Successful Within-patient Dose Escalation of Olipudase Alfa in Acid Sphingomyelinase Deficiency

Abstract

Background

Olipudase alfa, a recombinant human acid sphingomyelinase (rhASM), is an investigational enzyme replacement therapy (ERT) for patients with ASM deficiency [ASMD; Niemann-Pick Disease (NPD) A and B]. This open-label phase 1b study assessed the safety and tolerability of olipudase alfa using within-patient dose escalation to gradually debulk accumulated sphingomyelin and mitigate the rapid production of metabolites, which can be toxic. Secondary objectives were pharmacokinetics, pharmacodynamics, and exploratory efficacy.

Methods

Five adults with nonneuronopathic ASMD (NPD B) received escalating doses (0.1 to 3.0 mg/kg) of olipudase alfa intravenously every 2 weeks for 26 weeks.

Results

All patients successfully reached 3.0 mg/kg without serious or severe adverse events. One patient repeated a dose (2.0 mg/kg) and another had a temporary dose reduction (1.0 to 0.6 mg/kg). Most adverse events (97%) were mild and all resolved without sequelae. The most common adverse events were headache, arthralgia, nausea and abdominal pain. Two patients experienced single acute phase reactions. No patient developed hypersensitivity or anti-olipudase alfa antibodies. The mean circulating half-life of olipudase alfa ranged from 20.9 to 23.4 hours across doses without accumulation. Ceramide, a sphingomyelin catabolite, rose transiently in plasma after each dose, but decreased over time. Reductions in sphingomyelin storage, spleen and liver volumes, and serum chitotriosidase activity, as well as improvements in infiltrative lung disease, lipid profiles, platelet counts, and quality of life assessments, were observed.

Conclusions

This study provides proof-of-concept for the safety and efficacy of within-patient dose escalation of olipudase alfa in patients with nonneuronopathic ASMD.

INTRODUCTION

Acid sphingomyelinase deficiency [ASMD; Niemann-Pick disease (NPD) type A and B] is a rare lysosomal storage disorder that affects 0.25-.40 per 100,000 births depending on NPD subtype (1, 2). Mutations in the SMPD1 gene encoding the lysosomal enzyme acid sphingomyelinase (ASM) result in the accumulation of sphingomyelin primarily within reticuloendothelial cells and hepatocytes of affected patients (3). Sphingomyelin accumulation produces severe systemic manifestations, including hepatosplenomegaly, liver dysfunction, infiltrative lung disease, thrombocytopenia, anemia, and bone disease. In the most severe form with acute neuronopathic disease (NPD A), patients have little to no residual ASM activity, with onset of systemic manifestations, failure-to-thrive, and rapidly progressive neurodegeneration during infancy, and death by 3 years of age (4). In contrast, patients with chronic nonneuronopathic ASMD (NPD B) have higher residual ASM activity and variable ages of onset from infancy to adulthood. Systemic manifestations are heterogeneous and include hepatosplenomegaly, pancytopenia, infiltrative lung disease, and a pro-atherogenic lipid profile. Growth restriction during childhood and delayed puberty are common manifestations (5–7). Patients with NPD B may have a normal lifespan; however, some die prematurely from pulmonary or liver disease, or from hemorrhage, including splenic rupture (5).

There are no approved, etiology-based therapies for ASMD. Intravenous administration of enzyme replacement therapy (ERT) with olipudase alfa (recombinant human acid sphingomyelinase), has shown promising results in the ASM knock-out (ASMKO) mouse with dose-dependent reductions in tissue sphingomyelin levels up to 3.0 mg/kg (8, 9). Doses up to 3.0 mg/kg were able to reduce sphingomyelin levels in hard to reach target tissues, including lung. High doses (≥10.0 mg/kg), however, caused severe toxicity characterized by systemic inflammation and cardiovascular shock. Concomitant elevations in plasma ceramide and cytokine levels in ASMKO animals, along with the absence of toxicity in normal mice administered up to 30 mg/kg olipudase alfa, suggested that the observed toxicity in the ASMKO mice was a result of rapid production of sphingomyelin catabolite(s) rather than a nonspecific drug reaction. Consistent with this hypothesis, the high-dose toxicity could be completely prevented by prior treatment of ASMKO mice with several low doses of olipudase alfa to gradually debulk accumulated sphingomyelin (9).

Olipudase alfa is in clinical development for treatment of the systemic (nonneurological) manifestations of ASMD, as the enzyme is unable to cross the blood-brain barrier. A phase 1 single-ascending-dose study of olipudase alfa in adult patients with NPD B identified 0.6 mg/kg as the maximum tolerated first dose (10). The dose-limiting toxicity manifested as hyperbilirubinemia in a patient later diagnosed with Gilbert syndrome who received a dose of 1.0 mg/kg. At doses ≥ 0.3 mg/kg, constitutional symptoms (e.g. pyrexia, myalgia, and nausea) and changes in inflammatory mediators (e.g., C-reactive protein and cytokines) began 12 to 24 hours postinfusion and were consistent with an acute phase response triggered by the innate immune system (10). This is in contrast to the typical hypersensitivity-type, infusion-associated reactions (IARs), e.g., pyrexia, rash, urticaria, chills, and chest tightness, observed with other ERTs that develop in some patients after several infusions, and which are often associated with the formation of antibodies against the infused protein.

The unique adverse drug reactions to olipudase alfa, which were dose-dependent and attributable to the rapid production of sphingomyelin catabolites, coupled with the variable amount of sphingomyelin storage across patients, led to a novel phase 1b study design that employed within-patient, dose-escalation to gradually debulk accumulated sphingomyelin, thereby mitigating toxicity and enabling efficacious enzyme concentrations to be achieved in target tissues, including the lung.

