ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r. com/ loca te /yc l im

Clinical Immunology (2007) 123, 139–147

SHORT ANALYTICAL REVIEW

Management options for adenosine deaminasedeficiency; proceedings of the EBMT satelliteworkshop (Hamburg, March 2006)Claire Booth a,b, Mike Hershfield c, Luigi Notarangelo d,1, Rebecca Buckley e,Manfred Hoenig f, Nizar Mahlaoui g,h, Marina Cavazzana-Calvo g,h,Alessandro Aiuti i, H. Bobby Gaspar a,b,⁎

a Molecular Immunology Unit, Institute of Child Health, University College London, London WC1N 1EH, UKb Department of Clinical Immunology, Great Ormond Street Hospital NHS Trust, London WC1N 3JH, UKc Department of Medicine, Duke University Medical Center, Durham, NC 27710, USAd Università di Brescia, Dept. of Pediatrics, Piazzale Civili, 1, 25123 Brescia, Italye Department of Pediatrics, Duke University Medical Center, Durham, NC 27710, USAf Department of Pediatrics, University of Ulm, Eythstrasse 24, 89073 Ulm, Germanyg Department de Biothérapie, d'Hématologie Pédiatriques, AP-HP Hôpital Necker, Université René Descartes Paris V,75743 Paris Cédex 15, Franceh Unité d'Immunologie et d'Hématologie Pédiatriques, AP-HP Hôpital Necker, Université René Descartes Paris V,75743 Paris Cédex 15, Francei San Raffaele Telethon Institute for Gene, Therapy (HSR-TIGET), Milan, Italy

Received 8 November 2006; accepted with revision 7 December 2006Available online 14 February 2007

Abbreviations: ADA, adenosine deaPEG, polyethylene glycol; MSD, matcheImmunodeficiencies; EBMT, European G⁎ Corresponding author. Molecular Im1 Current address: Division of Immun

1521-6616/$ – see front matter © 200doi:10.1016/j.clim.2006.12.009

Abstract Adenosine deaminase (ADA) deficiency is a disorder of purine salvage that has its mostdevastating consequences in the immune system leading to severe combined immunodeficiency(SCID). Management options for ADA SCID include hematopoietic stem cell transplantation, enzymereplacement therapy and gene therapy. Formal data on the outcome following each of the threetreatment modalities are limited, and this symposium was held in order to gather together theexperience from major centers in Europe and the US. Transplantation for ADA-SCID is highlysuccessful with survival rates of∼90% if a matched sibling ormatched related donor is available butsurvival following matched unrelated donor or haploidentical procedures is 63% and 50% respec-tively with a significant rejection/non-engraftment rate in unconditioned procedures. Successfullytransplanted patients demonstrated good immunological recovery with normal cellular andhumoral function in the majority of cases. PEG-ADA has been used in over 150 patients worldwideeither as an alternative to mismatched transplant or as a stabilizing measure prior to transplant.

KEYWORDSAdenosine deaminasedeficiency;Hematopoietic stemcell transplantation;PEG-ADA;Gene therapy

minase; SCID, severe combined id sibling donor; MFD, matched famroup for Blood and Marrow Transmunology Unit, Institute of Childology, Children's Hospital, Harvard

6 Elsevier Inc. All rights reserved

mmunodeficiency; HSCT, hematopoietic stem cell transplantation;ily donor; MUD, matched unrelated donor; ESID, European Society ofplantation.Health, University College London, London WC1N 1EH, UK.Medical School, Boston, USA.

.

Overall, approximately two thirds of patients treated with PEG-ADA have survived with themajority of patients showing good clinical improvement. The level of immune recovery longterm was less than that seen after transplant and ∼50% of patients continued to receiveimmunoglobulin replacement. Gene therapy has been used as an experimental procedure intwo centers in Europe. Early results from 9 patients suggest that the treatment is safe and thatthe majority have shown recovery of cellular immune function. Long-term follow-up of treatedpatients highlights a significant incidence of non-immunological problems with cognitive,neurological and audiological abnormalities most prominent.© 2006 Elsevier Inc. All rights reserved.

140 C. Booth et al.

Introduction

Adenosine deaminase-deficient severe combined immunode-ficiency (ADA-SCID) is a rare primary immunodeficiencycharacterized by severe and recurrent opportunistic infection,failure to thrive and metabolic abnormalities. The immunolo-gical phenotype is characterized by profound lymphopenia andvery low immunoglobulin levels of all isotypes. ADA-SCIDaccounts for ∼10–15% of all cases of SCID and the incidence isestimated to be between 1:200,000 and 1:1,000,000 live births[1]. Unlike many other forms of SCID, the systemic nature ofADA expression results in non-immunological abnormalities ina significant proportion of children, who may have neurode-velopmental deficits [2], sensori-neural deafness [3] orskeletal abnormalities [4]. A single gene defect is responsiblefor the condition and the ADA gene has been mapped tochromosome 20q13.11 [5] leading to an autosomal recessivepattern of disease inheritance.

ADA is a purine salvage enzyme expressed in all tissuesof the body and catalyses the deamination of deoxydeno-sine (dAdo) and adenosine (Ado) to deoxyinosine andinosine respectively [6]. The lack of ADA in patients withADA deficiency results in elevation in levels of Ado anddAdo. An increase in the intracellular conversion of dAdo todeoxyadenosine trisphosphate (dATP), with resultingexpansion of the dATP pool, causes several potentiallydeleterious effects, including abnormalities of lymphocytedevelopment, viability and function, and is the majorcause of lymphocyte depletion and the resulting immuno-logical defects. Studies on ADA-deficient patients andmurine models have shown that dATP pool expansion caninhibit ribonucleotide reductase [7], an enzyme necessaryfor DNA replication and repair [8], induce apoptosis inimmature thymocytes and can interfere with terminaldeoxynucleotidyl transferase (TdT) activity thereby limitingantigen receptor diversity [9].

