plasma concentration of hgh and anti-hgh antibodies after subcutaneous administration of hgh for 3...

J Pharmacol Toxicol 41 (1999) 1–8

1056-8719/99/$–see front matter © 1999 Elsevier Science Ltd. All rights reserved.

PII: S 1 0 5 6 - 8 7 1 9 ( 9 9 ) 0 0 0 1 7 - 9

Plasma concentration of hGH and anti-hGH antibodies after subcutaneous administration of hGH for 3 weeks to immunosuppressed pigs

Henrik Agersø, DVM, Ph.D.

a,*

, Michael Wilken, M.Sc.

b

, Jørn Drustrup, M.Sc.

c

,

Poul-Martin Haahr, M.D.

d

, and Karin Damm Jørgensen, D.V.M.

e

Departments of

a

Pharmacokinetics,

b

Assay & Cell Technology,

c

Pharmaceutics,

d

Growth Hormone Pharmacology, Novo Nordisk A/S, Novo Nordisk Park, 2760 Maaloev, Denmark

e

Scantox, 36A Hestehavevej, Ejby, 4623 Lille Skensved, Denmark

Received November 16, 1998; revised and accepted March 4, 1999

Abstract

Objective:

The objective of the present study was to evaluate the dosing regimen of immunosuppressants necessary to avoid the forma-tion of anti-hGH antibodies in a pig model.

Animals:

Sixteen pigs were divided into four groups.

Procedure:

Three different immunosuppressive treatments were tested (group 1: Control (no treatment); group 2: 10 mg; group 3: 20mg; and group 4: 40 mg cyclosporine; combined with 2 mg azatioprine and 2 mg prednisolone p.o./kg/day). The treatments were given

from days

2

7 to 22.All groups were dosed subcutaneously (s.c.) with 0.5 mg hGH/kg once daily from days 1 to 22. On the first and the last days of dosing

blood samples were collected to describe the hGH concentration versus time profile. Before dosing and on days 5, 10, and 15 blood sam-ples were collected for measuring hGH antibody formation.

Results:

A dose-dependent decrease in white blood cell counts was observed in all immunosuppressive-treated groups. Groups 1 and 2produced antibodies against hGH during the 22 days of dosing while the formation of antibodies was suppressed in groups 3 and 4. In thecontrol group and group 2 the pharmacokinetic parameters of hGH were influenced by the formation of anti-hGH antibodies. In groups 3and 4, the pharmacokinetic parameters were comparable on the first and the last day of dosing.

Conclusion:

The formation of anti-hGH antibodies influenced the pharmacokinetics of hGH in pigs, but it could be prevented by im-munosuppressive therapy. From the present experiment, a dose of 20 mg cyclosporine, 2 mg azatioprine, and 2 mg prednisolone p.o./kg/day was able to prevent the pigs from producing antibodies without having severe adverse effects. This model may by useful in future ex-periments using sustained release formulations of hGH, and possibly for other compounds that may induce antibody production inpigs. © 1999 Elsevier Science Inc.

Keywords:

Immunosuppressed pig model; antibodies; anti-hGH; pharmacokinetics; immunosuppressant treatment

1. Introduction

Growth hormone (GH) is the major growth promotinghormone in postnatal life. The pulsatile pattern of GH-re-lease by the anterior pituitary is regulated by the interactionof stimulating and inhibiting hormones and influenced byphysiological factors. GH promotes longitudinal bonegrowth mainly through activation of insulin- like growthfactor I (IGF-I), but there may also be a direct effect. Themost important clinical feature of GH deficiency is a lowgrowth velocity resulting in short stature. Since 1958, GH-deficient children have been treated with human GH. Early

diagnosis, treatment, and optimisation of GH dose are im-portant to achieve near normal final height. The dosing reg-imen currently used includes daily doses administered sub-cutaneously (de Muinck Keizer-Schrama and Rikken,1992). Several approaches have been used in attempts to op-timise the dosing regimen; one of the latest approaches hashGH encapsulated into microspheres (Johnson et al., 1996,1997). Results show that serum levels of a therapeutic pro-tein can be sustained for an extended period when encapsu-lated into different formulations of injectable, biodegrad-able microspheres (Lee et al., 1997). It is well known thathuman proteins may be antigenic in animals and cause for-mation of antibodies (Dorchy et al., 1989). This may causeproblems if animals models are used to screen prolongedformulations of hGH because antibodies against a drug canalter the pharmacokinetic parameters of a drug substance

