Académique Documents
Professionnel Documents
Culture Documents
Pharmacology 2011;88:193200
DOI: 10.1159/000331341
Fresenius-Kabi Deutschland GmbH, Oberursel, and b Parexel International GmbH, Berlin, Germany;
Clintec Division of Anesthesia and Intensive Care Medicine, Karolinska University Hospital, Huddinge, Sweden;
d
J&P Medical Research Ltd., Vienna, Austria; e Beth Israel Deaconess Medical Center, and f Harvard Medical School,
Boston, Mass., USA
Key Words
Parenteral nutrition Phosphorus Interaction
Abstract
Objective: The primary aim of the present investigation was
to determine and compare the pharmacokinetic (PK) profiles
of inorganic phosphate in serum and urine after intravenous
administration of sodium glycerophosphate and inorganic
sodium phosphate. Additionally, study product safety profiles were evaluated. Subjects and Methods: In total, 27
healthy, white volunteers (17 male/10 female) were enrolled
in this double-blinded, randomized, 2-sequence, crossover
study and were assigned to receive an organic test drug (sodium glycerophosphate) and an inorganic reference preparation (sodium phosphate) on 2 occasions. Validated analytical methods were used, and concentrations of total inorganic phosphate in serum and urine were determined over 24 h
following a single 4-hour continuous intravenous infusion of
test and reference drugs at a dose of 80 mmol. Study days
were separated by washout periods of 7 days. An analysis of
variance, based on population means and 90% confidence
intervals (CIs), was used for testing bioequivalence (BE; range
0.81.25) between investigational products. Results: The
geometric means of the ratio of the point estimates and corresponding 90% CIs for the area under the concentration-
Part of the present work was presented at the 2nd World Congress
on Bioavailability and Bioequivalence, Pharmaceutical R&D Summit,
on June 68, 2011, in Las Vegas, Nev., USA.
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM
Pharmacology 2011;88:193200
if the drug is given in combination with essential components of TPN solutions such as amino acids, fats, glucose and electrolytes [68].
Against this background, we carried out the present
study to measure and compare the PK profiles of inorganic phosphate in serum and urine following intravenous administration of the two investigational formulations. Additionally, total calcium, sodium, potassium in
serum and urine as well as serum levels of potential surrogate markers of phosphate homeostasis such as PTH
and calcitonin were compared in both treatment groups.
Topp et al.
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM
Introduction
Study Drugs
Study drugs were provided by Fresenius-Kabi Deutschland
GmbH, Bad Homburg, Germany. Glycophos (sodium glycerophosphate; 20 mmol/20 ml), manufactured by Fresenius-Kabi
Norge AS (Halden, Norway; batch No. 12DDL03) diluted with
physiological sodium chloride for infusion served as the test
drug. The reference drug was pure sodium phosphate (Sodium
Phosphates Injection, USP, 3 mmol/ml), manufactured by Hospira
Inc. (Lake Forest, Ill., USA; batch No. 87-630-DK). The drug was
diluted appropriately with physiological sodium chloride to achieve
an equimolar intravenous phosphate dose of 80 mmol over 4 h.
No prescribed concomitant medication or therapy was allowed 4 weeks prior to the start of the experimental period up
until the completion of the end-of-study visit. No over-the-counter drugs including herbals such as St. Johns wort, vitamins and
minerals except for paracetamol or nasal anticongestion medicine were allowed 2 weeks prior to the first administration of
the investigational medicinal product until completion of the
study. Occasional oral intake of paracetamol of up to 1 g/day was
permitted.