The immunological phenotype in ADA-SCID is character-ized by an absence of T and B lymphocytes and, in mostcases, absent NK cells. The diagnosis is usually made byfinding low or undetectable ADA activity in erythrocytesand can be confirmed by raised levels of dATP and reducedS-adenosyl homocysteine hydrolase (SAHH) activity in redcells and elevated amounts of deoxyadenosine in urine [1].ADA gene mutation analysis is available in certain centers[10]. Prenatal diagnosis can be undertaken either throughmutation analysis or measurement of enzyme activity introphoblasts cultured from chorionic villus sampling or incultured amniocytes [11].

Management options for ADA-SCID

Three management options exist for ADA-SCID: allogeneichematopoietic stem cell transplantation (HSCT), enzymereplacement therapy (ERT) and somatic gene therapy (GT).HSCT has been the mainstay of therapy for many years.Survival following allogeneic matched family donor HSCT is73% but the 3-year survival following mismatched andhaploidentical transplants is approximately 29% [12],although the survival rate is higher in some centers ([13]and Buckley, unpublished data). The above data stretch backto 1968 and it is likely that the survival rate for patientsundergoing HSCT in the present day from either donor sourceis likely to be higher. There are presently no survival outcomedata on the use of unrelated or umbilical cord blood trans-plants, both ofwhich procedures have been used in increasingnumbers for immunodeficiencies over the past decade.

Enzyme replacement therapy (ERT) with pegylatedadenosine deaminase enzyme has been used in the treat-ment of ADA deficiency since 1987 [14]. Purified bovine ADAis conjugated with polyethylene glycol to decrease clearanceof the compound. PEG-ADA is administered weekly or twiceweekly by intramuscular injection. Cellular uptake of PEG-ADA is not significant, but maintaining high ectopic ADAactivity in plasma (>100 fold normal) eliminates extracel-lular Ado and dAdo, leading to normalization of intracellulardATP levels. Metabolic correction is then followed over a 2–4 month period by cellular and humoral immune reconstitu-tion. The extent of immune recovery is variable and althoughmany children recover full immunity in the short term, asignificant number (∼50%) remain on immunoglobulinreplacement. Over a longer time period, some patientsshow a decline in T-cell numbers and remain lymphopenic[15]. Despite this, follow-up suggests that most patientsremain clinically well and free of infection with normalgrowth parameters.

A number of gene therapy trials for ADA deficiency wereinitiated in the early 1990s [16–19]. Trials in the US andEurope used retroviral mediated gene transfer of the ADAgene into different autologous cellular fractions (peripheralblood lymphocytes, umbilical cord blood, bone marrow andCD34+ selected stem cells). In all these studies PEG-ADA wascontinued in addition to gene therapy and the effect of genetherapy alone on correction of the immune parameters wasnot clearly demonstrated. More recently two studies [20,21]have shown gene therapy to be efficacious in correcting ADAdeficiency. The process appears to be more successful if PEG-ADA is withdrawn prior to gene therapy to allow a selective

141Management options for ADA deficiency

advantage to gene transduced cells [22]. A mild conditioningregime was also used to allow engraftment of a greaternumber of gene modified cells.

With three different treatment modalities available, theexact role of each is yet to be determined. A study on thepractices of various different centers in Europe suggeststhat, when a matched related donor is available, all wouldproceed with HSCT (Gaspar, unpublished data). However,there is considerable discrepancy in practice when amatched related donor is not available. Making an informedchoice is also hampered by the lack of formal data and, inmost centers, treatment decisions are often based onpersonal experience. This symposium was organized toaddress these difficult issues and attempted to drawtogether the management experience from major centersin Europe and the USA.

Allogeneic HSCT for management of ADA-SCID

The HSCTexperience from 4major centers in Europe and 1 UScenter was presented; the results of which are summarizedin Table 1.

The survival following MSD/MFD transplants is extremelygood and most probably relates to the lack of pre-transplantconditioning and the rapid engraftment of mature T cellsfrom the full donor graft. Only 8 MUD transplants wereperformed at the 4 centers and, in all cases, pre-treatmentchemotherapy was used. The 3 deaths were from Gram-negative sepsis, CMV and adenoviremia and unexplainedsepsis with collapse at the time of conditioning. The poorestresults follow mismatched or haploidentical family donorgrafts. The majority of these transplants in Europe wereundertaken before 1990 and deaths were related to viralpneumonitis, GvHD and aspergillosis. As a result of early poorexperience, there have been fewer haploidentical trans-plants for ADA-SCID performed since that time, and becauseof its availability and lower toxicity profile, PEG-ADA hasbeen used as an alternative in all of the four above Europeancenters. In Ulm, haploidentical transplant is still seen as analternative to long-term PEG-ADA therapy.

Immune reconstitution in survivors of HSCT has in generalbeen very effective. In the 11 survivors of MSD/MFDtransplants in the London series, follow-up at a mean of

Table 1 HSCT transplant outcome for ADA-SCID

Total MSD/MFD MUD

Number Alive a Number A

London 31 13 11 5 3Ulm 15 b 7 7 2 1Brescia 4 3 3 1 1Paris 7 3 2 0 0Duke c 23 4 4 0 0Total 80 30 27 (90%) 8 5

MSD—matched sibling donor.MFD—matched family donor.MUD—matched unrelated donor.a All patients alive with T-cell engraftment.b Two patients from Munich.c All patients transplanted without conditioning.

6.25 years shows mean ALC (1970 cells/mm3), CD3+ (1407cells/mm3), CD4+ (607 cells/mm3) and CD19+ (254 cells/mm3)numbers in the normal range and 10 of 11 patients havediscontinued immunoglobulin replacement. Similarly in theUlm series, analysis of immune parameters in the 12 long-term survivors of all transplants (mean follow-up of11.7 years) shows a mean CD3+ T-cell count of 1576 cells/mm3 with normal T-cell proliferation to mitogen stimulation.B-cell recovery is more variable with a mean count 393 cells/mm3 but all patients have specific antibody responses totetanus and diphtheria vaccines and all have stoppedimmunoglobulin replacement. A mean ALC of 1400 cells/mm3 was seen in patients surviving HSCT in Brescia at 2 yearsfollow-up. Chimerism data from Ulm and London demon-strate that chimerism is 100% following conditioned trans-plants but recipients of an unconditioned MSD/MFDtransplant exhibit 100% donor chimerism only in T cells withthe remaining cell lineages predominantly of recipient origin.

