducational purposes Click on the Patient Page icon on the home page (wwwneurologyorg) for a complete index ofPatient Pages


DOI 10121201wnl0000251268411880420076899-109 Published Online before print December 6 2006Neurology

P S Kishnani D Corzo M Nicolino et al Pompe disease

-glucosidase Major clinical benefits in infantile-onsetαRecombinant human acid

This information is current as of December 6 2006

ServicesUpdated Information amp

httpnneurologyorgcontent68299fullincluding high resolution figures can be found at

References httpnneurologyorgcontent68299fullref-list-1

This article cites 23 articles 5 of which you can access for free at

Citations httpnneurologyorgcontent68299fullotherarticles

This article has been cited by 28 HighWire-hosted articles

Subspecialty Collections

httpnneurologyorgcgicollectionmuscle_diseaseMuscle disease

httpnneurologyorgcgicollectionglycogenosesGlycogenoses

_cohort_case_controlhttpnneurologyorgcgicollectionclinical_trials_observational_studyClinical trials Observational study (Cohort Case control)

httpnneurologyorgcgicollectionall_clinical_trialsAll Clinical trialsfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

Pompe disease also known as glycogen storage dis-ease type II or acid maltase deficiency is a rareprogressively debilitating and often fatal lysosomalstorage disorder Patients with Pompe disease havean autosomal-recessively inherited deficiency of theenzyme acid -glucosidase (GAA) which hydrolyzesglycogen to glucose This GAA deficiency causes gly-cogen to accumulate in multiple tissues especiallyskeletal and cardiac muscle Glycogen deposits dis-rupt muscle cytoarchitecture and function causingprogressive motor respiratory and cardiacdysfunction1

Patient age at the onset of Pompe disease symp-toms and the rate of deterioration can vary consider-ably1 Patients with infantile-onset Pompe diseaseexperience symptoms within the first year of life andprogress rapidly to death In other patients symp-toms are exhibited later and the disease progressesmore slowly but it is still associated with significantmorbidity Patients with early onset of symptomstypically manifest hypotonia muscle weakness andcardiomyopathy within the first months of life Theoutlook for these patients is grim they fail to acquireor lose motor developmental milestones and cardio-respiratory failure and death usually occur before 1year of age2-4 Slonim et al5 described a subset ofinfantile-onset patients with less severe cardiac in-volvement (ie less severe cardiomyopathy and alack of cardiomegaly before the age of 6 months) whosurvived longer with ventilatory support However arecent retrospective chart review found early cardio-megaly even among first-birthday survivors6 In thishistorical cohort early symptom onset was associ-ated with early demise

In early clinical trials IV enzyme replacementwith different forms of recombinant human (rh) GAAhas improved survival cardiac and respiratory func-tion and motor development in severely affectedinfants7-11 A subset of these treated patientsachieved independent ambulation In a recent trial ofrhGAA derived from transfected Chinese hamsterovary cells eight of eight infants showed cardiac im-provement six were alive and five were free of inva-sive ventilation after 52 weeks of treatment and fiveattained new motor milestones11 Of the three pa-tients who eventually achieved the ability to walkindependently all had begun treatment prior to theage of 6 months suggesting that early treatmentmay be critical for the optimal motor response totreatment11 In this multicenter multinationalopen-label dose-ranging study we examined thesafety and efficacy of rhGAA treatment in a largercohort of severely affected patients with Pompe dis-ease who began treatment prior to 6 months of age

Methods Study design and treatment Local institutional re-view boards or independent ethics committees approved protocolsand consent forms at each of 13 primary sites (six centers in theUntied States five in Europe one in Taiwan and one in Israel)Parents or guardians gave written informed consent for patientsrsquoparticipation An independent data safety monitoring board re-viewed safety an independent allergic reaction review board was

consulted if required for issues related to serious infusion-associated reactions (IARs)

Eligible patients had documented symptoms of infantile-onsetPompe disease including skin fibroblast GAA activity 1 of thenormal mean and hypertrophic cardiomyopathy (left ventricularmass index 65 gm2 by echocardiogram) they were no older than26 weeks at enrollment Exclusion criteria included respiratoryinsufficiency (O2 saturation 90 or CO2 partial pressure 55mm Hg [venous] or 40 mm Hg [arterial] in room air or anyventilator use) a major congenital anomaly or clinically signifi-cant intercurrent illness unrelated to Pompe disease or any priorGAA treatment A total of 18 patients were studied

Parents or legal guardians were informed that the protocol-defined stopping rules related to the occurrence of serious adverseevents (AEs) and allowed for temporary or permanent patientdiscontinuation if the investigator safety review board and spon-sor determined there was a significant risk to patient safety