The dietary regimen was similar for all subjects in both study
periods. The standardized diet consisted of 3 meals served approximately 4 h (12.00 a.m.), 8 h (4.00 p.m.) and 12 h (8.00 p.m.)
after the start of study drug infusion. The start of infusion of
study drugs was approximately at 8.00 a.m. The amount and composition of this diet were based on the normal caloric need of an
adult with normal weight and slight physical activity. Standardized meals contained regular amounts of phosphorus. A volume
of 200 ml of phosphorus-reduced still water was provided before
dosing and every 2 h of the 12-hour study period after study drug
administration. Afterwards, there was no restriction on water
consumption. The inorganic phosphate concentration in the water was 0.12 mg/l (0.0039 mmol/l), which was considered negligible compared to the amount of phosphate administered intravenously (80 mmol).
Study Day Flowchart
The clinical study flowchart included a screening phase, an
experimental in-house phase of 10 days comprising two treatments, separated by a washout phase of 7 days, and a follow-up
examination period of 47 days after the last infusion. Healthy
volunteers were admitted to the clinical research ward 3 days prior to the first study drug administration. Before study drug administration an indwelling catheter was inserted into a peripheral forearm vein to monitor serum concentrations of inorganic
phosphate at defined time points. Heart rate and blood pressure
were measured after resting for a few minutes in a supine position.
Glycerophosphate Is Interchangeable
with Inorganic Phosphate
According to the randomization list, the appropriate study medication was administered to subjects. Volunteers were scheduled
to undergo a final physical examination, including an electrocardiogram and a laboratory parameter assessment within the follow-up period.
Blood/Urine Sampling Procedures and Analytics
For the measurement of inorganic phosphate serum levels, venous blood was withdrawn from a forearm vein at baseline and 1,
2, 3, 3.92 (5 min before end of infusion), 4.5, 5, 6, 7, 8, 10, 12, 16,
24 and 30 h after administration of study preparations. In total,
15 blood samples were taken for PK analysis for each of the 2 study
drugs. Blood samples were collected in sterile one-way tubes
(Sarstedt Serum-Monovette Serum-Gel, Nmbrecht, Germany).
Blood samples were allowed to clot for 30 min before processing.
Blood specimens were centrifuged at 1,500 g for 10 min; cells were
discarded, and approximately 2.6 ml of serum was retained and
quick frozen at 20 C. Sampling periods and handling procedures of the serum samples were identical on both study drug
administration days. Thus, the total blood loss for PK analysis was
approximately 150 ml.
After study drug administration, a 24-hour urine collection
was performed from before dosing to 8, 810, 1016 and 1624 h
for the determination of the total amount of inorganic phosphate,
calcium, sodium and potassium excreted. At the end of the collection period, the urine was gently shaken and a 30-ml volume
sample was acidified by adding 200 l of 6 mol/l HCl. Ten milliliters of the acidified urine were stored at room temperature until
shipment to the analytical laboratory.
Specimens were analyzed by use of a Roche Cobas 6000/c501TM
analyzer according to standard operating procedures. Inorganic
phosphate was measured indirectly by photometric detection of
ammonium phosphomolybdate. Ammonium phosphomolybdate
is the result of a chemical reaction of inorganic phosphate with
ammonium molybdate. The lower limit of detection defined by
the manufacturer is 0.10 and 1.1 mmol/l for serum and urine samples, respectively. All specimens were handled as described for
study drug administration days.
Pharmacokinetics
Noncompartmental analysis was used in the present study.
The area under the concentration-versus-time curves from time
0 to 24 h (AUC024) was estimated by using the log-linear trapezoidal rule. The maximum concentration (Cmax) and the concentration 5 min before the end of the infusion (Css) after drug administration were determined directly from the serum concentrationtime curves. The terminal elimination rate constant (k z) was
estimated from the slope of the terminal exponential phase of the
semilogarithmic serum concentration-versus-time profile using
more than 3 data points. The time to reach Cmax (Tmax) and the
apparent terminal elimination half-life (t1/2 z) and k z were defined
as secondary PK parameters. t1/2 z was calculated as 0.693/k z. The
total amount of inorganic phosphate excreted in urine from 0
24 h (Ae024) was calculated for both study drug administration
days. Urine data were corrected for individual circadian baseline
values derived from measurements from 24 to 48 h before study
drug administration (Ae24 bc).