Normalization of metabolic abnormalities is achievedeffectively after HSCT. In a series of 11 patients in Londonwho had received either MSD/MFD or haploidentical trans-plant, there was reduction of dATP levels from a mean of1352 μmol/l at diagnosis to a mean of 103 μmol/l correctionfollowing HSCT (Fig. 1). These levels are not normal since redcell dATP is undetectable in normal individuals but these datarepresent a significant reduction in metabolite levels.

Long-term experience with treatment ofADA-deficient SCID by unconditioned HSCT

Twenty-three ADA-deficient SCID patients have been trans-planted at Duke University Medical Center in Durham, NorthCarolina in the period from November 1984 until the presenttime. Patients did not receive pre-transplant chemotherapyor post-transplant GvHD prophylactic drugs. The types oftransplant included 4 HLA-identical related and 19 rigorouslyT-cell-depleted haploidentical parental marrow transplants.Currently 18 patients (78%) survive from 3 to 21.9 years post-transplantation. The 5 deaths occurred from viral infections(4 cases) and pulmonary hypertension [1]. All deaths were inpatients who received haploidentical transplants. Elevenchildren (48%) are successfully immune reconstituted byrelated donor bone marrow transplantation, including the 4

Haploidentical

livea Number Alive Alive+T engraftment

13 3 36 4 40 04 0

19 14 7(63%) 42 21 (50%) 14 (33%)

Table 2 PEG-ADA treatment outcome

PEG-ADAtreatment

Progressionto othertreatments(HSCT/GT)

Alive onPEG-ADAalone

Deaths onPEG-ADA

London 12 6/2 2 2Ulm 3 2 1 0Brescia 8 2/2 4 0Paris 8 0 7 1Total 31 14 14 3

Figure 1 Levels of erythrocyte dATP in 11 patients pre- andpost-HSCT.

142 C. Booth et al.

who received HLA-identical transplants and in 7 of the 19(36.8%) who received T-cell-depleted haploidentical paren-tal transplants. Ten of these eleven engrafted children do notrequire IVIg and all 11 have excellent T-cell function. Five ofthe surviving but non-engrafted recipients of haploidenticaltransplants are being treated with PEG-ADA and arecandidates for gene therapy in a recently initiated US ADAgene therapy trial [23]. The other two have already receivedsuccessful gene therapy from the Milan group in Italy (seebelow). Considering the high survival rate with thisapproach, Duke University Medical Center advocates per-forming either an HLA-identical or haploidentical parentalbone marrow transplant from a relative without pre-transplant chemotherapy or post-transplantation GvHDprophylaxis as first-line therapy, using PEG-ADA only ifengraftment does not occur. This approach may change asgene therapy becomes more readily available.

PEG-ADA therapy for management of ADA-SCID

ERT with PEG-ADA (Adagen, Enzon Inc; obtained outsideNorth America through Orphan Europe) has been used for20 years to treat ADA-deficient SCID patients who lack anMFD and are often too ill to undergo a conditioned, non-MFDtransplant. About 30% of ERT patients have been olderchildren with a delayed onset phenotype, who are often poortransplant candidates because of chronic pulmonary disease,and because residual immune function may interfere withengraftment. PEG-ADA has also been a secondary therapy inseveral patients after failed unconditioned haploidenticaltransplants. During its first years of use, a number of casereports, described the immunologic and clinical response toERT, generally with relatively short follow-up (reviewed in[15]). The majority of patients receiving ERT as their onlytherapy recovered partial, but protective immune function,and showed good clinical improvement.

Approximately 150–160 patients worldwide have nowreceived PEG-ADA, and ∼90 are currently under treatment.Approximately 65% of the latter have received PEG-ADA for>5 years, and one third for 10–19 years. Tracking long-termclinical and immunologic status has been more difficult thanfor transplantation since many ERT patients, at least in theUS, are not followed for long periods at single locations,owing to the high mobility of patients and their physicians.Recently two reports on the long-term immune reconstitu-tion during PEG-ADA therapy have been published, involving5 patients treated in Italy [24] and 9 in California [25].

The laboratory at Duke University has served as a resourcefor biochemical monitoring of ERT and for analyzing ADAgenotype. Data held at Duke and information from thephysicians caring for patients have allowed a summary of theoverall experience with PEG-ADA, including information onage at start of ERT, length of therapy, survival, progression totransplantation and significant complications. The lastupdate covered the period from 1986 through to 2002 [15].

A preliminary review of information available through tothe end of 2005 indicates that approximately two thirds of allPEG-ADA treated patients (n=150–160) survive, includingabout half of those patients (15 out of 29) who eventuallydiscontinued PEG-ADA to undergo transplantation. Mostdeaths while receiving PEG-ADA have been in the first 6months of treatment and mainly within the first month dueto infections present at diagnosis. Deaths beyond 2 years ofERT have been related mainly to progression of pulmonarydisease already present when ERT was initiated. Threepatients have died from lymphoproliferative disorders diag-nosed after receiving PEG-ADA for 8 to 15 years [15,25,26].

ERT has generally been well tolerated and no allergic orhypersensitivity reactions to PEG-ADA have been reported.About 6–8% of patients, mostly with a delayed or late onsetphenotype, have developed neutralizing IgG antibodies thatreduce or eliminate the efficacy of PEG-ADA [15]. Transientimmune dysregulation may occur during the first months ofERT, but several patients have developed persistent hemo-lytic anemia, which in some cases started during a viralinfection or an episode of central catheter sepsis [15,27].Although PEG-ADA is effective and life-saving, there isconcern that immune function may eventually becomeinadequate beyond 10–15 years of treatment. Since it isnot curative, the expense of ERT is a barrier to its use in somecountries, although the cost is comparable to that forreplacement therapy in other rare inherited disorders.