METHODS

Study Design and Participants

This phase 1, open-label, within-patient, repeat-dose, dose-escalation study was conducted at two sites in the US and UK between March 2013 and January 2014. Adults between 18 and 65 years of age with confirmed nonneuronopathic ASMD, diffusing capacity for carbon monoxide (DL_CO) >20% and ≤80% of the predicted normal value, spleen volume ≥6 multiples of normal (MN), platelet count ≥60 × 10³/μL, and international normalized ratio (INR) >1.5 were eligible to participate. Patients with a history of major organ transplant, or who required medications that could decrease olipudase alfa activity (e.g., chlorpromazine, imipramine, or desipramine), were excluded. Patients on a stable dose and regimen of lipid-lowering therapy were eligible.

Procedures

Patients received intravenous infusions of olipudase alfa once every 2 weeks for 26 weeks at an initial dose of 0.1 mg/kg (day 1) followed 2 weeks later by 0.3 mg/kg. After tolerating two consecutive 0.3 mg/kg doses, dose escalation proceeded to 0.6, 1.0, 2.0, and 3.0 mg/kg, which was maintained for the remainder of the 26-week study. All patients were monitored in the hospital pre-infusion and for 72 hours after each infusion until the second 3.0 mg/kg infusion, after which they were monitored for at least 3 hours post-infusion. Patients who experienced adverse events greater than mild in severity were to repeat the same dose or receive a reduced dose at the next infusion. Dose escalation could only proceed if adverse events were mild or absent.

Outcome Measures

Safety was the primary objective and assessments included physical examinations and vital signs, cardiac evaluations, clinical laboratory tests, safety biomarkers [ceramide, high-sensitivity C-reactive protein (hsCRP), interleukin (IL)-6, IL-8], liver function [aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma-glutamyl transferase (GGT)], immune responses, continuous monitoring of treatment emergent adverse events (TEAEs) and infusion associated reactions (IARs), and histopathological evaluations of liver biopsy samples.

Secondary objectives included pharmacodynamic and pharmacokinetic assessments of olipudase alfa. Pharmacodynamic parameters included measurement of plasma sphingomyelin, and ceramide levels by liquid chromatography-tandem mass spectrometry (LC/MS/MS) and liver sphingomyelin content by both LC/MS/MS and computer-assisted morphometric analysis of hepatic histopathology (MetaMorph Imaging Processing and Analysis software (Version 6.3; Universal Imaging Corporation) (11). Serial blood samples for pharmacokinetic evaluations were collected at doses ≥0.3 mg/kg before and immediately after infusion plus 1, 4, 8, 12, 24, 48, and 72 hours after infusion. Plasma concentrations of olipudase alfa were determined using a validated enzyme linked immunosorbent assay (ELISA) at Genzyme.

Exploratory efficacy outcomes included quantitative measurement of spleen and liver volumes using abdominal MRI; qualitative assessment of infiltrative lung disease using high-resolution computed tomography (HRCT) (12) and chest x-ray; and pulmonary function tests (PFT) using American Thoracic Society guidelines (13). Fasting lipid profiles included measurement of total cholesterol (TC), low density lipoprotein (LDL-C) very low density lipoprotein (VLDL-C) high density lipoprotein (HDL-C), apo-lipoprotein B100 (Apo B) and triglycerides. Exploratory disease biomarkers included those related to macrophage proliferation [chitotriosidase, CCL18, and angiotensin converting enzyme (ACE)] and bone-specific biomarkers including alkaline phosphatase and C-telopeptide.

The validated Brief Fatigue Inventory (BFI) and Brief Pain Inventory-Short Form (BPI-SF) questionnaires (14, 15) were used by patients to rate the severity of their fatigue and pain and assess interference with daily activities at baseline and week 26. Patients rated fatigue from 0 (no fatigue) to 10 (worst fatigue) and scores were categorized as mild (1–3), moderate (4–6), and severe (7–10). Patients rated average, current and last 24-hour pain intensity from 0 (no pain) to 10 (worst pain), which was categorized as mild (1–4), moderate (5–6) and severe (7–10) (16). BPI pain interference was graded from 0 (no interference) to 10 (completely interferes) as the mean of 7 interference items reported during the last 24 hours: general activity, mood, walking ability, normal work (both work outside the home and housework), relations with other people, sleep, and enjoyment of life.

Analyses

All patients were included in efficacy, safety, pharmacokinetic, and pharmacodynamic analyses. Histopathological evaluations of inflammation and fibrosis in liver samples were performed using the Laennec scoring system (17) to grade the extent of fibrosis on a scale from 0–4 (0, no fibrosis; 1, minimal; 2, mild; 3, moderate; 4, cirrhosis). Percent predicted values were calculated for PFT parameters (18, 19). Raw data and change (absolute and percent) from baseline for exploratory efficacy endpoints were summarized by time point using descriptive statistics. Liver and spleen volumes were reported as absolute volumes (cm³) and were calculated as MN where normal spleen volume was assumed to be 0.2% of body weight and normal liver volume to be 2.5% body weight.

Study Approval

The Institutional Review Board (US) or Ethics Committee (UK) at each site approved the protocol, and all patients provided written informed consent prior to inclusion in the study. The study was conducted according to Good Clinical Practice and in accordance with the principles of the Declaration of Helsinki.

RESULTS

Patient Demography and Baseline Characteristics

Five of six screened patients were enrolled in the study. One patient was ineligible due to prohibited medication use and excluded baseline hematology and bilirubin levels. Patients ranged in age from 23 to 48 years, included 3 males and 2 females, and all were Caucasian. Residual ASM enzyme activity in leukocytes ranged from 12 to 29% of normal in patients for whom historical data were available (n = 4), and all had low ASM activity in dried blood spots (DBS) consistent with ASMD (20). Patient genotypes are shown in Table 1.

Table 1.