Pharmacology 2011;88:193200
195
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM
Randomization List
Subjects were scheduled to receive a single intravenous dose
of both study preparations. For that purpose, healthy volunteers
were randomly assigned either to sequence 1 or sequence 2 according to the randomization list (SAS for Microsoft Windows
version 9.2, SAS Institute Inc., Cary, N.C., USA).
Subjects randomized to sequence 1 received the test drug on
study day 3, while the reference drug was administered after a
washout period of 7 days (test/reference regimen). Subjects randomized to sequence 2 were handled as described for sequence 1,
but the treatment allocation was reversed (reference/test regimen).
Statistical Calculations
Initially, a sample size of 18 subjects was considered appropriate for the present investigation. However, a blinded interim analysis of urine concentrations of inorganic phosphate showed that
interindividual variability was higher than expected, and thus a
recalculation of the sample size was necessary. The procedure of
sample size adjustment was approved by the local ethics committee and did not require the adaption of the -error of 0.05, due to
the blinded re-estimation procedure. The final sample size considered adequate was 22 (11 subjects/group). In total, 25 subjects
were included in PK and safety analyses [11].
An ANOVA for a 2-sequence crossover design was employed
to test for differences in the means of each primary PK parameter.
This traditional analysis tested for treatment and sequence (fixed
Parameter
Sequence 1
Sequence 2
Total
(TR regimen) (RT regimen)
Males/females, n
Age, years
Height, cm
Weight, kg
BMI, kg/m2
4/10
34.086.6
176.4811.2
74.389.5
23.982.1
6/7
33.187.7
172.1811.1
75.6815.3
25.483.3
10/17
33.687.1
174.3811.2
74.9812.4
24.682.8
Data are presented as arithmetic means 8 SD, except for gender. TR = Test/Reference; RT = Reference/Test; BMI = body mass
index.
Results
Demography and PK
Healthy volunteers had a mean age of 33.6 8 7.1 years
(range 1945 years), a mean weight of 74.9 8 12.4 kg
(range 54.898.0 kg), a mean height of 1.74 8 0.11 m
(range 1.571.94 m) and a mean body mass index of 24.6
8 2.8 kg/m2 (range 20.529.5; table1).
Table 2. PK outcome parameters for inorganic phosphate in serum and urine after intravenous administration of study drugs to healthy
subjects
PK parameter
Study drugs
Serum (n = 25)
AUC024, mmolh/la
Cmax, mmol/l
Css, mmol/l
Tmax, h
t1/2 z (n = 24), h
Urineb (n = 21)c
Ae24 bc, mmol
T drug
R drug
T/R ratio
90% CI
38.70
2.87
2.87
4.0
2.06
37.34
3.36
3.35
3.97
2.25
1.04
0.85
0.85
n.d.
n.d.
1.001.07
0.840.87
0.840.87
n.d.
n.d.
52.87
62.92
0.84
0.770.92
196
Pharmacology 2011;88:193200
Topp et al.
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM
PK data are presented as geometric means and 90% CIs. T = Test (Glycophos); R = Reference (Sodium Phosphates Injection, USP);
Cmax = peak concentration; Css = concentration 5 min before end of infusion (3.92 h); AUC = area under the concentration-versus-time
curve; Ae24 bc = total amount of phosphate excreted in urine in 24 h; n.d. = not determined.
a ANOVA was calculated with AUC
024 as dependent variable and Css as covariate.
b Parameter was corrected for individual baseline levels.
c Due to incomplete urine samples in 4 subjects.