European center experience of PEG-ADA

At present, there is no formal guideline or consensus on whenand for how long PEG-ADA should be used. The data from thefour European centers (see Table 2) suggest that, in somecenters, PEG-ADA is used primarily to stabilize a child beforeprogression to MUD HSCTor GTor while an appropriate donoris identified. In some centers (Paris, London, Brescia), PEG-ADA is also seen as an alternative to haploidenticaltransplant and faced with this donor option, PEG-ADA iscontinued indefinitely. Of the three deaths seen on PEG-ADA

143Management options for ADA deficiency

alone, two occurred within 3 months of starting PEG-ADA andin both cases were related to viral (in one case CMV)encephalitis. A third patient died of RSV pneumonitis after apoor immunological response to PEG-ADA.

The metabolic detoxification after starting PEG-ADA wasrapid and effective with reduction in erythrocyte dATPlevels in all treating centers. dATP levels following PEG-ADA treatment are in general lower than that seenfollowing HSCT. Immune reconstitution in PEG-ADA treatedpatients was variable. In the first 6 months after starting ageneral increase in the ALC (absolute lymphocyte count)was seen including a rapid but transient increase in B-cellnumbers. Over a longer period, a similar experience wasseen in Paris, London and Brescia. Patients remainedclinically well but showed a gradual reduction in ALC andT-cell counts. Of 8 patients treated in Paris, 7 are alive onPEG-ADA (mean follow-up=8 years) with a median CD3+

count of 650 cells/mm3 and a median CD4+ count of 400cells/mm3 and all 7 show good proliferative responses toPHA and 3 remain on Ig replacement. Two of the eightpatients showed a poor response in terms of absolutelymphocyte counts and 1 died of RSV pneumonitis. The CD4+

responses to PEG-ADA in this Paris cohort are seen in Fig. 2.Data from the London cohort suggest that, in the first yearafter starting PEG-ADA, there is initiation of thymopoiesis asdemonstrated by an increase in TREC (T-cell receptor excisioncircles) numbers and in naive T cells. However, in 5 patientstreated for a more prolonged period of time in Brescia, TRECnumbers were reduced in comparison to normal controlssuggesting compromise of thymic function over a longer term[24].

A number of adverse events were seen in patientsreceiving PEG-ADA therapy. Two children experiencedsevere Coomb's positive autoimmune hemolytic anemiaalthough a viral etiology could not be excluded. Anotherchild developed vasculitis and macrophage activationsyndrome requiring treatment with steroids and withdrawalof PEG-ADA. These three individuals proceeded to genetherapy. One child developed anti-PEG-ADA antibodieswhich resolved after a course of steroids and an increasein the dose of PEG-ADA. Three children suffered a transientthrombocytosis with no clinical sequelae and one childexperienced a transient acute neurological disorder withcomplete recovery, but the relationship of this event toPEG-ADA therapy is not clear.

Figure 2 Reconstitution of CD4+ T cells with time afterinitiation of PEG-ADA therapy.

European survey of PEG-ADA treatment

The lack of formal data on the outcome of treatment withPEG-ADA, particularly in terms of short-term and long-termimmune recovery, prompted the ESID/EBMT Inborn ErrorsWorkshop to produce a questionnaire which was sent tomajor centers around Europe in November 2005. Preliminarydata from 42 patient questionnaires (including those patientspresented in Table 2) were presented. In 27 patients PEG-ADA was started in the first 6 months of life, 6 patientsstarted at between 6 and 12 months of age and 8 patientscommenced treatment in the second year of life. Approxi-mately 60% of patients received only PEG-ADA with theremainder progressing to receive HSCTs or gene therapy. Fourdeaths have been reported while, on PEG-ADA alone, thesebeing due to CMV viremia, encephalitis, pneumonitis and RSVinfection. Immune recovery is variable with low levels ofCD3+ (460 cells/mm3) and CD4+ cells (258 cells/mm3) inpatients treated for longer than 12 months (n=15, meanfollow-up=76.8 months). In the first year of starting PEG-ADA, 75% of patients received immunoglobulin (Ig) replace-ment but after 12 months of treatment only 40% remained onIg therapy. Overall survival was 85% in patients receivingPEG-ADA alone (n=26), 70% in those who went onto HSCT(n=10) and 100% in those treated by gene therapy (n=7).These preliminary data are the subject of further analysis asmore datasets are received.

Gene therapy for ADA-SCID

Gene therapy for ADA-SCID in Europe has been undertaken inMilan and London and both centers presented their data. Inboth studies, retroviral vectors encoding the ADA gene wereused to modify autologous HSCs. Prior to return of genemodified cells, patients were given a low dose conditioningregime to facilitate engraftment of corrected cells. Con-ditioning protocols differed with the use of Busulphan 4 m/kgin Milan and a single of dose of Melphalan at 140 mg/m2 inLondon. In both trials, PEG-ADA was either stopped or insome cases not started prior to treatment by gene therapy tofully exploit the selective advantage for gene correctedcells. Milan have now enrolled 8 children, 4 of whom hadreceived PEG-ADA prior to gene therapy and 3 of whom hadundergone a previously unsuccessful haploidentical HSCT([20] and unpublished data). One child had not received anytreatment prior to gene therapy. The mean age at genetherapy was 2.2 years (range 0.6–5.5 years). All childrenenrolled in the trial are currently healthy and thriving, withthe longest follow-up reaching 64 months. In the six childrenwith a follow-up >6 months after gene therapy, vector-ADA+

cells became progressively the large majority of T, B and NKlymphocytes. The patients have now stable engraftment ofgene corrected cells with 0.1%–10% of myeloid cells contain-ing vector-ADA (unpublished data). There does appear to bea correlation between the number of CD34+ cells infused andthe percentage of myeloid engraftment observed. Genetherapy has resulted in a substantial increase of lymphocytecounts, development of polyclonal thymopoiesis and normal-ization of T-cell functions. Currently in half of the childrenIVIg infusion has been discontinued with evidence of specificantibody production, while two other patients have planned

144 C. Booth et al.

discontinuation. One child has suffered from varicellainfection and overcame this without complication. ADAenzyme activity was restored in lymphocytes and improvedin RBC, leading to a substantial decrease in the levels of toxicdAXP metabolites in RBC, and clinically all patients havenormal development thus far. The patient who was treatedwith the lowest stem cell dose has received PEG-ADA in thepast months in the attempt to provide full detoxification, butthere has been no change in lymphocyte number or immunefunctions. Finally, vector integrations were highly hetero-geneous in circulating T cells with no evidence of clonalexpansion. No adverse events or toxicity related to genetherapy have been observed.

