Hematopoietic stem cell transplantation improves the high incidence of neutralizing allo-antibodies observed in Hurler’s syndrome after pharmacological enzyme replacement therapy

Abstract

Background Mucopolysaccharidosis type I is caused by deficiency of α-L-iduronidase. Currently available treatment options include an allogeneic hematopoietic stem cell transplant and enzyme replacement therapy. Exogenous enzyme therapy appears promising but the benefits may be attenuated, at least in some patients, by the development of an immune response to the delivered enzyme. The incidence and impact of alloimmune responses in these patients remain unknown.Design and Methods We developed an immunoglobulin G enzyme-linked immunosorbent assay as well as in vitro catalytic enzyme inhibition and cellular uptake inhibition assays and quantified enzyme inhibition by allo-antibodies. We determined the impact of these antibodies in eight patients who received enzyme therapy before and during hematopoietic stem cell transplantation. In addition, 20 patients who had previously received an allogeneic stem cell transplant were tested to evaluate this treatment as an immune tolerance induction mechanism.Results High titer immune responses were seen in 87.5% (7/8) patients following exposure to α-L-iduronidase. These patients exhibited catalytic enzyme inhibition (5/8), uptake inhibition of catalytically active enzyme (6/8) or both (4/8). High antibody titers generally preceded elevation of previously described biomarkers of disease progression. The median time to development of immune tolerance was 101 days (range, 26–137) after transplantation. All 20 patients, including those with mixed chimerism (22%), tested 1 year after transplantation were tolerized despite normal enzyme levels.Conclusions We found a high incidence of neutralizing antibodies in patients with mucopolysaccharidosis type I treated with enzyme replacement therapy. We also found that allogeneic hematopoietic stem cell transplantation was an effective and rapid immune tolerance induction strategy.

Introduction

Mucopolysaccharidosis type I (MPSI) is a lysosomal storage disorder (LSD) caused by deficiency of α-L-iduronidase (IDUA). The deficiency of IDUA results in intracellular accumulation of dermatan sulphate (DS) and heparan sulphate and a progressive, multisystem clinical disorder. Recombinant human enzyme replacement therapy (ERT) for LSD first became available in the 1990s1 and is currently used clinically for MPSI, MPSII, MPSVI, Pompe, Gaucher and Fabry disease at a considerable financial cost (estimated at $150,000–300,000 per patient per annum in MPSI).2 It is essential that the ERT is functionally active (catalytic activity) and able to target and penetrate enzyme-deficient cells. In MPSI, enzyme is taken up by cells via a mannose-6-phosphate receptor-mediated mechanism.

The formation of allo-antibodies has been reported to occur in all LSD treated with ERT.3 The clinical importance of antibodies in ERT-treated patients is uncertain but the existence of an alloimmune response to infused exogenous enzyme and recombinant human proteins is well described.4–5 In early clinical trials the reported incidence of IDUA-specific antibodies was low (40%) and there was no evidence of immunoprecipitation of enzyme or inhibition of catalytic activity.6 Subsequently, a prospective, open-label, multinational study revealed a suboptimal biomarker response in the presence of high antibody titers (> 1:10,000) when compared to that in patients with no alloimmune response.7 More recent studies in canine models of MPSI demonstrated up to 90% inhibition of enzyme uptake by MPSI fibroblasts by serum taken from dogs that had high titer anti-IDUA antibodies.8 There is increasing evidence, mostly derived from animal models of LSD, of an inverse correlation between an observed antibody response and metabolic and clinical outcome. Even though the immune response currently reported in patients with MPSI is 91%,9 the true incidence of functionally active (neutralizing) antibodies is unknown.