300
1.6
250
Test drug
Reference drug
200
150
100
50
0
1.5
1.4
1.3
1.2
1.1
1.0
0.9
0.8
12
16
20
Time (h)
12
16
20
Time (h)
Glycerophosphate Is Interchangeable
with Inorganic Phosphate
3.5
3.0
Test drug
Reference drug
2.5
2.0
1.5
1.0
0.5
0.0
0.5
0
Pharmacology 2011;88:193200
12
16
20
24
Time (h)
197
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM
4.0
Table 3. PK parameters for calcium, sodium and potassium in serum and urine after intravenous administration of test and reference drugs to healthy subjects
PK parameter
Serum
Sodium AUC024, mmolh/l
Calcium AUC024, mmolh/l
Potassium AUC024, mmolh/l
Urinea
Sodium Ae024, mmol
Calcium Ae024, mmol
Potassium Ae024, mmol
Study drugs
T drug
R drug
T/R ratio
90% CI
3,318
52.74
100.6
3,323
52.67
101.5
1.00
1.00
0.99
1.001.00
1.001.01
0.981.01
211.1
2.61
62.50
187.2
2.89
61.06
1.13
0.90
1.02
1.051.21
0.811.00
0.961.09
Discussion
The central finding of this analysis was that point estimates and 90% CIs of the ratios of test to reference for
AUC024 and Cmax of inorganic phosphate in serum were
within the BE range from 0.80 to 1.25. Baseline-corrected
(Ae24 bc) excretion of phosphate in urine marginally failed
to demonstrate BE between study formulations. Total calcium, potassium, sodium (table 3) and PTH in serum
were also measured during this investigation and proved
to have value as a secondary outcome parameter to serum
phosphate PK. The extraordinarily high intravenous
dose of 80 mmol of phosphate administered to subjects in
the present investigation seems to have had a minimal-tomoderate effect on the changes in the serum levels of sodium, potassium and total calcium versus baseline (fig.2).
For example, the small elevation of sodium in serum detected in the present investigation is likely the direct result of the use of sodium phosphate solutions and, therefore, simply indicates that a sodium-loaded drug was administered. Similarly, the observed minor, not clinically
relevant changes in concentrations of potassium in serum and urine appear to be the result of study-drug-related sodium load counterregulation.
198
Pharmacology 2011;88:193200
0.6
0.4
Test drug
Reference drug
0.2
0.0
0.2
0.4
0
12
16
20
24
Time (h)
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM
PK data are presented as geometric means and 90% CIs. T = Test (Glycophos); R = Reference (Sodium Phosphates Injection, USP); AUC = area under the concentration-versus-time curve; Ae 024 = total amount excreted
in urine in 24 h.
a
Parameter was not corrected for individual baseline levels.
at doses of up to 80 mmol/l (fig.1c). Only a modest, statistically not significant difference in the means of the
serum inorganic phosphate AUC024 of approximately
3.5% was evident between groups; the AUC in the test
group was smaller when compared with the reference
drug (table2). We are speculating that this small difference is due to the differences in the protein binding of test
and reference drugs in serum. An alternative or complementary explanation could be that the test drug is excreted in urine in its nonhydrolyzed form in very low proportions of approximately 5% or less. This concept is strengthened if one keeps in mind that the phosphate assay utilized
in the present study was practically unable to detect the
nonhydrolyzed glycerophosphate molecule, neither in serum bound to serum proteins nor in urine. This could
help to understand why Ae24 bc in urine failed to demonstrate BE between test and reference drugs, while estimations of the AUC024 of inorganic phosphate in serum did
not (table2).
With regard to safety of the investigational drugs, no
obvious difference between formulations was detected.
Both study drugs were well tolerated, despite the relatively high dose of 80 mmol as a 4-hour infusion. A total of
15 treatment emergent untoward effects were observed,
all of which were mild to moderate in intensity and comprised symptoms such as headache, nausea, dry mouth,
gastrointestinal disorders, hypocalcemia and others, but
resolved either spontaneously or after having taken appropriate medical treatment measures. Both study groups
were equally affected.
One minor limitation of the present examination is
the relatively large number of different laboratory parameters we were looking at. However, this approach was
considered most useful in the identification of the most
suitable parameter for testing the BE of phosphate formulations. The explorative character of the present study
supported our previous expectations and confirmed that
the measurement of phosphate levels in serum is a costefficient, reproducible and reliable method for BE testing
of phosphate.