In London, one patient has been treated [21]. This childreceived PEG-ADA for 3 years but showed a gradual decline inT-cell numbers and lack of thymic activity. PEG-ADA wasstopped 1 month prior to treatment. Two years after genetherapy, there has been an increase in T-cell numbers,normalization of proliferative responses and re-initiation ofthymopoiesis. The patient is clinically well and has stoppedprophylactic antibiotic therapy. Metabolic detoxification wascomparable to that seen after HSCT with an initial burst ofRBC ADA activity followed by a decline to levels seen post-transplant.

Non-immunological consequences of ADA-SCID

Despite good immune reconstitution following HSCTand PEG-ADA therapy, all centers noted a high incidence of non-immunological problems in patients during long-term follow-up. Significant cognitive and behavioral abnormalitiespersist and London has documented a significant reductionin both verbal and performance IQ levels (patients fallingmore than 2 standard deviations below the control group) incomparison to a case-matched controlled group of non-ADA-SCIDs undergoing HSCT [2]. There was no difference inchimerism, type of conditioning or donor origin to explainthe difference between the groups. Increased levels of dATPshowed an inverse correlation with IQ and are perhapsindicative of the systemic nature of the disease. In theLondon series bilateral sensori-neural, high frequency hear-ing loss was found in 58% of those treated [3].

In Ulm, 6 of 12 long-term survivors showed neurologicaland behavioral problems with all exhibiting developmentaldelay. The most seriously affected patient suffers frommuscular hypertonia severe enough to require a wheelchair.In the majority of cases hypertonia, gait abnormalities,reduced verbal skills (despite normal hearing) and behavioralproblems were observed although no standardized testingwas performed. Neuro-imaging was uninformative in rele-vant cases. Two children require continuous care and all ofthe above 6 patients as well as another one who has learningdifficulties attend special needs schools. One child suffersfrom seizures. Similar to the London experience, four of theseven children suffered from sensori-neural deafness.

In Munich a pair of identical twins showed very differentneurological outcomes despite receiving identical treatmentby HSCT (see Case 3 discussion below). There was noevidence of CNS infection or adverse neonatal events. Thetwins received HSCTs at the same time, following the sameconditioning, exhibited identical chimerism and spent the

same time in hospital. The neurological problems describedare not seen frequently in other forms of SCID.

An unusual non-immunological complication was describedby a number of groups. One patient in Brescia, aged 20 monthsand transplanted at 4 months of age developed multiplecalcified nodules over her body. These lesions were fixed tounderlying soft tissue and appeared at areas of microtrauma.There have been reports in the past of children with ADA-SCIDdeveloping fibrosarcoma, although biopsy in this particularcase ruled out this particular diagnosis. Prof. Hershfielddescribed the case of two sisters from Indiana with unusualADA mutations (late onset disease) who developed recurrentfibrosarcomas. Dr. Gaspar has also observed a case in London ofa child initially thought to have a lipoma which was excised.This recurred and histology pointed to a dermatofibrosarcomaprotuberanswith the child eventually receiving chemotherapy.

Case discussions

Case 1

Dr. Speckman (University of Freiburg) described a 21-year-old woman who presented with moderate immunodefi-ciency and immune dysregulation. She had remainedhealthy until 3 years of age when she developed recurrentbacterial sinopulmonary infections. Since 5 years of age shehad experienced autoimmune hemolytic anemia (treatedwith cyclosporin), severe eczema and elevated IgE levels(>20,000 U/ml). She also exhibited persistent lymphopeniawith few thymic naive cells, impaired T-cell function andsignificantly reduced TRECs. Immunoglobulin levels werenormal. She was initially misdiagnosed with MDS (frombone marrow findings) and a subsequent diagnosis of ADAdeficiency was made. Recurrent infections were not hermain clinical problem and she required only one admissionwith pneumonia over the past few years.

Dr. Speckman asked the panel whether this patient wouldbenefit from PEG-ADA as she had not yet been started ontreatment and was clinically well. Both Prof. Hershfield andDr. Gaspar agreed that pregnancy might be a difficult time interms of autoimmunity. One London center (Royal FreeHospital) had experience of two late onset ADA deficiencypatients; both women in the third decade of life [28]. Bothreceived PEG-ADA at diagnosis but gained no significantbenefit with regard to immune recovery. Their clinicalproblems deteriorated during pregnancy and both hadmore severe lung disease than the patient in Freiburg. Oneof the patients from London died of chronic lung disease andthe other still has significant respiratory impairment. In thiscase the center stopped PEG-ADA due to costs and thenrepeated a trial of PEG-ADA for 1 year. No improvement wasobserved. Prof. Hershfield described a 16-year-old patient inCalifornia presenting with autoimmune phenomena andchronic lung disease (and oxygen requirement) whichimproved after PEG-ADA. The patients described by Dr.Gaspar and Prof. Hershfield are still alive and do not receiveintravenous immunoglobulin replacement (IVIg).