The consequences of a refractory immune response can be serious, ranging from treatment failure to catastrophic rapid progression of disease and high mortality in some LSD.7,10–12 Several clinical protocols for the induction of immune tolerance to exogenous enzyme have been reported for these diseases and others, including hemophiliac disorders13 in which there is a similar host immune response against a foreign protein. Allogeneic hematopoietic stem cell transplantation (HSCT) can replace the recipient’s immune system with that of the donor, thereby tolerizing the individual to the replaced enzyme (pharmacological or cellular). The donor’s immune system is naturally tolerized to this protein as the donor possesses normal levels of the enzyme. However, data on allogeneic HSCT as a means of immune tolerance induction are scarce.

In children with severe MPSI (MPSI H, Hurler’s syndrome) the standard of care is allogeneic HSCT. Engrafted donor leukocytes deliver enzyme (cellular ERT) to the brain. Pharmacological ERT is unable to correct neurological disease because of its inability to cross the blood-brain barrier. It has been our practice recently to give ERT to patients with MPSI H in the interval between diagnosis and transplantation in order to improve the somatic (including cardiac) manifestations of the disease and to prepare the child for transplant conditioning therapy.14 We have developed quantitative and functional immunological assays and studied the incidence, pattern and impact of the immune response in MPSI patients prior to receiving any treatment and following ERT and HSCT. We have also determined the relationship between these antibody responses and metabolic biomarkers of the disease.

Design and Methods

Patients’ samples

Blood samples were collected from 28 MPSI H patients treated and followed up at the Royal Manchester Children’s Hospital over a 4-year period. The samples were collected with informed consent and under permission REC 08H101063 of the hospital’s ethical commission. The patients’ blood samples were processed within 6 h and the serum was separated and stored at −80°C. One group of patients was followed longitudinally and comprised patients who received ERT prior to HSCT (n=8). The second was a cross-sectional group of patients (n=20) studied at least 1 year after HSCT. One patient (patient 5) rejected the donor cells following HSCT and had autologous cells returned to restore hematologic function. This patient subsequently received a second allogeneic HSCT which was successful. Longitudinal data for this patient during both transplants are included in the results. The demographics, enzyme levels, details of HSCT, immune reconstitution, and immunosuppression for all patients in the longitudinal series are presented in Table 1. The transplant details and information on immunosuppression for all patients in the cross-sectional series are shown in Table 2.

Immunoglobulin G enzyme-linked immunosorbent assay

For the immunoglobulin G (IgG) enzyme-linked immunosorbent assay (ELISA), 96-well plates were coated with recombinant human IDUA enzyme (Aldurazyme) (Genzyme, Framingham, USA) in enzyme-carbonate solution (10 μg/mL in 0.1 M sodium bicarbonate, pH 8.5), overnight and blocked with blocking agent containing 1% human serum albumin. The plates were washed with washing buffer [phosphate-buffered saline (PBS), 0.1% Tween at pH 7] three times and patient’s serum (50 μL) was added (two-fold serial dilutions) in duplicate and incubated at room temperature for 1 h. The plates were then washed three times with washing buffer, and goat anti-human IgG-horseradish peroxidase antibody (1:5000 dilutions in dilution buffer containing PBS, 0.05% Tween, 0.01% human serum albumin) was added and incubated at room temperature for 1 h. Plates were washed again, o-phenylenediamine dihydrochloride solution was added and the reaction stopped with 2.5 M H₂SO₄. The plates were read at 492 nm immediately. The normal range was determined for each dilution by testing 13 normal sera (from normal volunteers). A positive cut-off value was defined as two standard deviations above the absorbance value for the normal sera at that concentration. To quantify ELISA, the lowest titer with absorbance above the cut-off was reported as a positive serum dilution for that patient’s sample. Positive responses were confirmed by western blotting to show the specificity of the IgG antibody to Aldurazyme.

Catalytic inhibition

Functional enzyme activity was measured using 4-methylumbelliferyl α-L-iduronide (4MU) substrate (Glycosynth, Warrington, UK) at 37°C based on previously described methods.15–16 We modified the protocol to develop and optimize mixing assays in 96-well plates and assessed inhibition of enzyme activity by patients’ sera.