Patient Demographics and Baseline Characteristics

Patient ID
	1	2	3	4	5
	Male	Female	Female	Male	Male	Mean (SD)
Symptom onset age (years)	2	1	6	0	12	4.2 (4.9)
Diagnosis age (years)	2	2	12	8	12	7.2 (5.0)
Age at first olipudase alfa infusion (years)	31	32	47	28	22	32.6 (9.4)
ASM activity in leukocytes (% of normal)a	NA	17%	29%	13%	12%	17.8 (7.8)
ASM activity in DBS (μmol/L/hr) b	1.00	1.67	1.15	1.12	0.25	1.04 (0.46)
Genotype	G232D/E515V	fsP330/ΔR608	G242R/N383S	R600H/ΔR608	fsV143/R441X
Spleen Volume (MN)c	14.49	17.92	7.41	16.07	7.96	12.77 (4.81)
Liver Volume (MN)c	2.23	2.20	1.21	1.76	1.29	1.74 (0.48)
Hemoglobin (g/L)d	136	139	132	134	152	139 (8)
Platelet Count (109/L)e	73	119.5	127	169	101	117.6 (35.3)
DLCO (% predicted)f	43.7	48.0	77.0	43.0	80.0	58.3(18.5)
TC(mmol/L)g	4.70	3.83	5.26	4.66	3.63	4.42 (0.67)
HDL-C(mmol/L)h	0.32	0.36	0.96	0.31	0.57	0.50 (0.28)
TC/HDL-C Ratio	14.7	10.6	5.5	15.0	6.4	10.4 (4.5)
LDL-C (mmol/L)i	3.38	2.59	3.32	2.69	2.25	2.85 (0.49)
VLDL-C (mmol/L)j	0.88	0.88	0.98	1.66	0.80	1.04 (0.35)
Triglycerides (mmol/L)k	2.20	1.55	1.14	4.35	1.38	2.12 (1.31)
Apo B (g/L)l	1.41	0.96	0.92	1.32	0.59	1.04 (0.33)

Apo B = apo-lipoprotein B100; ASM= acid sphingomyelinase; C = cholesterol; DBS = dried blood spot; DL_CO=lung diffusion of carbon monoxide; HDL=high-density lipoprotein; LDL=low-density lipoprotein; MN=multiples of normal; NA = not available; SD=standard deviation; TC =total cholesterol; VLDL=very low-density lipoprotein.

^aHistorical data

^bASM activity in normal adults: 4.96 μmol/L/hr (21)

^cMN, multiples of normal calculated assuming normal spleen volume is 0.2% body weight (kg), and normal liver volume is 2.5% body weight (kg)

^dHemoglobin normal range: UK Male 130 - 180 g/L, US female 117 - 150 g/L, US Male 139 - 163 g/L

^ePlatelet count normal range: UK 150 – 400×10⁹/L, US 150 – 450×10⁹/L

^fNormal DLco >80%; Mildly reduced >60% to ≤ 80%; Moderately reduced 40–60%; Severely reduced < 40%

^gTotal cholesterol normal range: US <5.18 mmol/L; UK 0–3.9 mmol/L

^hHDL normal range: US male >0.777; US female >0.9065 mmol/L; UK >1.2 mmol/L

ⁱLDL normal range: US <3.3411 mmol/L; UK 0–2 mmol/L

^jVLDL normal range: US <0.518 mmol/L; UK 0.09–0.71 mmol/L

^kTriglycerides normal range: <1.7 mmol/L

^lApo B normal range: US male 0.55–1.4 g/L; US female 0.55–1.25 g/L; UK 0.52–1.09 g/L

All patients had splenomegaly (range 7.4 to 16.1 MN), hepatomegaly (range 1.2 to 2.2 MN), impaired gas exchange (range 43 to 80% of predicted DLco), and a pro-atherogenic lipid profile (Table 1). All patients had TC to HDL-C ratios above 5 (range 5.5 to 15.0) despite two being on a stable regimen of lipid-modifying agents. Four patients had thrombocytopenia and none were anemic. All patients had a history of hepatosplenomegaly, excessive bleeding/bruising and joint/abdominal/back pain; four had a history of delayed puberty and osteopenia/osteoporosis, and three had a history of short stature, thrombocytopenia, shortness of breath, and/or fatigue.

Dose Escalation and Drug Exposure

All patients successfully escalated from the initial dose of 0.1 mg/kg to the target dose of 3.0 mg/kg and completed 26 weeks of olipudase alfa treatment. There were 70 total infusions with 5, 10, 6, 6, 6, and 37 infusions at the 0.1, 0.3, 0.6, 1.0, 2.0 and 3.0 mg/kg dose levels, respectively (Table 2, part C).

Table 2.

Summary of TEAE Profile and Common Events by Olipudase Alfa Dose

TEAE Profile	Patients n (%)	Events n
Any TEAE	5 (100)	216
Mild	5 (100)	210
Moderate	2 (40)	6
Severe	0	0
Related TEAE	4 (80)	107
Related TEAEs considered IARs	4 (80)	55
Mild	4 (80)	49
Moderate	2 (40)	6
Serious adverse events, deaths, discontinuations	0	0

Commona TEAEs	Patients n (%)	Events n	Common IARsb	Patients n (%)	Events n
Headache	4 (80)	18	Headache	3 (60)	14
Arthralgia	4 (80)	16	Nausea	2 (40)	7
Nausea	3 (60)	14	Abdominal pain	2 (40)	6
Abdominal pain	4 (80)	14	Musculoskeletal pain	1 (20)	6
Back pain	5 (100)	8	Arthralgia	2 (40)	3
Pain in extremity	3 (60)	6
Abdominal discomfort	3 (60)	5
Pyrexia	3 (60)	5