The overall opinion of the panel was to continue presentconservative management. It was suggested that the patientbe monitored closely regarding her immunological para-meters. If her clinical or immunological status should change

145Management options for ADA deficiency

(especially with worsening of her pulmonary disease), thepanel suggested that a trial of PEG-ADA, initially over a pre-defined period of time (at least 6 months), may beappropriate. The panel was cautious in recommending PEG-ADA therapy since her severe autoimmunity might worsenwith ERT, and it cannot be excluded that clones contributingto autoimmunity might expand and become moreproblematic.

Case 2

Dr. Huck (Düsseldorf) described an 8-year-old Somalian boywho presented with a history of recurrent pneumonia,septicemia and pseudomonal skin abscesses. He had alsosuffered from severe varicella infection. Three siblings haddied in Somalia from infection in the first year of life. Sincethe age of two he had received IVIg and prophylacticantibiotics. Immunophenotyping revealed absent B cells andlow CD4+ T cells. Levels of IgA and IgM were normal with lowIgM and raised IgE and T-cell function is impaired. Clinicallyhowever, the child had remained well other than a chroniccough. No HLA-identical donor had been found but two cordshad been identified (5/6 and 4/6 matches). He is notreceiving PEG-ADA due to significant cost.

There was no consensus regarding treatment for this boybut if the economic barrier were not a major limitation, atrial of PEG-ADA may be useful. Dr. Gaspar advised againstHSCT due to the poor outcome seen in mismatchedtransplants and agreed that PEG-ADA would be useful andwould not preclude entry into a gene therapy trial at alater stage. Looking towards the future it was anticipatedthat the cost of gene therapy would be similar to an HSCTbut probably less than the cost for 1 year of treatment withPEG-ADA.

Case 3

Identical twins, briefly discussed in an earlier section, werepresented by Dr. Albert (Munich). They presented at 4monthsof age with identical phenotypes and no neurologicalimpairment. They underwent a split MSD graft 8 years agoon the same day, without conditioning or graft versus hostdisease (GvHD) prophylaxis. Several months after transplantthe first signs of developmental delay were observed in bothgirls. One sister is mildly affected with behavioral abnorm-alities and deficits in motor, verbal and social skills. Hersister is severely delayed with refractory seizures and nospeech or motor skills. MRI scan reveals no abnormalities.Viral PCR has not been performed on CSF. Firstly, Dr. Gasparsuggested that the twins undergo full neurological, meta-bolic and radiological investigation to rule out any co-existing pathology. Prof. Hershfield knew of several exam-ples of siblings with discordant phenotype. Dr. Gasparspeculated whether PEG-ADA post-transplant might changethe neurological status but was concerned it might affectimmune recovery. Prof. Notarangelo has experience of PEG-ADA administration in patients who had poor immunerecovery post-HSCT in an attempt to boost immunity.Although ERT did not correct immune function in thesepatients, it did not affect engraftment either. It wassuggested that PEG-ADA therapy be administered for a

defined period of time with close monitoring of allparameters including biochemical markers.

Concluding remarks

The workshop raised several issues relating to both manage-ment options for ADA-SCID and the disease itself. DifferentEuropean centers have different approaches to the manage-ment of ADA-SCID. Experience from Duke University favorsHSCT (of any type) prior to the initiation of PEG-ADA therapyas transplant offers hope of a permanent correction. InEurope, centers proceed to HSCT if an MSD/MFD is available.If an MSD/MFD is not available, most centers favor stabiliza-tion of the patient on PEG-ADA followed by HSCT, or morerecently gene therapy. There is no evidence that priortreatment with PEG-ADA is associated with graft failure.Haploidentical transplants are not regularly performed inEurope for this condition due to poor historical results. Datasuggest that the outcome following matched sibling andmatched family transplants is good and GvHD is not a majorproblem. There are insufficient data currently on outcomefollowing MUD HSCT or following haploidentical transplantsbut a questionnaire-based survey sent out to major centerswill help provide evidence in the near future. If childrensurvive HSCT (family donors or conditioned haploidenticaltransplants) reconstitution of both cellular and humoralimmunity is good and well sustained at 5 years aftertransplant.

PEG-ADA has now been used for over 20 years and ∼60patients have been treated in Europe. PEG-ADA appears toprovide a better overall survival outcome than mismatchedtransplants but despite better metabolic detoxification,long-term immune reconstitution is lower than that follow-ing HSCT. Some patients respond more effectively to PEG-ADA, and at present it is unclear as to the determinants ofthe response. This may relate to the degree of thymicreserve at the time of PEG-ADA initiation and thus it may beimportant to identify those patients who will benefit mostfrom enzyme replacement, perhaps by evaluating TRECactivity or other markers of thymic function. Funding is anissue for many centers but as suggested several times duringthis workshop a trial of PEG-ADA for a pre-defined period oftime at diagnosis may be very useful. Gene therapy offers anexciting new treatment modality and potential curealthough follow-up is short in comparison to the two othertreatment modalities. Gene therapy trials are also ongoing inthe USA (Children's Hospital Los Angeles, NIH) [23] and inJapan (Hokkaido University School of Medicine), and thecumulative results from the 4 different centers will be highlyinformative. Gene therapy appears to be safe and efficaciousin patients treated so far, but the long-term effects must beevaluated. In the medium term, immune recovery iscomparable to that seen after transplant and in the futurethe cost may also be less than HSCT.

Late onset ADA-SCID poses a difficult problem in terms ofmanagement. Often patients are relatively well and suffermore from immune dysregulation, with manifestation ofautoimmune symptoms, than from severe infections. Under-taking an HSCT in such patients is risky and may bedetrimental to their quality of life. In such cases a trial ofPEG-ADA may be valuable. There does appear to be a

146 C. Booth et al.

genotype/phenotype correlation for ADA-SCID and this mayalso help in treatment decisions for late onset patients.

A fairly consistent finding among all groups was thepresence of sensori-neural deafness and neurological andbehavioral abnormalities which may be disease specific andunrelated to treatment. ADA-SCID is a systemic disease andas children are, on the whole, very young and severely ill atpresentation, developmental and neurological assessment atdiagnosis can be difficult. It is therefore unclear what roleviral infection (particularly of the CNS) plays in theneurological and behavioral abnormalities seen, and it isimportant that these patients should be investigatedextensively for evidence of such infections. It is also difficultto ascertain to what degree toxic metabolites (dATP) areresponsible for complications observed. It is believed thatbetter metabolic detoxification will achieve a greaterclinical benefit but it is impossible to ascertain how muchdamage has been done prenatally.