The assay was performed in duplicate on the highest titer patient’s sample in the longitudinal series. A series of two-fold dilutions (starting from 30 ng/mL to optimize the catalytic inhibition) of Aldurazyme in dilution buffer (PBS, 0.05% Tween, 0.01% human serum albumin) were mixed with the same volume of patient’s sera or normal sera in serial dilutions in 96-well black plates (Sterilin, South Wales, UK). The sera were added at fixed dilutions (1:4) and incubated at room temperature for 2 h in an incubator kept under agitation. The 4MU substrate (20 μL) in substrate buffer (0.4 M formate buffer, pH 3.5, 0.9% NaCl) was added to each well and incubated for 1 h at room temperature in the dark. The enzyme activity in the absence and presence of sera (for patient’s serum and normal serum) was measured. Normal serum was used as a standard to compare and quantify the inhibition caused by the antibodies in the patients’ sera. The enzyme activity in the presence of a patient’s serum was compared with the enzyme activity of the plain enzyme solution for each dilution. We also used IDUA from human embryonic kidney cells (HEK) and mouse liver cells to assess the inhibition of innate enzyme by patients’ sera.

The percentage inhibition of the catalytic activity for each enzyme dilution was determined as follows:

where EA_Ps is the enzyme activity in the presence of the patient’s serum, EA_Ns is the enzyme activity in the presence of normal serum, BEA_Ps is the background enzyme activity in the patient’s serum and BEA_Ns is the background enzyme activity in normal serum.

Cellular uptake inhibition

This assay determines the inhibition of mannose-6-phosphate-mediated enzyme uptake and subsequent activity by binding antibodies in non-FCR expressing, IDUA-deficient fibroblasts.

Table 1.Patients’ demographics (longitudinal series of patients).

MPSI fibroblasts (MPSI-A171) from a patient with Hurler’s syndrome (genotype W402X/W402X) were established and maintained in Dulbecco’s modified Eagle’s medium, 10% fetal bovine serum, 1% glutamine at 37°C in 5% CO₂. The assays were performed in duplicate. Six-well culture plates were seeded with 1.5×10 fibroblasts/well and grown to 95% confluence. The enzyme was diluted in culture medium (Dulbecco’s modified Eagle’s medium with 1% glutamine) to below half the Km for Aldurazyme (100 ng/mL). Patient’s serum or normal serum was added at a volume of 10 μL (1 in 100 dilutions) to the diluted enzyme solution (1000 μL volume) and incubated for 2 h at room temperature. The culture medium was replaced with the enzyme mix medium and incubated for 1 h in 5% CO₂ at 37°C. The cells were washed with PBS and harvested by trypsinization. The cells were re-suspended and washed twice with PBS and a final suspension was made in homogenizing buffer (0.5 M NaCl/0.02 M Tris pH 7–7.5). Following a cycle of freezing and thawing, the cells were sonicated and centrifuged at 2045 g for 10 min. The supernatant was tested for enzyme activity using the 4MU assay described earlier. The protein concentration was determined using a bicinchoninic acid assay, as described elsewhere.17 No uptake inhibition was observed with normal sera and hence the percentage inhibition was calculated against plain enzyme solution as follows:

Biomarkers and chimerism analysis

The ratio of DS to chondroitin sulphate (CS), a glycosaminoglycan (GAG) which is not stored in cells in MPSI, is a better marker of disease progression than total urinary GAG.18–21 We, therefore, studied DS/CS in our longitudinal series of patients. Total urinary GAG and DS/CS ratio were determined using two dimensional electrophoresis (originally described by Whiteman22) and LabWorks software (Anachem, Luton, UK).19 Donor engraftment was evaluated in our clinical laboratory through analysis of variable number tandem repeats using a UVP gel photography system and LabWorks software.

Table 2.Patients’ demographics (cross-sectional series of patients).