*Corresponding author. Henrik Agersø. Tel.:

1

45 4443 4617; fax:

1

454466 3939; e-mail: [email protected]

2

H. Agersø et al./J. Pharmacol. Toxicol. 41 (1999) 1–8

(Rehlaender and Cho, 1998). Previously, the formation ofanti-insulin antibodies has been demonstrated to be closelycorrelated with significant changes in the pharmacokineticparameters of insulin; the half-life and the volume of distribu-tion increased significantly after the formation of antibodies(Van Haeften et al., 1986). The pharmacokinetics of neo-carzinostatin has been compared to the pharmacokinetics ofneocarzinostatin conjugated with monoclonal antibodies.Data revealed that high molecular weight conjugates per-sisted in the body for longer and at a higher concentrationthan the free drug (Gottaschalk et al., 1991). Likewise, inexperiments examining the clearance of DNA, it was re-ported that the clearance of DNA coupled to anti-DNA anti-bodies was markedly suppressed compared to the clearanceof DNA (Emlen and Mannik, 1980).

These results suggests that circulating immune com-plexes may suppress the normal clearance of drugs andthereby contribute to the persistence of the drug for a pro-longed period of time in the circulation. Generally, immunecomplexes would be expected to be rapidly cleared by thebody, but under certain circumstances the opposite effectmay be observed. If the binding of the drug to the antibodydoes not trigger clearance, then the antibody would act as acarrier protein for the drug, similar to any other protein inthe blood that binds drugs (Rehlaender and Cho, 1998).

The present study was undertaken to evaluate an immuno-suppressed animal model that may be useful in future stud-ies to evaluate sustained release formulations of human pro-teins in laboratory animals.

2. Materials and methods

2.1. Animals

Sixteen female SPF-pigs (Duroc/Yorkshire/Danish Land-race) weighing about 10 kg at the start of the experimentwere included in the study at Scantox, DK-4654. The pigswere supplied by Ellegaard Göttingen Minipigs ApS, DK-4261. The animals were fed a commercially available diet(Altromin 9033, from Chr. Petersen A/S, DK-4100) twicedaily. Furthermore, the animals were given a supply of 200 gautoclaved hay daily.

2.2. Drugs and chemicals

The aqueous test solution of hGH article was supplied byNovo Nordisk A/S. All other chemicals were obtained fromcommercial sources and were of analytical grade.

2.3. Experimental design

The pigs were divided into four groups, consisting offour pigs each. All the pigs included were dosed subcutane-ously in the neck once daily from days 1 to 22. The dosewas 0.5 mg hGH/kg; given in a volume of 0.075 mL/kg.The immunosuppressant dosing regimen consisted of cy-closporine (Sandimmun Neoral), azatioprine (Imural), and

prednisolone (Prednisolon) which were administered oncedaily in corn oil by gavage from days

2

7 to 22 to groups 2,3, and 4 (Table 1).

EDTA-stabilized blood samples for hGH and anti-hGHantibody measurement were collected before dosing and ondays 5, 10, and 15. Additionally, on days 1 and 22 bloodsamples for hGH measurement were collected before dos-ing and at 15, 30, 60, 120, 180, 240, 360, and 480 min afterdosing. Plasma was separated and stored frozen pendinganalysis.

On days

2

7, 1, 8 15, and 22, EDTA-stabilized bloodsamples were collected for white blood cell count (WBC)(Cobas Minos Roche), differential count (May-GrünwaldGiemsa smear staining), and serum creatinine (BoehringerMannheim).

2.4. Analytical methods

Concentrations of hGH in EDTA plasma were measuredin a two-site ELISA. The monoclonal antibody ES3 wascoated onto microtiter plates (Immunoplate, Nunc, Den-mark) in a concentration of 5

m

g/mL in phosphate-bufferedsaline (PBS) (pH 7.4). Another monoclonal antibody, ES7,labelled with biotin was used as secondary antibody, con-centration 1

m

g/mL. NordiTropin (Novo Nordisk A/S) wasused as calibrator material. The assay was performed as atwo-step assay. In the first step, 100

m

L buffer was mixedwith the 10

m

L sample/calibrator in the coated plate and al-lowed to react overnight. After removal of unbound mate-rial, 100

m

L ES7-biotin was added and allowed to react for1 h. Streptavidin-peroxidase followed by TMB substratewas used to visualize the reaction. No cross-reactivity topGH could be detected.