Recently adenosine has been implicated in remodelingof airways and this may lead us to a greater understandingof pulmonary disease also seen in ADA-SCID [29]. Chest X-rays of patients who present early in life with lung diseaseare characteristic of pneumonitis but despite extensiveinvestigation a pathogen may not be isolated. This mayrepresent intrinsic pulmonary pathology specific to thiscondition. Several centers have observed soft tissuesarcomas in patients with ADA-SCID. The pathogenesis ofthis is unknown; there may be a relation with certain ADAgene mutations. In any patient with SCID who develops softtissue swellings biopsy should be considered. More informa-tion is needed on the systemic nature of ADA-SCID and theoutcome associated with various treatment options. Genetherapy adds another dimension to management of thiscondition and further formal evidence is required in orderto identify which patients will benefit most from thedifferent modalities.

Acknowledgments

This meeting was sponsored and organized by OrphanEurope, distributors of ADAGEN (PEG-ADA).

Declared interests: Dr Gaspar has received consultancyfees from Enzon Inc. Prof. Hershfield has received con-sultancy fees and an educational grant from Enzon Inc.

References

[1] M.S. Hershfield, B.S. Mitchell, Immunodeficiency diseasescaused by adenosine deaminase deficiency and purinenucleoside phosphorylase deficiency, in: C.R. Scriver, A.L.Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic andMolecular Basis of Inherited Disease, McGraw-Hill, NewYork, 1995, pp. 1725–1768.

[2] M.H. Rogers, R. Lwin, L. Fairbanks, B. Gerritsen, H.B. Gaspar,Cognitive and behavioral abnormalities in adenosine deaminasedeficient severe combined immunodeficiency, J. Pediatr. 139(2001) 44–50.

[3] W. Albuquerque, H.B. Gaspar, Bilateral sensorineural deafnessin adenosine deaminase-deficient severe combined immuno-deficiency, J. Pediatr. 144 (2004) 278–280.

[4] S.D. Cederbaum, I. Kaitila, D.L. Rimoin, E.R. Stiehm, Thechondro-osseous dysplasia of adenosine deaminase deficiency

with severe combined immunodeficiency, J. Pediatr. 89 (1976)737–742.

[5] M.B. Petersen, L. Tranebjaerg, N. Tommerup, P. Nygaard, H.Edwards, New assignment of the adenosine deaminase genelocus to chromosome 20q13×11 by study of a patient withinterstitial deletion 20q, J. Med. Genet. 24 (1987) 93–96.

[6] R. Hirschhorn, Overview of biochemical abnormalities andmolecular genetics of adenosine deaminase deficiency, Pediatr.Res. 33 (1993) S35–S41.

[7] N. Lee, N. Russell, K. Ganeshaguru, B.F. Jackson, A. Piga, H.G.Prentice, R. Foa, A.V. Hoffbrand, Mechanisms of deoxyadeno-sine toxicity in human lymphoid cells in vitro: relevance to thetherapeutic use of inhibitors of adenosine deaminase, Br. J.Haematol. 56 (1984) 107–119.

[8] S.G. Apasov, M.R. Blackburn, R.E. Kellems, P.T. Smith, M.V.Sitkovsky, Adenosine deaminase deficiency increases thymicapoptosis and causes defective Tcell receptor signaling, J. Clin.Invest. 108 (2001) 131–141.

[9] L. Gangi-Peterson, D.H. Sorscher, J.W. Reynolds, T.B. Kepler,B.S. Mitchell, Nucleotide pool imbalance and adenosinedeaminase deficiency induce alterations of N-region inser-tions during V(D)J recombination, J. Clin. Invest. 103 (1999)833–841.

[10] F.X. Arredondo-Vega, I. Santisteban, S. Daniels, S. Toutain, M.S.Hershfield, Adenosine deaminase deficiency: genotype–phe-notype correlations based on expressed activity of 29 mutantalleles, Am. J. Hum. Genet. 63 (1998) 1049–1059.

[11] T. Dooley, L.D. Fairbanks, H.A. Simmonds, C.H. Rodeck, K.H.Nicolaides, P.W. Soothill, P. Stewart, G. Morgan, R.J. Levinsky,First trimester diagnosis of adenosine deaminase deficiency,Prenat. Diagn. 7 (1987) 561–565.

[12] C. Antoine, S. Muller, A. Cant, M. Cavazzana-Calvo, P. Veys,J. Vossen, A. Fasth, C. Heilmann, N. Wulffraat, R. Seger,S. Blanche, W. Friedrich, M. Abinun, G. Davies, R. Bredius,A. Schulz, P. Landais, A. Fischer, Long-term survival andtransplantation of haemopoietic stem cells for immunodefi-ciencies: report of the European experience 1968–99, Lancet361 (2003) 553–560.

[13] R.H. Buckley, Molecular defects in human severe combinedimmunodeficiency and approaches to immune reconstitution,Annu. Rev. Immunol. 22 (2004) 625–655.

[14] M.S. Hershfield, R.H. Buckley, M.L. Greenberg, A.L. Melton, R.Schiff, C. Hatem, J. Kurtzberg, M.L. Markert, R.H. Kobayashi,A.L. Kobayashi, et al., Treatment of adenosine deaminasedeficiency with polyethylene glycol-modified adenosine dea-minase, N. Engl. J. Med. 316 (1987) 589–596.

[15] M.S. Hershfield, Combined immune deficiencies due to purineenzyme defects, in: E.R. Stiehm, H.D. Ochs, J.A. Winkelstein(Eds.), Immunologic Disorders in Infants and Children, W.B.Saunders, Philadelphia, 2004, pp. 480–504.