Commona TEAEs by Olipudase Alfa Dose (mg/kg) and Exposure (# infusions)
Dose (Exposure)	TEAE	Patients n (%)	Events n
0.1 (5 infusions)
	Headache	1 (20)	1
	Abdominal pain	1 (20)	1
	Pain in extremity	1 (20)	1
0.3 (10 infusions)
	Headache	2 (40)	2
	Abdominal pain	2 (40)	2
	Nausea	1 (20)	2
	Back pain	1 (20)	1
0.6 (6 infusions)
	Arthralgia	2 (40)	3
	Nausea	2 (40)	3
	Abdominal discomfort	2 (40)	3
	Headache	2 (40)	2
	Abdominal pain	1 (20)	1
	Fatigue	1 (20)	1
	Pyrexia	1 (20)	1
1.0 (6 infusions)
	Arthralgia	2 (40)	3
	Headache	1 (20)	1
	Nausea	1 (20)	1
	Abdominal pain	1 (20)	1
	Abdominal discomfort	1 (20)	1
	Pyrexia	1 (20)	1
2.0 (6 infusions)
	Pain in extremity	3 (60)	3
	Headache	2 (40)	2
	Back pain	2 (40)	3
	Arthralgia	1 (20)	1
3.0 (37 infusions)
	Headache	4 (80)	10
	Arthralgia	4 (80)	12
	Nausea	3 (60)	5
	Abdominal pain	3 (60)	4
	Back pain	3 (60)	4
	Pyrexia	2 (40)	3
	Pain in extremity	1 (20)	2
	Abdominal discomfort	1 (20)	1

IAR=infusion associated reaction; TEAE=treatment emergent adverse events.

^aCommon TEAEs defined as 5 or more events in greater than 2 patients for all doses combined.

^bCommon IARs defined as 3 or more events for all doses combined

Three patients completed dose escalation by week 12 (seven doses) as planned and remained at the 3.0 mg/kg dose. Two patients had IARs requiring a longer period of dose escalation. Patient 1 had an IAR (flu-like symptoms) with moderate pyrexia following the week 8 infusion of 1.0 mg/kg and was administered a reduced dose of 0.6 mg/kg at week 10. Patient 2 had an IAR (moderate adverse events of worsening migraine headache, abdominal gas pain, splenic pain, nausea, and headache) following the week 10 infusion of 2.0 mg/kg and repeated the same dose at week 12. Both patients resumed dose escalation without incident and remained at the 3.0 mg/kg dose through week 26.

Pharmacokinetics

Olipudase alfa plasma levels increased in a close to dose-proportional manner from a mean (SD) of 3270 (458) ng/mL after the first 0.3 mg/kg infusion to 23,100 (2230) ng/mL after the final 3.0 mg/kg infusion at week 26. The C_max of olipudase alfa was at the end of the infusion for each patient. Mean t_1/2z values ranged from 20.9 to 23.8 hours across all doses. There was no apparent relationship between dose and terminal t_1/2z, clearance (Cl), or volume of distribution at steady state (V_ss). All plasma concentrations measured before each infusion were below the lower limit of quantitation (LLOQ), indicating no accumulation of olipudase alfa in the circulation with biweekly dosing.

Adverse Events

The dose escalation regimen was well-tolerated with no serious or severe TEAEs, and no TEAEs resulted in study discontinuation. Most adverse events (210/216, 97%) were mild. The overall TEAE profile is shown in Table 2. There were 6 adverse events of moderate severity associated with IARs in two patients: Patient 1, a 31-year-old male, had pyrexia after the 1.0 mg/kg infusion, and Patient 2, a 32-year-old female, had 5 adverse events (abdominal gas pain, spleen pain, nausea, headache, and migraine) following the 2.0 mg/kg infusion. The most common TEAEs and IARs (including headache, arthralgia, nausea and abdominal pain) and the common TEAEs reflecting cumulative exposure by dose are shown in Table 2. The TEAE profile was similar at each dose.

Approximately half (107/216, 49.5%) of all TEAEs were considered related to olipudase alfa treatment, with 55 identified as IARs in 4 patients during 32 of the 70 infusions. The IAR profile and common IARs are also summarized in Table 2 parts A and B. The majority of IARs (39/55, 67%) occurred >3 hours post-infusion: 10 during infusion, 6 between 0–3 hours post-infusion, 17 between 3–24 hours post-infusion, 20 between 24–72 hours post-infusion, and 2 >72 hours post-infusion. There were 0, 2, 12, 10, 9, and 22 IARs at the 0.1, 0.3, 0.6, 1.0, 2.0, and 3.0 mg/kg dose levels, respectively. The number of IARs per infusion was lowest at the 0.1 and 0.3 mg/kg dose levels (0 and 0.2 IARs/infusion, respectively), and highest at the intermediate dose levels of 0.6, 1.0 and 2.0 mg/kg doses (2, 1.7 and 1.5 IARs/infusion, respectively). The number of IARs/infusion at the 3.0 kg/kg dose level was 0.6, even though the greatest exposure was at this dose with 37 infusions.

Patient 1 had laboratory findings and clinical symptoms consistent with an acute phase reaction, including elevated ferritin (from 37.3 μg/L preinfusion to 153.4 μg/L postinfusion day 2), and hsCRP (from 0.4 mg/L preinfusion to 31.9 mg/L postinfusion day 2) following the 1.0 mg/kg infusion at week 8. Following dose reduction to 0.6 mg/kg at week 10 without incident, dose escalation at 1.0 mg/kg resumed at week 12. Findings consistent with an acute phase reaction also occurred in Patient 5, a 23-year-old male, following the 2.0 mg/kg dose. Elevations in hsCRP postinfusion day 2 (from 0.8 mg/L preinfusion to 16.5 mg/L) at week 10 were associated with changes in at least one of the negative and/or positive acute phase reactants, including decreased iron (from 14 μmol/L preinfusion to 5.4 μmol/L postinfusion day 2) and increased ferritin levels (from 14 μg/L preinfusion to 27 μg/L postinfusion day 2). No dose reduction or repetition was required.

No adverse events, including IARs associated with acute phase reactions, were consistent with hypersensitivity reactions or cytokine release syndrome, and no patient developed antibodies (IgG or IgE) to olipudase alfa during the 26-week study.

With the exception of the two patients with acute phase reactions, no clinically significant changes in laboratory findings or safety biomarkers were observed. Transient elevations from baseline and from pre-infusion levels were observed in IL-6, IL-8, and hsCRP during dose escalation. The peak elevations in these biomarkers occurred between weeks 6 and 10 and were associated with the acute phase responses that occurred in Patient 1 at the 1.0 mg/kg and Patient 5 at the 2.0 mg/kg dose. Figure 1 shows the mean hsCRP levels pre- and post-infusion for weeks 0 through 16. No transient increases in hsCRP occurred after week 14, by which time all patients were receiving the maximum 3.0 mg/kg dose.