The assay for measuring hGH antibody formation wasperformed by incubating diluted plasma samples (1

1

14) onmicrotiter plates coated with hGH (Norditropin) in a con-centration of 1

m

g/mL in PBS. After incubation for 1 h, theplates were washed to remove unbound material and thepresence of pig antibody demonstrated by a peroxidase la-belled second antibody (rabbit anti-swine immunoglobu-lins/HRP, P0164, Dako A/S). TMB was used as substrate.The relative antibody titer was calculated directly from ab-sorbance data with the response of a sample prior to the firstdose of hGH taken as 100%.

Table 1Treatment regimen: Four pigs were included in each group, the immunosuppressive treatment was initiated 7 days before the first dose of human growth hormone was administered

Group no.hGH (mg/kg)

Cyclosporine (mg/kg)

Azatioprine (mg/kg)

Prednisolone (mg/kg)

1 0.5 — — —2 0.5 10 2 23 0.5 20 2 24 0.5 40 2 2


3

2.5. Pharmacokinetic Analysis

The concentration of hGH in plasma was analysed forthe individual pigs in each group by use of noncompartmen-tal methods using the PC-based software WinNonlin (Ver-sion 1.5, Scientific Consulting Inc. USA). The areas underthe plasma hGH concentration versus time curves were cal-culated according to the trapezoidal rule.

Eq. (1)

The infinite part of the curve was determined as C

last

/

l

z

,with C

last

5

the last concentration and

l

z

5

the slope for thelast phase.

The area under the first moment curve (AUMC), wascalculated as the area under the curve of the product of time(t) and plasma concentration (C

p

), from zero time to infinity:

Eq. (2)

The infinite part of the first moment curve was calculated as:

Eq. (3)

with t

last

5

the last sample time.The mean residence time (MRT), was calculated as follows:

Eq. (4)

The peak concentration (C

max

) and times (t

max

) were readfrom the individual plasma concentrations versus time curves.

3. Results

3.1. Clinical signs

No adverse signs were seen in the control group. Ingroups 2 and 3, minor sporadic symptoms were observed(e.g., swollen left foreleg), and these symptoms may havebeen aggravated by the immunosuppressive treatment. Ingroup 4, the pigs had soft stools and became subdued by thehigh immunosuppressive dosing regimen. Furthermore, twopigs were euthanised, one on day 20 and one on day 22;consequently, the mean values from group-4 pigs on the lastday of dosing only represent two pigs.

The body weight gain tended to decrease with the in-crease in the immunosuppressive load. The mean bodyweight gain was 8

6

1, 6

6

1, 6

6

3, and 3

6

4 kg forgroups 1, 2, 3, and 4, respectively. The mean weight gainduring the experimental period for group 4 was 44% of themean body weight gain of the control group.

3.2. Haematology and serum creatinine

On day 1, after 1 week of immunosuppressive treatment,the white blood cell count in groups 3 and 4 tended to dropcompared with the control group. The WBC count in groups3 and 4 stayed below the counts obtained in groups 1 and 2

AUC∞ CP0∞∫ dt=

AUMC∞ t C⋅ P td0

∞∫=

tlast Clast⋅λz

------------------------Clast

λz2

-----------,+

MRT AUMCAUC

-------------------=

[Fig. 1(a)]. The drop in WBC was primarily caused by adrop in the lymphocytes [Fig. 1(B)], while the neutrophilicgranulocytes stayed rather constant during the dosing period[Fig. 1(C)].

When initiating the immunosuppressive treatment (day

2

7), the mean (SD) serum creatinine level was 120

6

11,125

6

8, 127

6

17, and 145

6

39

m

mol/L for groups 1, 2,3, and 4, respectively. On the last day of dosing (day 22),the mean serum creatinine level was 129

6

18, 99

6

17, 96

6

9, and 95

6

8 for groups 1, 2, 3, and 4, respectively.

3.3. Antibody formation

During the 22 days of dosing, the anti-hGH antibody titerincreased with about 250- and 65-fold in groups 1 and 2, re-spectively (Fig. 2). In groups 3 and 4, the anti-hGH anti-body titer did not increase to the same values; in group 3,the anti-hGH antibody titer stayed below a twofold increaseand in group 4 below a fourfold increase.