[16] R.M. Blaese, K.W. Culver, A.D. Miller, C. Carter, T.A. Fleisher, M.Clerici, G. Shearer, L. Chang, Y. Chiang, P. Tolstochev, J.J.Greenblatt, S.A. Rosenberg, H. Klein, M. Berger, C.A. Mullen,W.J. Ramsey, L. Muul, R.A. Morgan, W.F. Anderson, Tlymphocyte-directed gene therapy for ADA-SCID: initial trialresults after 4 years, Science 270 (1995) 475–480.

[17] C. Bordignon, L.D. Notarangelo, N. Nobili, G. Ferrari, G.Casorati, P. Panina, E. Mazzolari, D. Maggioni, C. Rossi, P.Servida, Gene therapy in peripheral blood lymphocytes andbone marrow for ADA-immunodeficient patients, Science 270(1995) 470–475.

[18] D.B. Kohn, K.Weinberg, J.A. Nolta, L.N. Heiss, C. Lenarsky,G.M.Crooks, M.E. Hanley, G. Annett, J.S. Brooks, A. El-Khoureiy, K.Lawrence, S. Wells, R.C. Moen, J. Bastian, D.E. Williams-Herman, M. Elder, D. Wara, T. Bowen, M. Hershfield, C.A.Mullen, R.M. Blaese, R. Parkman, Engraftment of gene-modifiedumbilical cordblood cells in neonateswith adenosine deaminasedeficiency, Nat. Med. 1 (1995) 1017–1023.

147Management options for ADA deficiency

[19] P.M. Hoogerbrugge, V.W. van Beusechem, A. Fischer, M.Debree, F. Le Deist, J.L. Perignon, G. Morgan, H.B. Gaspar,L.D. Fairbanks, C.H. Skeoch, A. Moseley, M. Harvey, R.J.Levinsky, D. Valerio, Bone marrow gene transfer in threepatients with adenosine deaminase deficiency, Gene Ther.(1996) 179–183.

[20] A. Aiuti, S. Slavin, M. Aker, F. Ficara, S. Deola, A. Mortellaro, S.Morecki, G. Andolfi, A. Tabucchi, F. Carlucci, E. Marinello, F.Cattaneo, S. Vai, P. Servida, R. Miniero, M.G. Roncarolo, C.Bordignon, Correction of ADA-SCID by stem cell gene therapycombined with nonmyeloablative conditioning, Science 296(2002) 2410–2413.

[21] H.B. Gaspar, E. Bjorkegren, K. Parsley, K.C. Gilmour, D. King, J.Sinclair, F. Zhang, A. Giannakopoulos, S. Adams, L.D. Fairbanks,J. Gaspar, L. Henderson, J.H. Xu-Bayford, E.G. Davies, P.A.Veys, C. Kinnon, A.J. Thrasher, Successful reconstitution ofimmunity in ADA-SCID by stem cell gene therapy followingcessation of PEG-ADA and use of mild preconditioning, Mol.Ther. 14 (2006) 505–513.

[22] A. Aiuti, S. Vai, A. Mortellaro, G. Casorati, F. Ficara, G. Andolfi,G. Ferrari, A. Tabucchi, F. Carlucci, H.D. Ochs, L.D. Notar-angelo, M.G. Roncarolo, C. Bordignon, Immune reconstitutionin ADA-SCID after PBL gene therapy and discontinuation ofenzyme replacement, Nat. Med. 8 (2002) 423–425.

[23] B.C. Engel, G.M. Podsakoff, J.L. Ireland, E.M. Smogorzewska,D.A. Carbonaro, K. Wilson, A.J. Shah, N. Kapoor, M. Sweeney,M. Borchert, G.M. Crooks, K.I. Weinberg, R. Parkman, H.M.Rosenblatt, S.Q. Wu, M.S. Hershfield, F. Candotti, D.B. Kohn,Prolonged pancytopenia in a gene therapy patient with ADA-

deficient SCID and trisomy 8 mosaicism—a case report, Blood109 (2006) 503–506.

[24] F. Malacarne, T. Benicchi, L.D. Notarangelo, L. Mori, S. Parolini,L. Caimi, M. Hershfield, L.D. Notarangelo, L. Imberti, Reducedthymic output, increased spontaneous apoptosis and oligoclo-nal B cells in polyethylene glycol-adenosine deaminase-treatedpatients, Eur. J. Immunol. 35 (2005) 3376–3386.

[25] B. Chan, D. Wara, J. Bastian, M.S. Hershfield, J. Bohnsack, C.G.Azen, R. Parkman, K. Weinberg, D.B. Kohn, Long-term efficacyof enzyme replacement therapy for adenosine deaminase(ADA)-deficient severe combined immunodeficiency (SCID),Clin. Immunol. 117 (2005) 133–143.

[26] D.A. Kaufman, M.S. Hershfield, J.A. Bocchini, I.J. Moissidis, M.Jeroudi, S.L. Bahna, Cerebral lymphoma in an adenosinedeaminase-deficient patient with severe combined immunode-ficiency receiving polyethylene glycol-conjugated adenosinedeaminase, Pediatrics 116 (2005) e876–e879.

[27] E. Lainka, M.S. Hershfield, I. Santisteban, P. Bali, A. Seibt, J.Neubert, W. Friedrich, T. Niehues, Therapy provides temporaryimmune reconstitution to a child with delayed-onset ADAdeficiency, Clin. Diagn. Lab. Immunol. 12 (2006) 861–866.

[28] C.L. Shovlin, H.A. Simmonds, L.D. Fairbanks, S.J. Deacock,J.M. Hughes, R.I. Lechler, A.D. Webster, X.M. Sun, J.C.Webb, A.K. Soutar, Adult onset immunodeficiency caused byinherited adenosine deaminase deficiency, J. Immunol. 153(1994) 2331–2339.

[29] M.R. Blackburn, R.E. Kellems, Adenosine deaminase defi-ciency: metabolic basis of immune deficiency and pulmonaryinflammation, Adv. Immunol. 86 (2005) 1–41.