Figure 1. Mean High Sensitivity C-Reactive Protein (hsCRP) Levels Pre- and Post-Infusion from Baseline to Week 16.

Arrow indicates when patients escalated to the 3.0 mg/kg dose (3 patients at week 12 and 2 patients at week 14) Post-infusion hsCRP levels remained in the normal range for the remainder of the study (data not shown).

Mean liver function enzyme levels were within normal ranges at baseline (AST, ALT, GGT) and bilirubin was slightly elevated. All trended downward from baseline over time with mean values within normal ranges (data not shown). Liver pathology assessments showed no signs of acute inflammation at baseline. Patient 1 had a single focus of lymphocytic infiltrate at week 26 consistent with chronic inflammation.

Fibrosis scores varied at baseline from no fibrosis (0) to moderate fibrosis (3). A one-point change in score was reported for Patient 3, who had no fibrosis (0) at baseline and a score of 1 (mild, periportal fibrosis) at week 26. Patient 4 had a score of 2 (mild) at baseline and had a change to 3 (moderate) at week 26, although the liver biopsy sample at baseline was less than the minimum length (1.5 cm) adequate for scoring fibrosis. Baseline and week 26 fibrosis scores for Patient 1 were 3 (moderate) and 4 (cirrhosis) at baseline and week 26 respectively, however, these liver biopsy samples were also less than 1.5 cm.

There were no clinically significant changes in vital signs, physical findings or cardiac safety parameters.

Sphingomyelin Metabolism

Plasma sphingomyelin levels remained within the normal range (<200 to 703 μg/mL) for all patients throughout the study, with levels tending to peak transiently 48 hours following each olipudase alfa infusion. The mean (SD) plasma ceramide level was slightly elevated at baseline (6.62 (1.08) μg/mL; normal range, 1.8 to 6.5 μg/mL). All doses of olipudase alfa elicited similar and transient increases in plasma ceramide levels that generally peaked at 48 hour post-infusion. Both pre- and post-infusion ceramide levels steadily decreased with each successive olipudase alfa infusion and then plateaued after week 14 (Figure 2).

Figure 2. Mean Plasma Ceramide Levels Pre- and Post-Infusion from Baseline to Week 26.

Absolute values ± SD.

Biomarkers

Changes in the exploratory biomarkers chitotriosidase, CCL18, and ACE after 26 weeks of treatment with olipudase alfa are presented in Table 3, and over time in Figure 3. Chitotriosidase and CCL18, which are secreted by activated macrophages, were moderately elevated at baseline (4-fold and 6-fold the upper limits of normal, respectively) and steadily decreased through week 26 by 56% and 38%, respectively. Mean levels for both biomarkers remained above normal at week 26. Mean ACE activity, which reflects the size of the macrophage pool, was mildly elevated at baseline (1.2-fold the upper limit of normal). Although levels fluctuated over the course of treatment, mean ACE activity was decreased by 33% at week 26 and was within normal limits. Bone-specific alkaline phosphatase, a bone formation biomarker, and C-telopeptide, a bone resorption biomarker, were within normal limits at baseline and week 26 (data not shown).

Table 3.

Effect of Olipudase Alfa on Exploratory Efficacy Endpoints

Endpoint	Baseline	Week 26	Week 26 Change from Baseline	Week 26 % Change from Baseline*
Exploratory Biomarkers Normal ranges	Mean (SD)
Chitotriosidase (nmol/hr/mL) 15–181	735.0 (494.8)	319.8 (232.5)	−415.2 (298.9)	−56.0 (22.4)
CCL18 (μg/mL) 16.5–169.5	1033.14 (623.75)	573.46 (274.04)	−459.68 (480.78)	−38.3 (23.3)
ACE (IU/L) 12–70	91.56 (50.76)	62.16 (41.85)	−29.40 (27.52)	−33.4 (19.2)
Lipid Profile
TC (mmol/L) <5.18 US. 0–3.9 UK	4.42 (0.67)	3.37 (0.75)	−1.05 (0.86)	−22.8 (18.0)
HDL-C (mmol/L) >0.777 US male, >0.9065 US female, >1.2 UK	0.50 (0.28)	0.71 (0.28)	0.21 (0.10)	49.5 (27.0)
Ratio**	10.4 (4.5)	5.3 (2.3)	−5.2 (3.6)	−46.2 (16.8)
LDL-C (mmol/L) <3.3411 US, 0–2 UK	2.85 (0.49)	2.21 (0.67)	−0.64 (0.82)	−20.7 (24.3)
VLDL-C (mmol/L) <0.518 US, 0.09–0.71 UK	1.04 (0.35)	0.43 (0.28)	−0.67 (0.20)	−62.1 (16.9)
Triglycerides (mmol/L) < 1.7	2.12 (1.31)	1.06 (0.41)	−1.06 (0.98)	−43.4 (19.5)
Apo-lipoprotein B100 (g/L) 0.55–1.4 US male; 0.55–1.25 US female; 0.52–1.09 UK	1.04 (0.33)	0.77 (0.20)	−0.27 (0.38)	−20.6 (26.9)

^*Mean percent changes from baseline over time for biomarkers and lipids are shown in Figures 3 and 6, respectively.

^**Ratio of TC to HDL-C

ACE = angiotensin converting enzyme; C = cholesterol; CCL18 = chemokine (C-C motif) ligand 18; HDL = high density lipoprotein; LDL = low density lipoprotein; VLDL = very low density lipoprotein; TC = total cholesterol

Figure 3. Exploratory Disease Biomarkers. Mean Percentage Changes in ACE, Chitotriosidase, and CCL18 Levels from Baseline to Week 26.