Results

Pattern of immune response

In the longitudinal series of patients, high titers of IDUA reactive antibodies were seen in 87.5% (n=7) of patients during IgG ELISA. Figure 1A shows the immune response after first exposure to ERT up to the time of HSCT in these patients. One patient (patient 8) had detectable IgG antibodies only at a low titer (1:256). These patients were followed up for a median period of 216 days (range, 155–685). The median time between first exposure to ERT and the first positive IgG antibody result was 38.5 days and while that to the peak response was 55.5 days (range, 17–129). Antibody titers in seven patients (immediate post-transplant data were not available for patient 6 in this group) who received HSCT had dropped to less than 1:10,000 by a median time of 101 days (range, 26–137) despite these patients having a normal level of innate IDUA. A summary of the immune responses in all these patients is shown in Table 3. The specificity of the ELISA was demonstrated using western blotting, as shown in Figure 1B. This Figure shows that in the absence of antibodies in the pre-ERT sample, there is no antibody binding to Aldurazyme (recombinant IDUA) or Myozyme (recombinant human alglucosidase-α, Genzyme, Framingham, USA). After exposure to ERT, the IgG antibodies in patients’ sera only bound to Aldurazyme and not to Myozyme. ELISA results from all patients over the entire follow-up period are shown in Figure 1C base-lined to the first HSCT. None of the patients had detectable antibodies on the most recent ELISA performed a year or more after the HSCT. It is interesting to note that patient 5 in this series had primary graft rejection with autologous hematologic recovery after infusion of stored autologous cells. This resulted in an escalation of the immune response following subsequent exposure to IDUA which was later eradicated by a second unrelated donor transplant. In this cohort, the median CD3 and CD19 counts were 1140×10/L (range, 330×10/L − 3994×10/L) and 452×10/L (range, 68×10/L − 1470×10/L), respectively, at the time of the last assessment.

Analysis of the sera from patients in the cross-sectional group (n=20) showed that 85% (n=17) of patients tested 1 year or more after HSCT had no detectable antibodies despite normal IDUA levels. Eight of these patients had received ERT prior to HSCT. Three patients in this group had weakly positive IgG antibody responses to IDUA (<1:250). Chimerism data were not available for two patients in this group. The majority of patients (n=14, 78%) had achieved full donor engraftment while 22% (n=4) were chimeric with donor leukocyte engraftment of 45%, 70%, 90% and 90%. The median duration of immune suppression for graft-versus-host disease prophylaxis after HSCT in the cross-sectional and longitudinal series of patients was 150 days (range, 90–330) and 169 days (range, 110–250), respectively (see Tables 1 and 2). The median time to assessment in this cohort was 66 months (range, 12–181).

Enzyme neutralization by antibodies

Figure 2A illustrates the catalytic inhibition in all patients in the longitudinal series. In this series 62.5% of patients demonstrated catalytic inhibition. The inhibition was more pronounced at lower enzyme concentrations. Three patients (patients 4, 5 and 8) did not show enzyme inhibition at any concentration whereas another two patients (patients 7 and 3) showed over 75% inhibition of enzyme activity at concentrations of less than 1 ng/mL. Enzyme inhibition at higher concentrations (7.5–30 ng/mL) was up to 20% in comparison to that of a standard (normal serum). A similar pattern of inhibition was seen by antibodies when human and mouse enzymes were used (data not shown).

Figure 1.(A) Longitudinal data demonstrating the immune response in eight patients (1–8) in the longitudinal series from start of ERT to just before HSCT are shown baselined to the time of first starting ERT. All patients raised antibody responses to the enzyme. (B) A western blot for two patients depicting the specificity of the ELISA for IDUA (Aldurazyme). In the absence of antibodies in the pre-ERT sample, there was no antibody binding to IDUA or Myozyme (MZM). After exposure to ERT, the IgG antibodies in the patients’ serum only bound to IDUA and not to MZM. Mkr (Marker) region shows the position of various bands based on their molecular weight (kDa) according to the ladder. (C) Longitudinal data demonstrating the immune response in eight patients (1–8) in the longitudinal series are shown baselined to the time of first HSCT. Antibody titer is presented on the primary vertical axis in a logarithmic scale. Note a rapid decline in antibody titer after HSCT (following approximately 3 months of ERT). Patient 5 rejected his first graft which was followed by an escalation of antibody titers. This immune response resolved after a second transplant on day 328.

