Pharmacology Journal

Reactive oxygen species (ROS), like superoxide anion (O2.−) and hydrogen peroxide (H2O2), play an important role in health and disease. They have been implicated in the pathophysiology of different disease states, including anthracyclin-induced cardiotoxicity (AIC), inflammatory bowel disease, ischaemia/reperfusion injury and neurodegenerative conditions. The hypothesis is that in these pathological conditions, relatively large amounts of ROS are produced which cause functional damage to many tissues and even apoptosis. The underlying mechanism for the deleterious effect of ROS on tissues is not fully understood, but includes cell membrane damage due to lipid peroxidation, and direct damage to proteins and DNA.

There are different endogenous defence mechanisms against the ROS damage such as superoxide dismutase (SOD), catalase, peroxidases and vitamin A and E which all share free radical scavenger properties.

Three isoforms of SOD exist in humans: cytosolic Cu,Zn SOD (SOD1), mitochondrial MnSOD (SOD2) and extracellular Cu,Zn SOD (SOD3), of which the intracellular forms are the most abundant. The endothelial cell surface is protected by SOD3, but this protection seems insufficient in many clinical conditions, and it has therefore been suggested that additional protection may be of benefit. Indeed, over the last decade therapeutic use of SOD has been explored, but there is consensus that up to now this has been of limited value. Likely explanations for the limited success of exogenously administered SOD are that its intracellular isoforms hardly bind to the endothelium and that they are relatively short-lived. In addition, particularly for SOD3, which is an attractive candidate for therapeutic use, the manufacturing process is difficult.

Therefore, there is a need for SOD preparations that are relatively easy to manufacture, show a reasonably long residence time in the body and will be taken up by organs that are relatively poorly protected against free radicals. This has resulted in the development of PC-SOD (recombinant human SOD1 covalently coupled to an average of four molecules of lecithin) and a chimeric recombinant superoxide dismutase consisting of SOD2 and SOD3.

PC-SOD has a higher affinity to the cell membrane, enhanced distribution to various tissues and a prolonged systemic half-life compared with SOD1 alone. In addition, it has a 4.5-fold increase in oxygen-radical scavenging effects resulting in a 100-fold increase in protective effects against vascular endothelial cell injuries, compared with unmodified SOD. Preclinical data have shown that PC-SOD is effective in several models, including inflammation, chemotherapy-induced cardiotoxicity, ischaemia-reperfusion injury and motor dysfunction after spinal cord injury. The preclinical data also indicated that PC-SOD is well tolerated, although multiple doses to primates were associated with the presence of lipid inclusion bodies in renal tubular cells. However, this was entirely reversible and not associated with functional impairment or necrosis of cells. Thus, PC-SOD is a potentially protective agent in pathological conditions mediated by free radical overproduction.

In a previous study in Japanese volunteers, where doses up to 20 mg were investigated, PC-SOD was well tolerated, but the duration of increased elevation of SOD activity was only 3 h, which is too short to be of likely clinical relevance. The current study was performed to assess the tolerability, pharmacokinetics (PK) and effects of single (higher) ascending doses of PC-SOD in healthy White volunteers. The study was designed so that detectable SOD activity would be present for a period of 12–24 h. Furthermore, special attention was given to the effects of the compound on renal function and tubular integrity, as this was an issue with very high doses of PC-SOD in preclinical experiments. The effects on renal function were assessed by measurement of the urinary excretion of specific markers for tubular damage (N-acetyl-β-glucosaminidase (NAG), α- and π-glutathione S-transferase (GST) and microalbumin.

Subjects and methods

The study protocol was approved by the Medical Ethical Committee of Leiden University Medical Center (LUMC) and performed according to the principles of the International Conference on Harmonization and Good Clinical Practice and the Helsinki Declaration. Written, informed consent was obtained from all subjects before study entry.

Subjects

Eight healthy subjects (four female and four male) aged between 18 and 45 years and within 20% of the normal body weight range relative to height and frame size were included in this double-blind, placebo-controlled, four-way cross-over study. Subjects were included after a full medical screening showing no clinically significant abnormalities. Subjects were excluded in case of a history of drug allergy or hypersensitivity, or drug, alcohol or nicotine abuse.

Study medication

The subjects were dosed four times using an ascending dose schedule with randomized placebo as summarized. Dose escalation was performed when no significant clinical abnormalities were observed after the previous lower dose. The wash-out period between doses was at least 1 week.

The PC-SOD preparation consists of an average of four molecules lecithin derivative covalently bound to the human derived CuZn-SOD, produced by genetic recombination using Escherichia coli as a host cell. The lecithinized product has 3 × 10³ U SOD activity per mg. For this study, a single batch of the lyophilized formulation also containing sucrose was used. Placebo consisted of sucrose. The final preparation that was administered consisted of PC-SOD or placebo diluted with distilled water and 5% mannitol.

Study days

The subjects were admitted to the research unit after an overnight fast. After preparation and baseline measurements, the study drug was administered intravenously over 60 min. During the study days, frequent measurements of vital signs, 12-lead ECG recording and evaluation of adverse events, blood sampling and fractionated urine collection took place. The subjects remained in the unit for 24 h and returned for follow-up assessments and blood sampling at 48 and 96 h after dosing. During the study days, subjects used standard meals and abstained from using xanthine-containing drinks or food.

Sampling and assays

Serum PC-SOD concentrations and SOD activity were measured in venous blood samples taken predose (twice), at 20, 40, 60, 65, 75, 90 min, and at 2, 3, 4, 8, 12, 24, 48, 96 and 168 h after start of the infusion. The last time point coincided with the first predose sample of the subsequent study day. After collection, the tubes were kept at 4°C for and subsequently centrifuged at 2000 g for 10 min at 4°C. The separated serum was stored at −20°C until analysis within 1 month after sampling.