It was considered unethical to include a placebo group as partof the study design for the following reasons 1) infantile-onsetPompe disease is a rapidly fatal disorder and 2) early clinicaltrials have shown that treatment with various forms of rhGAAcan improve survival cardiac and respiratory function growthand motor development in severely affected infants7-11 Thus ahistorical control group of 61 severely affected infants 6 monthsold and younger was identified by applying the study inclusionand exclusion criteria to a group of 168 patients with infantile-onset Pompe disease identified through a retrospective chart re-view6 The cohort included 168 patients from nine countries 33different sites with birth dates ranging from before 1985 to 20026

Drug effects on overall survival (with or without ventilator use)and invasive ventilator-free survival were compared betweentreated patients and this historical cohort our group of 18 treatedpatients provided 98 power to detect a difference betweentreated and control groups Table 1 summarizes the screeningcriteria applied to the original historical patient population inorder for patients to be included in the historical control subgroup

Alglucosidase alfa which contains rhGAA produced in Chinesehamster ovary cells was supplied by Genzyme Corporation Asimple randomization scheme was used to assign patients to thetwo dose groups Upon confirmation that patients met all eligibil-ity criteria and completion of pretreatment screening and baselineassessments eligible patients were randomized in a 11 ratio toreceive an IV infusion of either 20 mgkg or 40 mgkg of rhGAAevery other week Clinicians and patients were not blinded to thedose of rhGAA used The total amount of rhGAA administeredcould be adjusted every 4 weeks to account for changes in bodyweight

Clinical assessments of safety and efficacy Patients were ob-served and their vital signs were monitored during rhGAA infu-sion and for 2 hours afterward Safety assessments includedevaluating blood and urine chemistry (data not shown) the occur-rence and timing of the development of anti-rhGAA IgG antibod-ies (assayed at first rhGAA infusion and every 4 weeksthereafter) and the incidence and nature of AEs including IARs

Efficacy was measured by assessing survival ventilator useleft ventricular mass by echocardiography growth (weight andlength) muscle GAA activity and glycogen levels motor develop-ment (using the Alberta Infant Motor Scale [AIMS]12) and level ofdisability (using an adaptation of the Pediatric Evaluation of Dis-ability Index [PEDI]13 developed for Pompe disease1415 Althoughnot an efficacy endpoint cognitive development (using BayleyrsquosScales of Infant Development Second Edition [BSID-II]) was alsoserially evaluated16 Echocardiograms and glycogen content inmuscle biopsy samples were centrally read by a pediatric cardiol-ogist and a pathologist who were blinded to dose patient andtime point Motor and cognitive evaluations were centrally scoredby a nonblinded central clinician

Survival and ventilation data were analyzed up to 18 monthsof age as compared to survival of the historical control group allother efficacy data were analyzed with respect to changes frombaseline after 52 weeks of treatment The primary efficacy end-point was the Kaplan-Meier17 proportion of patients alive and freeof invasive ventilation at 18 months of age Safety data wereanalyzed for the duration of treatment Because patients wereenrolled over a period of 1 year safety data ranged from 52 weeksfor the last patient randomized to treatment to 106 weeks for thefirst patient treated with rhGAA















Inclusion criteria


Yes 115 (685)

No 53 (315)


Yes 112 (667)

No 56 (333)


Yes 153 (911)

No 15 (89)

Exclusion criteria


Yes 0

No 168 (1000)


Yes 1 (06)

No 167 (994)


Yes 12 (71)

No 156 (929)


Yes 37 (220)

No 131 (780)


Yes 25 (149)

No 143 (851)


Yes 62dagger (369)

No 106 (631)


















Parameter



n 62)


(n 168)

Gender n ()

Male 11 (611) 28 (452) 83 (494)

Female 7 (389) 34 (548) 85 (506)

Race n ()

White 7 (389) 31 (500) 80 (476)

Black 4 (222) 4 (65) 22 (131)

Hispanic 2 (111) 1 (16) 1 (06)

Asian 3 (167) 18 (290) 52 (310)

Other 2 (111) 1 (16) 1 (06)

Unknown 00 7 (113) 12 (71)


n 18 62 166

Mean 16 19 27

Median 10 16 20

SD 178 179 248

Min max 00 55 00 59 00 120


n 18 62 165

Mean 37 36 60

Median 43 41 47

SD 217 194 883

Min max 02 68 44 66 51 842


n 18 NA NA

Mean 46 NA NA

Median 53 NA NA

SD 167 NA NA

Min max 12 61 NA NA


n NA 61sect 163


Min max NA 03 439 16 569

Kishnani et al6



NA not applicable
























Patient

Dosegroup

(mgkg)GAA activity

(nmolhg tissue)