To confirm that the inhibition in a patient’s serum was due to antibodies, we assessed the enzyme inhibition in the same patient before and after HSCT i.e. the specimen with the highest titer of antibodies and no detectable antibodies on ELISA. The data confirmed that there was no catalytic inhibition in the absence of antibodies (Figure 2B).

Inhibition of cellular uptake of catalytically active enzyme

Cellular uptake inhibition was also assessed in the samples with the highest antibody titer for each patient in the longitudinal series. Inhibition of enzyme uptake by MPSI fibroblasts was seen in 75% (n=6) of patients in this series. Figure 3A illustrates the results of the uptake inhibition assay in eight patients in the longitudinal series. Two patients (patients 6 and 8) did not show any inhibition. It is interesting to note that one of these patients (patient 6) had evidence of some enzyme inhibition on catalytic inhibition assays. In contrast the two patients (patients 4 and 5) who did not have any evidence of catalytic inhibition showed 25% and 76% inhibition of enzyme uptake, respectively. Patients 2 and 7 who showed over 75% inhibition of enzyme activity also demonstrated high levels of cellular uptake inhibition (over 70%). There was no correlation between the catalytic inhibition and cellular uptake inhibition in the rest of the patients suggesting that the antibodies are polyclonal. The cellular uptake inhibition assay was repeated in non-tolerized specimens (patient on ERT pre-transplantation and with a high-titer immune response) and tolerized specimens (post-HSCT with no antibodies) from the same patient to confirm that the inhibition was due to the antibodies in the serum (Figure 3B).

Immune response and disease biomarkers

Longitudinal biomarker data (DS/CS ratio) were available for six patients in this series. Figure 4 shows the relationship between antibody titers and DS/CS ratio over the follow-up period. Patient 2 showed an immediate improvement in DS/CS ratio after starting ERT. This improvement in biomarkers was halted by a high-titer immune response. This pattern continued even after the HSCT was performed and a further improvement in DS/CS ratio resumed only after the immune response was completely abrogated (Figure 4A). The DS/CS ratio in patient 3 stayed high despite HSCT, with a continuing high-titer immune response. The slight improvement in DS/CS ratio seen in this patient was suboptimal and could be considered as a therapeutic failure at this stage. The DS/CS ratio after resolution of the immune response is not yet available for this patient (Figure 4B). The improvement in biomarker levels in patient 4 (Figure 4C) was significant after ERT was commenced; however, this progress was halted and partially reversed after a high-titer immune response was generated. The subsequent improvement in DS/CS ratio followed resolution of the immune response. Patient 5 rejected his first transplant and was given a second transplant which was succcessful. His ERT was interrupted during this period. The initial improvement in biomarker levels plateaued during the high-titer immune response despite restarting ERT. After the second HSCT, the biomarker levels showed some deterioration despite ERT and HSCT and improved only after the immune response was completely resolved (Figure 4D). Patient 6 showed a suboptimal biomarker response to ERT during the high-titer immune response (Figure 4E). The DS/CS ratio in patient 7 dropped significantly after ERT but further follow-up data were not available during the immune response and following the HSCT (Figure 4F).

Table 3.Summary of immune responses in the longitudinal series of patients.

Figure 2.(A) Catalytic enzyme inhibition in eight patients (1–8) compared to a standard (normal serum) across various enzyme concentrations. (B) Inhibition due to tolerized and non-tolerized serum. Non-tolerized serum taken from a patient (patient 3) on ERT before HSCT with high titer antibodies is compared to tolerized serum taken from the same patient a year after HSCT (with no antibodies) which shows no significant inhibition.