Urine was collected during the study period over the following time spans: 0–4, 4–8, 8–12, 12–24 and 24–48 h. Immediately after voiding, urine samples were stored at 4°C and aliquots of 2 ml were taken from each collection period and stored at −20°C until analysis within 1 month after sampling. Samples to assess antibody formation were taken at completion of the last administration and at 1 and 3 weeks after the last dosing.

Blood samples for routine haematology and biochemistry were taken before and at 24 h after each infusion.

Serum and urinary PC-SOD concentrations were measured using an enzyme-linked immunosorbant assay consisting of an antibody against human Cu, Zn-SOD, and a second antibody against human Cu, Zn-SOD conjugated with horseradish peroxidase. The assay has a lower limit of quantification 626 ng ml^−1. Intra- and interassay variability were investigated at PC-SOD concentrations of 626, 2500 and 10 000 ng ml^−1 for serum and 626, 5000 and 20 000 ng ml^−1 for urine; each concentration in triplicate. The coefficients of variation (CV) for the intra-assay variability for the respective concentrations were 5.6, 3.2 and 1% in serum, and 7.3%, 2.3% and 2.3% in urine. The CVs for the interassay variability in serum and urine were 7.9, 2.7 and 1.3% and 4.9%, 8.2% and 1.2%, respectively. Repeated freezing and thawing had no appreciable effects (CV < 10% after three freeze–thaw cycles).

PC-SOD activity was measured using a nitrite method previously described.

The test is based on the principle that when hypoxanthine and xanthine-oxidase are brought together, superoxide anion is formed. When superoxide anion reacts with hydroxylamine, nitrite is formed and this can be measured by colour densitometry with the aid of a colouring reagent. SOD present in serum will inhibit the formation of nitrite by reacting with the superoxide anion. Serum SOD activity was quantified using the reduction in superoxide anion generation caused by serum added to the system. The assay had a lower limit of quantification of 3 µg ml^−1. Intra- and interassay variability was 3.9% and 7.5% for serum and 6.8% and 10.9% for urine, respectively. Both assays were performed at Daiichi Pure Chemicals Co. Ltd (Ibaraki, Japan).

Antibody formation against PC-SOD was measured by quantification of specific IgE, IgG and IgM titres. For anti-PC-SOD IgE antibody measurement, antihuman IgE mouse monoclonal antibody (alkaline phosphatase labelled) was used as secondary antibody. The titre was qualitatively judged using the level of the positive control (human antiperennial rye class IgE antibody) as the reference value and was described as positive if the titre was >0.2 IU ml^−1. For anti-PC-SOD-IgG+IgM measurements, antihuman IgG and IgM mouse monoclonal antibody (alkaline phosphatase labelled) was used as secondary antibody. The titre was qualitatively judged in reference to the antibody level of a pooled normal human serum sample (negative control) and indicated as positive if the value exceeded by 3.1-fold the average value of four normal human serum samples.

Urinary NAG activity was measured using a commercially available colorimetric assay (Roche Diagnostics, Basel, Switzerland; reference value 1.39–3.23 U per 24 h, detection limit 1 U l^−1). Urinary excretion of α-GST and π-GST was determined using validated quantitative enzyme immunoassays (Biotrin, Dublin, Ireland; limit of detection α-GST 0.09 µg l^−1 and π-GST 1.72 µg l^−1 and both intra- and interassay variability <6.9%). Urinary microalbumin and creatinine concentrations were measured using routine methodology at the central laboratories for clinical chemistry of LUMC.

Data analysis

Vital signs, ECG and laboratory parameters were analysed by generating average graphs of parameters over time per treatment. If these graphs suggested possible differences between treatments, areas under the effect curve over the first 12 h divided by the corresponding time span (AUC) were calculated and compared between treatments using factorial analysis of variance (factors subject and treatment).

The cumulative urinary excretion of NAG, α-GST, π-GST and creatinine over 0–4 h and over 0–48 h were calculated. For values below the detection limit, the detection limit was used. The cumulative 24-h microalbumin excretion was evaluated as the microalbumin over creatinine ratio. The values were compared between treatments using factorial analysis of variance (factors subject and treatment).

The PK of PC-SOD was assessed using a noncompartmental PK approach for C_max, AUC_{0−48 h} and AUC_{0−7 days}. These parameters were compared between doses after dividing the parameter by the doses using factorial analysis of variance (anova; factors subject and dose) to assess dose linearity. Within-individual ratios for the different doses were compared using paired Student t-tests.

Compartmental PK (using a two-compartment open model) was performed on all of the profiles by analysing the data as arising from a multiple dose sequence. The analyses were performed using nonlinear mixed effect modelling, which estimates all curves for all subjects simultaneously. First-order conditional error estimation with the ‘interaction’ option was used and residual error was modelled as the sum of an additive and a constant coefficient of variation component.

Multiplying the urine weights with the associated concentrations and summing over 48 h calculated the cumulative excretion of PC-SOD. Average renal clearance over this period was calculated by dividing the cumulative renal excretion by the serum AUC over the same time span. Renal clearance was compared between doses using factorial analysis of variance (factors subject and treatment).

The relationship between activity and serum concentration was investigated using graphical and regression techniques. Linear mixed effect modelling was performed to examine the relationship between PC-SOD concentration and SOD activity.

The compartmental PK analyses were performed using NONMEM version V (GloboMax LLC, Hanover, MD, USA). All statistical calculations were performed using SPSS for Windows software (SPSS, Inc., Chicago, IL, USA).

Filed under: Pharmacokinetics