Wk 52 1409


Wk 52 699


Wk 52 750


Wk 52 689


Wk 52 1388


Wk 52 740


Wk 52 1448


Wk 67Dagger 263


Wk 52 490


Wk 52 12983


Wk 52 670


Wk 52 ND


Wk 52 1527


Wk 52 1333


Wk 52 704


Wk 52 5588


Wk 52 554


Wk 52 1858











































































Reprints

















Inclusion criteria


Yes 115 (685)

No 53 (315)


Yes 112 (667)

No 56 (333)


Yes 153 (911)

No 15 (89)

Exclusion criteria


Yes 0

No 168 (1000)


Yes 1 (06)

No 167 (994)


Yes 12 (71)

No 156 (929)


Yes 37 (220)

No 131 (780)


Yes 25 (149)

No 143 (851)


Yes 62dagger (369)

No 106 (631)


















Parameter



n 62)


(n 168)

Gender n ()

Male 11 (611) 28 (452) 83 (494)

Female 7 (389) 34 (548) 85 (506)

Race n ()

White 7 (389) 31 (500) 80 (476)

Black 4 (222) 4 (65) 22 (131)

Hispanic 2 (111) 1 (16) 1 (06)

Asian 3 (167) 18 (290) 52 (310)

Other 2 (111) 1 (16) 1 (06)

Unknown 00 7 (113) 12 (71)


n 18 62 166

Mean 16 19 27

Median 10 16 20

SD 178 179 248

Min max 00 55 00 59 00 120


n 18 62 165

Mean 37 36 60

Median 43 41 47

SD 217 194 883

Min max 02 68 44 66 51 842


n 18 NA NA

Mean 46 NA NA

Median 53 NA NA

SD 167 NA NA

Min max 12 61 NA NA


n NA 61sect 163


Min max NA 03 439 16 569

Kishnani et al6



NA not applicable
























Patient

Dosegroup

(mgkg)GAA activity

(nmolhg tissue)






Wk 52 1409


Wk 52 699


Wk 52 750


Wk 52 689


Wk 52 1388


Wk 52 740


Wk 52 1448


Wk 67Dagger 263


Wk 52 490


Wk 52 12983


Wk 52 670


Wk 52 ND


Wk 52 1527


Wk 52 1333


Wk 52 704


Wk 52 5588


Wk 52 554


Wk 52 1858











































































Reprints
















Parameter



n 62)


(n 168)

Gender n ()

Male 11 (611) 28 (452) 83 (494)

Female 7 (389) 34 (548) 85 (506)

Race n ()

White 7 (389) 31 (500) 80 (476)

Black 4 (222) 4 (65) 22 (131)

Hispanic 2 (111) 1 (16) 1 (06)

Asian 3 (167) 18 (290) 52 (310)

Other 2 (111) 1 (16) 1 (06)

Unknown 00 7 (113) 12 (71)


n 18 62 166

Mean 16 19 27

Median 10 16 20

SD 178 179 248

Min max 00 55 00 59 00 120


n 18 62 165

Mean 37 36 60

Median 43 41 47

SD 217 194 883

Min max 02 68 44 66 51 842


n 18 NA NA

Mean 46 NA NA

Median 53 NA NA

SD 167 NA NA

Min max 12 61 NA NA


n NA 61sect 163


Min max NA 03 439 16 569

Kishnani et al6



NA not applicable
























Patient

Dosegroup

(mgkg)GAA activity

(nmolhg tissue)






Wk 52 1409


Wk 52 699


Wk 52 750


Wk 52 689


Wk 52 1388


Wk 52 740


Wk 52 1448


Wk 67Dagger 263


Wk 52 490


Wk 52 12983


Wk 52 670


Wk 52 ND


Wk 52 1527


Wk 52 1333


Wk 52 704


Wk 52 5588


Wk 52 554


Wk 52 1858











































































Reprints


























Patient

Dosegroup

(mgkg)GAA activity

(nmolhg tissue)






Wk 52 1409


Wk 52 699


Wk 52 750


Wk 52 689


Wk 52 1388


Wk 52 740


Wk 52 1448


Wk 67Dagger 263


Wk 52 490


Wk 52 12983


Wk 52 670


Wk 52 ND


Wk 52 1527


Wk 52 1333


Wk 52 704


Wk 52 5588


Wk 52 554


Wk 52 1858











































































Reprints










































































Reprints




recombinant human acid -glucosidase - neurology · skeletal muscle weakness, and early death. the...

Documents