Discussion

The immune response to replacement recombinant human proteins has been widely reported.4 However, the clinical impact of antibodies ranges from no significant effect23 to almost complete failure of therapy.24 Hemophilia A and B are examples of congenital disorders in which the immune response (inhibitory antibodies) can make the replacement therapy (recombinant human factor VIII and IX) virtually ineffective. It is noteworthy that the true impact of neutralizing antibodies can only be measured accurately using robust functional assays, such as the Bethesda assays for hemophilia inhibitors. It is considerably more challenging to study the impact of antibodies in LSD because of the absence of standardized immune assays and the marked phenotypic heterogeneity in the natural course of a rare illness. While we studied only two mechanisms of enzyme neutralization, it is hypothesized that antibodies against ERT can inhibit the treatment by several mechanisms.25 The polyclonal nature of the IgG antibodies makes it imperative to develop multiple functional assays to encompass the full range of effects due to antibodies.26 Despite the heavy financial burden of ERT, there have been no collaborative efforts to develop and standardize quantitative and functional assays. As a consequence the true incidence and impact of these allo-antibodies remain unknown.

Figure 3.(A) Cellular uptake inhibition in all patients (1–8) in the longitudinal series is shown. Patients 6 and 8 showed no cellular uptake inhibition. (B) Tolerized serum (post-transplant with no antibodies) showed no inhibition compared to significant inhibition by non-tolerized serum (prior to transplant) in the same patient.

Figure 4.Correlation between biomarkers (DS/CS ratio) and the antibody titers in six patients in the longitudinal series (patients 2–7). The shaded areas represent the period of ERT and black arrows point to the time of HSCT. (A) Patient 2 showed an immediate improvement in the DS/CS ratio after starting ERT which was halted by high titer anti-IDUA antibodies. The biomarker levels improved after the immune response was abrogated. (B) Patient 3 showed minimal improvement in biomarkers despite HSCT and ERT during the high-titer immune response. (C) The improvement in biomarker levels in patient 4 was significant after ERT was commenced; however, this progress was arrested and reversed after the high titer immune response was generated. The biomarkers improved after resolution of the immune response. (D) Patient 5 rejected his first HSCT and had a second successful transplant. The initial improvement in biomarker levels plateaud during the high-titer immune response despite ERT (note a brief period when ERT was interrupted after the first failed HSCT) which improved only after the immune response was completely resolved after the second HSCT. (E) Patient 6 showed a suboptimal response to ERT during the high titer immune response. (F) The DS/CS ratio in patient 7 dropped to a significant level but further follow-up data were not available for the period during the immune response and following the HSCT. Biomarker data are not available for patients 1 and 8.

In recent years, the functionally active nature of allo-antibodies and the catastrophic effects of the alloimmune response in some LSD11–12 were recognized and led to efforts to develop an effective and quick immune tolerance induction regimen. These strategies had variable outcomes and serious side effects were reported in some cases.13,27–29 There is a risk that intensifying the tolerance induction protocols by combination chemotherapy and immune suppression would result in significantly increased morbidity, as has been reported in hemophilia.30,31 Development of immune tolerance with allogeneic HSCT has been reported in various autoimmune disorders and to date over a thousand HSCT have been carried out for this indication.32 The use of low intensity conditioning regimens, the ability to manipulate the graft and the very low rate of transplant-related mortality have led to the use of allogeneic HSCT as a long-term tolerance induction strategy in some solid organ transplant recipients.33,34 The role of HSCT as an immune tolerance induction mechanism in patients with neutralizing allo-antibodies has not yet been established.