Another question that should be addressed relates to
the dietary intake of phosphorus and its effects on serum
levels of phosphate. In this regard, a large survey of nutrient consumption in the USA found the average phosphorus intake to be approximately 1.4 and 1.0 g/day in men
and women, respectively [16]. The estimated total daily
oral intake of phosphorus after study drug infusion over
a period of 4 h was approximately 1.1 g (35.0 mmol),
which corresponds to about 44 % of the intravenously infused dosage of phosphate (80 mmol). However, this cal-
Glycerophosphate Is Interchangeable
with Inorganic Phosphate
Pharmacology 2011;88:193200
199
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM
quently, the assessment of these outcome parameters appears to be of limited value as a primary outcome parameter in BE studies on phosphate. Moreover, the urine profiles of inorganic phosphate, total calcium, potassium
and sodium did not provide additional information to
data derived from serum concentration measurements
alone. Obviously, levels of inorganic phosphate in serum
are a solid measure in the assessment of BE between different phosphate formulations after parenteral administration.
Another finding of the present study was the confirmation of previous studies results of circadian rhythm
changes of electrolyte serum levels [1214]. The levels of
phosphate in serum were lowest in the morning hours
and began to increase slowly during midday and afternoon before declining again during the evening and
night hours (fig.1b). The observed changes were moderate and within the normal physiological range of variability found in textbooks [2]. The reason for phosphate serum level increases in the afternoon hours has not yet
been clarified. Importantly, no significant differences
were detected between baseline-corrected and uncorrected data sets (p 1 0.05). Thus, the circadian fluctuations of
serum phosphate did not affect the reliability and robustness of BE testing, and therefore individual time-adjusted
corrections were performed exemplarily only for urine
excretion of phosphorus in study subjects (table2).
Organic glycerophosphate and inorganic sodium
phosphate are chemically different, and the conversion
from glycerophosphate into inorganic phosphate in blood
is the rate-limiting step, which is dependent on the presence and optimal activity of hydrolytic enzymes in serum
and target tissues [15]. The kinetics of glycerophosphate
were studied in previous in vitro studies, which have
shown that the hydrolysis rate of glycerophosphate to
glycerin and inorganic phosphate in serum is about 0.09
0.12 mmol/l/h, suggesting that approximately 1215
mmol of glycerophosphate is hydrolyzed per day [15].
However, the maximum amount of glycerophosphate
that can be hydrolyzed per day appears to mainly depend
on the activity of alkaline phosphatase in blood [15]. Given that these values are valid in clinical settings, the daily dose of 20 mmol inorganic phosphate equivalent should
be sufficient in the vast majority of patients in need of
phosphorus supplement. However, the accuracy of these
previous in vitro data in relation to the in vivo situation
has not been established in clinical studies yet.
From our own experiment, we learned that the hydrolytic conversion of glycerophosphate into inorganic phosphate is a fast, nonsaturable and almost complete process
Disclosure Statement
H.T., O.H., M.W., H.S. and M.R. are employees of FreseniusKabi Deutschland GmbH. M.G. is an employee of Parexel international GmbH, a contract research organization in Berlin, Germany. J&P Medical Research Ltd. is an international independent
life science institute basically operating according to the publicprivate-partnership concept. C.J. is the chief executive officer of
J&P Medical Research Ltd. S.B. has a contract as clinical consultant of Fresenius-Kabi Deutschland. Fresenius-Kabi Deutschland
GmbH is the sponsor of the present study and manufacturer of
the test drug. All authors, including S.B., declare having no relationship with other companies that make products relevant to the
paper and have no conflicts of interest with the present work.
References
200
Pharmacology 2011;88:193200
Topp et al.
Downloaded by:
St. Andrews University
138.251.14.35 - 1/15/2014 6:01:28 PM