Absolute values ± SD at Baseline and Week 26 are shown in Table 3. ACE = angiotensin converting enzyme; CCL18 = chemokine (C-C motif) ligand 18

Liver Sphingomyelin Content

Representative high-resolution light microscopy (HRLM) images of liver biopsy tissue sections (Patient 2) stained for sphingomyelin and used for histomorphometric analysis at baseline and week 26 are shown in Figure 4A. Abundant sphingomyelin staining was present in both hepatocytes (H) and Kupffer (K) cells at baseline, and staining was markedly decreased at week 26. In the four patients with evaluable liver biopsy samples, mean (SD) liver sphingomyelin levels measured histomorphometrically decreased from 33.3% (17.8%) of total tissue area occupied by sphingomyelin at baseline, to 4.3% (3.6%) of total tissue area occupied by sphingomyelin at week 26, corresponding to a relative decrease of 86.6% (4.29%) (Figure 4B). Liver sphingomyelin content measured by LC/MS/MS yielded similar results with reductions observed in all 5 patients. The mean (SD) sphingomyelin level decreased by 66.3% (24.6), from 2414.8 (1284.5) μg/mg protein at baseline to 968.4 (1000.4) μg/mg protein at week 26 (individual and LC/MS/MS data not shown).

Figure 4. Effect of Olipudase Alfa on Liver Sphingomyelin Content and Spleen and Liver Volume after 26 Weeks.

A. High-resolution light microscopy (HRLM) images of liver biopsy tissue sections stained with modified toluidine blue (600x) from Patient 2. Arrows indicate sphingomyelin (dark purple) accumulation in lysosomes of hepatocytes (H) and Kupffer cells (K). B. Mean sphingomyelin content in liver quantified using histomorphometric analysis of HLRM at baseline and week 26 using MetaMorph Imaging Processing and Analysis software (Version 6.3; Universal Imaging Corporation). All sphingomyelin identified and measured by this method represents the disease process. The normal value for this parameter in a non-ASMD liver sample is zero. C. By-patient spleen volumes in multiples of normal (MN) at baseline and week 26 (22); D. By-patient liver volumes (MN) at baseline and week 26. (22)

Hepatosplenomegaly

Corresponding reductions in splenomegaly and hepatomegaly were also observed after 26 weeks. Mean (SD) spleen volume decreased by 25.3% [1646.5 (665.1) cm³ to 1233.0 (505.8) cm³], and mean (SD) liver volume decreased by 17.1% [2800.6 (934.4) cm³ to 2191.9 (311.3) cm³]. Slightly greater reductions were observed when organ volumes were normalized for body weight owing to weight gain, with mean (SD) spleen volume decreasing by 29.4% [12.77 (4.81) MN to 9.07 (3.53) MN], and mean (SD) liver volume decreasing by 21.9% [1.74 (0.48) MN at Baseline to 1.30 (0.13) MN] (22). Spleen and liver volume reductions occurred in 5 and 4 patients, respectively, with one patient with mild hepatomegaly at baseline having no change at week 26 (Figure 4C and 4D).

Hematology

Hematology variables are negatively correlated with splenomegaly (2), and were therefore part of the exploratory efficacy analyses. There was a minor decrease in mean hemoglobin at the end of study (138.3 g/L at baseline versus 132.8 g/L at week 26), which remained within the normal range and was likely related to frequent phlebotomy (23). Mean ferritin concentration decreased over the course of the study (76.6 μg/L at baseline and 22.8 μg/L at week 26), but remained within the normal range and was not associated with a reduction in mean iron levels (12.5 μmol/L at baseline versus 14.1 μmol/L at week 26) or low hemoglobin levels. The mean platelet count at baseline of 117.6 × 10⁹/L indicated moderate thrombocytopenia and increased by 18% to 138.8 × 10⁹/L (mild thrombocytopenia) at week 26.

Pulmonary Imaging and Function

Baseline HRCT confirmed the presence of infiltrative lung disease in all 5 patients, ranging from mild (score of 1) to severe (score of 3) (12). Observed patterns of interstitial lung disease, reticulo-nodular densities, and ground glass appearance, were evident at all four anatomic levels (Figure 5A), with the lowest scores recorded for ground glass appearance and highest for interstitial lung disease. By week 26, consistent reductions were observed for all disease patterns at all anatomic levels and in both lungs (Figure 5B). Moreover, 2 patients with a severe interstitial lung disease pattern improved to mild or moderate, while 3 patients with a severe reticulo-nodular density pattern improved to mild or moderate at week 26. Chest X-rays corroborated the HRCT results, with measurable improvements from baseline in the patterns of interstitial lung disease, lung nodules, and reticulo-nodular densities noted in 4 of 5 patients (data not shown; ground glass pattern was not assessed by X-ray).

Figure 5. Effect of Olipudase Alfa on Lung Infiltrative Lung Disease and Function.

A. High-resolution CT images (Level 2) of lung at baseline and week 26 (Patient 1). The straight arrow in each image refers to reticular changes best seen at the periphery. The block arrow refers to ground glass opacity. Both were improved from baseline to week 26, with only mild residual abnormality at week 26. B. Qualitative assessment of infiltrative lung disease showing the interstitial, reticulonodular, and ground glass appearance components. There were no findings of consolidation, or pleural thickening. Areas of the right and left lungs were evaluated at 4 anatomic levels (1= level of the aortic arch; 2= level of the carina; 3= midway between the tracheal carina and 1 cm above the hemidiaphragm; 4=1 cm above the hemidiaphragm) based on a 4-point scale: 0=no disease; 1=mild disease affecting 1–25% of lung volume (green); 2=moderate, affecting 26–50% of lung volume (yellow); 3=severe, affecting 51–100% of lung volume(red). Imaging data were evaluated by external central readers blinded to patient and time point; C. By-patient percent predicted DLco levels at baseline and week 26. Normal DLco: >80%; Mildly reduced: >60% to ≤80%; Moderately reduced: 40–60%; Severely reduced: <40%

Spirometry was normal (percent predicted ≥80%) for forced vital capacity, forced expiratory volume in 1 second, and total lung capacity throughout the study. In contrast, all 5 patients displayed reduced percent predicted DLco at baseline with a mean (SD) of 58.3% (18.5%) indicating moderate impairment of gas exchange (24). By week 26, percent predicted DLco had increased in 4 of 5 patients, resulting in a mean group (SD) DLco of 64.4% (14.3%), reflecting a 13.4% relative increase from baseline and improvement to the mildly impaired range (Figure 5C).