All the patients in our study had the severe phenotype of MPSI H (Hurler’s syndrome). Seven patients were homozygous for null mutations, conferring a high risk of allo-immune responses. Consistent with previous reports, we found a high incidence of antibodies in this subgroup. All transplant recipients in the longitudinal series cleared their antibodies to clinically insignificant levels within a median period of 101 days despite normal enzyme levels. None of the 20 patients in the cross-sectional group had a high-titer immune response 1 year or more after HSCT and cessation of immunosuppression. Full donor engraftment was not required for immune tolerance as some patients were tolerized despite achieving only mixed chimerism. These data confirm that allogeneic HSCT is an effective and quick mechanism of inducing immune tolerance. These data also have important implications for CD34 gene therapy. In animal models of LSD and hemophilia, stem cell transplantation after hematopoietic stem cell-based gene therapy tolerized the immune system to the infused protein.35,36

The majority of patients (62.5%) demonstrated evidence of some catalytic inhibition by antibodies. Even though the catalytic inhibition appears weak at high enzyme concentrations, this in vitro inhibition may reflect partial neutralization of infused enzyme due to the very short half-life of Aldurazyme (3 h in the absence of antibodies) and a mean maximum plasma concentration (C_max) of 1.2–1.7 μg/mL (ALID-014–02: A phase II Open-Label Clinical Trial of Aldurazyme). The inhibition of endogenous (human and mouse) enzyme by antibodies suggests cross-reactivity of anti-IDUA antibodies to endogenous IDUA. This signifies a need for HSCT-induced immune tolerance, as the presence of antibodies at high titers after HSCT would potentially neutralize the cellular therapy (enzyme delivered by allogeneic HSCT).

The cellular uptake inhibition can be demonstrated at a much higher enzyme concentration. We used enzyme concentrations of less than half the Km (100 ng/ml) to optimize the uptake inhibition and serum at 1:100 dilution. Corrected for dilution, this is equivalent to just over six times the C_max (maximum plasma concentration after infusion) of Aldurazyme, making the treatment virtually ineffective in some patients with high-titer immune responses. The cellular uptake inhibition at a higher enzyme concentration compared to catalytic inhibition suggests stronger inhibition of the mannose-6-phosphate binding sites (and enzyme uptake) than the catalytic site of recombinant IDUA by anti-IDUA antibodies. Our study looked at the antibody neutralization of enzyme in a specific group of MPSI patients with a greater propensity to develop a high-titer immune response. It is possible that the variable effects and polyclonal nature of anti-IDUA antibodies might have resulted in underestimation of the effects of antibodies in previous studies because of the pooling of data from various groups of patients. It is, therefore, important to assess these patients individually.

Analysis of biomarker data and antibody titers show that the recovery is slowed or, in some cases, reversed in the presence of a high-titer immune response. Generally there appears to be a close relationship between the two, although the DS/CS ratio and biomarker responses usually lagged behind antibody responses by several days, confirming the in vivo neutralization of ERT by antibodies.

In conclusion, our data show that the high-titer immune responses in MPSI H patients treated with ERT can neutralize replacement therapy in a significant proportion of patients. In the past, the heterogeneous clinical course of the disease compounded by a lack of robust biomarkers and reliable functional immune assays made it very difficult to evaluate the effect of antibodies in LSD patients treated with ERT. There is now a dire need to standardize quantitative and qualitative immune assays in these patients. Given the remarkable improvement in the outcome of HSCT, this therapy is now a viable therapeutic modality as a mechanism for inducing immune tolerance in patients with refractory immune responses to ERT and other replacement therapies by substituting the enzyme-naïve immune system with that of the donor. The generation of high-titer neutralizing antibodies to ERT prior to HSCT makes it unnecessary to continue ERT infusions in the presence of an immune response, particularly if the transplant is carried out early after the diagnosis and in the absence of any serious co-morbidities. However, inevitably, some clinically unstable patients will benefit from ERT to optimize their clinical status prior to HSCT. Close biochemical and clinical monitoring of the immune response in ERT-treated LSD patients can help to determine the optimum therapy for this group of patients.

Footnotes

• ↵* These authors contributed equally
• Authorship and Disclosures The information provided by the authors about contributions from persons listed as authors and in acknowledgments is available with the full text of this paper at www.haematologica.org.
• Financial and other disclosures provided by the authors using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are also available at www.haematologica.org.

• Received November 17, 2011.
• Revision received January 26, 2012.
• Accepted February 13, 2012.