Plasma Lipid Levels

Patients were at mild-to-moderate risk for cardiovascular disease based on their baseline lipid profiles, with elevated mean TC, LDL-C, VLDL-C and triglyceride levels, and a low mean HDL-C level (Table 3). TC/HDL-C ratios were above 5 in each patient, with a mean TC/HDL-C ratio of 10.4. The mean Apo B level was normal at baseline. Treatment with olipudase alfa was associated with gradual improvements in all lipid parameters (Figure 6). By week 26, the mean TC (−22.8%), VLDL-C (−62.1%) and triglyceride (−43.4%) levels normalized, while LDL-C (−20.7%) and HDL-C (+49.5%) levels approached normal ranges, and the Apo B level (−20.6%) decreased (Table 3). TC and LDL-C levels normalized in all 5 patients and VLDL-C levels normalized in 3 patients by week 26. The mean ratio of TC/HDL-C at week 26 was 5.3, and the mean change from baseline at week 26 represented a 46% reduction in the TC/HDL-C ratio.

Figure 6. Mean Percentage Changes in Fasting Lipid Profiles from Baseline to Week 26.

Absolute values ± SD at baseline and week 26 are shown in Table 3. HDL = high density lipoprotein; LDL = low density lipoprotein; VLDL = very low density lipoprotein

Patient-Reported Outcomes

Global fatigue scores at baseline ranged from mild (score of 1–3) to moderate (score of 4–6). The mean (SD) baseline score was 3.0 (2.3) and decreased to 2.1 (2.1) at week 26, reflecting a reduction in global fatigue in 3 of 5 patients. Two patients who reported increase in fatigue at week 26 had no change in severity scale, remaining in the mild (0.4 to 0.8) and moderate (5.1 to 5.8) categories, respectively.

Pain severity ratings for worst pain at baseline ranged from no pain (score of 0) in 1 patient to severe pain (score of 7) in 2 patients. Median (range) “worst pain” severity score decreased by 3.0 points, from moderate pain of 6 (0, 7)) at baseline to mild pain of 3 (0, 5) at week 26. The two patients with severe pain at baseline (scores of 7) reported mild (score of 3) or moderate (score of 5) pain at week 26.

At baseline, pain interfered with life activities, work, and/or relationships in 4 patients. Concomitant with improvement in worst pain severity, median (range) scores for pain interference improved from 1.9 (0, 4.3) at baseline to 0 (0, 1.6) at week 26, and reflected improvement in all 4 patients.

Discussion

The primary objective of the study was to apply a within-patient dose escalation strategy for olipudase alfa to slowly reduce accumulated sphingomyelin in order to address dose-limiting toxicities observed in a previous single dose clinical trial and in nonclinical studies and enable the safe, tolerable, and effective repeat dosing of olipudase alfa in patients with ASMD (8–10). The study results support this approach as all patients reached and maintained the target dose of 3.0 mg/kg with no discontinuations or serious or severe adverse events. In contrast, the maximum tolerated single dose of olipudase alfa achieved in the absence of a regimen to gradually reduce sphingomyelin stores was five-fold lower at 0.6 mg/kg (10). Two patients with IARs required a repeat or reduced dose during dose escalation, and a longer period of dose escalation with an additional 0.6 mg/kg dose will be implemented in future clinical trials.

Almost all adverse events observed in the study were mild, with half considered related to treatment. The large number of reported adverse events likely reflects the extent of continuous monitoring (72 hours post-infusion) of hospitalized patients along with the stringent criteria used to identify IARs and determine whether the next ascending olipudase alfa dose could be administered. Headache was the most common event overall, as well as the most common IAR. Overall, IARs were predominantly mild and there were no hypersensitivity reactions or events of cytokine release syndrome. There were no adverse events associated with hematologic variables, and in fact, thrombocytopenia, which is common in patients with ASMD (2) improved from moderate at baseline to mild at week 26, with a mean increase in platelet count of 18%.

Hypersensitivity IARs arising after several doses and typically occurring during or immediately after infusions are common with ERT, and are often associated with the induction of enzyme-specific antibodies (25, 26). None of the patients in this study developed antibodies to olipudase alfa or displayed characteristic hypersensitivity IARs. Patients with ASMD are known to have residual levels of ASM protein (27). Given that the 5 patients enrolled in this trial have detectable residual enzyme activity (as shown in Table 1), and are therefore not naïve to the protein sequence of olipudase alfa, it is perhaps not surprising that antibodies to the enzyme were not generated, and immunologically-driven IARs were not observed. Single IARs occurring in two patients during dose escalation were consistent with an acute phase response involving transient elevations in total bilirubin (one patient), ferritin, and hsCRP peaking between 24–48 hours post-infusion. Acute phase responses characterized by “flu-like” symptoms attributed to increased levels of inflammatory mediators have been reported following other drugs, such as bisphosphonate administration (28), and similar responses accompanied by transient increases in hsCRP and ferritin were observed in four patients receiving single doses of olipudase alfa in the previous phase 1 study (10). A possible initiator of the acute phase reaction following olipudase alfa treatment is ceramide, a major product of sphingomyelin metabolism and a well-known signaling molecule in cytokine release, inflammation, and apoptosis (29, 30). Increased concentrations of plasma ceramide were observed by 24 hours in both patients with acute phase reactions in the present study and in 3 of 4 patients with acute phase reactions in the single-dose study (10), and tended to precede elevations in inflammatory mediators. However, in the current repeat-dose study, the acute phase reactions occurred at intermediate doses (1.0 and 2.0 mg/kg) when plasma ceramide levels were lower than baseline levels. If ceramide or another sphingomyelin metabolite is triggering the inflammatory response, the repeat-dosing results indicate that the pool of ceramide involved is distinct from that measured in plasma. The rises in hsCRP, IL6 and IL8 appear to be more informative biomarkers of the acute phase response. The source of the transient elevations in plasma ceramide also is unknown, but could reflect the metabolism of sphingomyelin in circulating cells, lipoproteins, or endothelial cell surfaces. Both pre- and 48-hour post-infusion ceramide levels declined with each successive dosing step and coincided with the reduction in organomegaly and normalization of the lipid profile, suggesting that the strategy of escalating olipudase alfa doses to gradually debulk accumulated sphingomyelin in tissues helped to mitigate the impact of ceramide release.

Exposure increased in an approximately dose-proportional manner with no accumulation following repeated olipudase alfa infusions every other week. Mean t_1/2z values ranged from 21 to 24 hours. The half-life following administration of single olipudase alfa doses of 0.1–1.0 mg/kg ranged from 10–15 hours (10). In contrast, the circulating half-lives of approved ERTs typically range from <10 minutes (mannose-exposed acid beta-glucosidase) to ~2 to 4 hours (all others). While the cause and biological significance of the long circulating half-life of olipudase alfa is unknown, binding of olipudase alfa to sphingomyelin contained within circulating lipoprotein particles or exposed on the surfaces of cell membranes (31) may be contributing to the enzyme’s longer circulating t_1/2.

Consistent with a recent assessment of liver pathology in adult patients with NPD B (11), variable degrees of fibrosis were present in patients at baseline, and single-point increases in fibrosis scoring were noted in three patients. However, due to the small patient population size, the single (6-month) time point, and tissue samples less than the recommended 1.5 cm for assessment of fibrosis in some instances, it is unclear whether fibrosis truly increased, or whether scoring changes reflected liver biopsy sampling variation due to the heterogeneous nature of liver fibrosis, or a change in the appearance of fibrosis resulting from a reduction in liver volume. In this regard, the decrease in liver transaminases suggests improvement rather than disease progression. One patient had a small inflammatory focus of lymphocytes at week 26 in the absence of changes in liver enzymes or bilirubin. A patient in the single-dose phase 1 study also had small inflammatory foci of lymphocytes (10). The relevance of these foci cannot be determined from these small studies. In the ASMKO mouse model, there was no liver-associated toxicity in animals administered up to 30 mg/kg following a debulking regimen using a lower dose (9).

The targeted maintenance dose of 3.0 mg/kg was selected in order to maximize penetration of olipudase alfa into hard-to-reach tissues such as lung. Although this study was not designed to evaluate efficacy or dose-response relationships, promising effects of olipudase alfa were observed at the cellular, organ, and functional levels, even with the relatively short duration of treatment. As observed in a natural history study of NPD B, spleen and liver volumes were enlarged in all patients at baseline (2, 6). Spleen volume has been proposed as a surrogate marker of NPD B disease burden and severity, correlating well with hepatomegaly, growth impairment, lipid profile, hematologic parameters, and pulmonary function (2). Following olipudase alfa administration, liver sphingomyelin assessed by histomorphometry decreased by 87%, while spleen and liver volumes (as MN) decreased by 29% and 22%, respectively. These results provide compelling pharmacodynamic evidence that olipudase alfa is engaging its intended target, lysosomal sphingomyelin, in biologically-relevant tissues. The reduction in sphingomyelin and organ volumes was accompanied by improvements in several biomarkers in the downstream pathophysiology of ASMD (32–34), including those related to macrophage proliferation (chitotriosidase, CCL18, and ACE) and hepatocyte function (ALT, AST, GGT and bilirubin). In most patients, ALT, AST, and bilirubin decreased relative to baseline by week 26 of treatment with olipudase alfa, with bilirubin that was elevated at baseline normalizing. These reductions tended to be associated with reductions in spleen and liver volume.

Improvements were also seen in radiographic and functional measures of infiltrative lung disease (including improvement in DLco from moderate to mild impairment in gas exchange), which, if untreated, tend to worsen with age in patients with NPD B (6). Coronary artery disease was identified as an under-recognized and serious morbidity in patients with NPD B (5). Lipid parameters, including HDL-C and LDL-C, improved with olipudase alfa treatment regardless of whether patients were on a stable regimen of lipid-modifying drugs. In untreated patients with NPD B, atherogenic lipid profiles typically worsen with age (6), and lipid abnormalities may be associated with early coronary artery disease as in the general population (35). While the impact of lipid profile improvements on cardiac disease risk following ERT is unknown, improved lipid profiles have also been reported with ERT in patients with Gaucher disease (36), indicating that lipoproteins can be responsive biomarkers for some lysosomal storage diseases. Complementing the improvements at the cellular and organ system level, positive trends in quality of life outcomes including reduced pain and fatigue were reported.

Enzyme replacement therapy is a first-line treatment for several lysosomal storage disorders including Gaucher, Pompe, and Fabry disease, and several mucopolysaccharidoses (I, II, IV, and VI) (37, 38). In the absence of any approved, etiology-based therapy, patients with ASMD must rely on palliative and supportive care for symptom management. This study demonstrates proof-of-concept for olipudase alfa in the treatment of nonneurological manifestations of ASMD using dose escalation to prevent toxicity. Over 26 weeks, treatment with every-other-week infusions of olipudase alfa was well tolerated with no serious adverse events or hypersensitivity reactions. Two patients required an expanded dose escalation period, thus, while the dose escalation approach was successful, the regimen may need to be tailored based on individual patient responses. Although interpretation of efficacy outcomes is limited by the small number of patients, olipudase alfa treatment was associated with improvement in objective biomarker responses, clinical measures, and patient-reported quality of life outcomes. While this was a small, short-term trial in adult patients only, its results support the further clinical development of olipudase alfa.

Highlights.

• Olipudase alfa is an ERT in development for ASMD

• Within-patient, dose-escalation was used to debulk accumulated sphingomyelin

• This novel dosing regimen enabled safe, tolerable, and effective repeat dosing