3.4. Pharmacokinetics of hGH

Some of the plasma samples contained plasma hGH con-centrations above the highest standard (groups 3 and 4, day1; group 2, day 22). These samples were included in the cal-culations as 500 ng/mL (the level of the highest standard).Extrapolation of the standard curve beyond the highest stan-dard (500 ng/mL) yielded concentrations between 500 and600 ng/mL for most of those samples that were set to 500ng/mL; however, it must be emphasised that some of thepharmacokinetic parameters will be biased by this approxi-mation (especially C

max

and AUC).

3.4.1. Group 1 (control group)

Following s.c. administration of hGH, a peak value ofabout 405–418 ng/mL was observed 56 min after dosing onboth days 1 and 22 (Table 2). After the peak value on day 1,the plasma concentration decreased with a half-life of 106min. On day 22, the plasma concentration level tended tostay constant after reaching the peak value [Fig. 3(a)]. Thus,the half-life could not be calculated on day 22, and theAUC(0-t) values were much higher on day 22 compared today 1 (Table 2).

3.4.2. Group 2 (10 mg cyclosporine/kg)

The concentration versus time curves obtained for group2 were comparable to the curves obtained for group 1 (Fig.3(b). The plasma concentration increased to a C

max

of 338–500 ng/mL approximately 86–105 min after dosing (t

max

).Following the plasma peak, hGH was eliminated with ahalf-life of 86 min on the first day of dosing (Table 2). Onday 22, the plasma concentration tended to stay constant af-ter reaching the peak value; thus, as described for group 1,the half-life could not be calculated, and the result was avery high AUC(0-t) value.

4


Fig. 1. Mean (6SD) differential counts in immunosuppressed pigs. (a) WBC counts. (b) Lymphocytes counts. ( c ) Neutrophilic counts. The immunosuppress-ing treatment was initiated on day 27 (n54). Group 1: control group; group 2: 10 mg cyclosporine/kg; group 3: 20 mg cyclosporine/kg; group 4: 40 mgcyclosporine/kg.


5


Following s.c. administration of hGH on both days 1 and22, the plasma concentration versus time curves were com-parable [Fig. 3(c)]. The time to reach the maximum concen-tration of approximately 459–500 ng/mL (C

max

) was about64–83 min (t

max

). After the peak concentration, hGH waseliminated with an elimination half-life of 71–97 min ondays 1 and 22; hence, the AUC values did not tend to in-crease on the last day of dosing (Table 2).


During the experimental period, two pigs in this groupwere euthanised; thus, from day 22 results represent themean of only two pigs. The results obtained from group 4are comparable to those from group 3. The plasma versustime curves on days 1 and 22 are comparable [Fig. 3(d)],and the pharmacokinetic parameters are almost the same onthe two days of dosing (Table 2).

4. Discussion

In the present experiment, different doses of the cy-closporine (combined with fixed doses of azatioprine andprednisolone) were tested in order to identify a dosing regi-men that would make it possible to immunosuppress pigswithout incurring serious adverse events. It was demon-strated that by immunosuppressing the pigs it was possibleto prevent antibodies from interfering with pharmacokineticparameters of hGH. Three different doses of cyclosporinewere tested, and while the lowest dose was too low to pre-vent the pigs from producing anti-hGH antibodies, the high-est dosing regimen caused too many adverse effects to beacceptable under experimental conditions. The intermediatedosing regimen was able to prevent the pigs from producingantibodies while maintaining the frequency of adverse ef-fects at an acceptable level. All three dose levels resulted ina suppression of the lymphocyte proliferation, which previ-

ously has been reported for cyclosporine-treated pigs (Tsaoet al., 1992). The immunosuppressant cyclosporine isknown to induce nephrotoxicity by causing a vasoconstric-tion in glomerular afferent arterioles (Andreoni et al., 1994).

Fig. 2. Mean (6SD) anti-hGH antibody titer in pigs after administering 0.5 mg hGH/kg once daily for 22 days (n54). Group 1: control group; group 2: 10 mgcyclosporine/kg; group 3: 20 mg cyclosporine/kg; group 4: 40 mg cyclosporine/kg. The immunosuppressing treatment was initiated on day 27.

Table 2Mean (

6

SD) pharmacokinetic parameters obtained on days 1 and 22 after s.c. administration of 0.5 mg hGH/kg/day to pigs for 22 days (

n

5

4, except for group 4 on day 22, where

n

5

2). group 1: Control group; group 2: 10 mg cyclosporine/kg; group 3: 20 mg cyclosporine/kg; group 4: 40 mg cyclosporine/kg.

Day 1 Day 22

Group no. Parameter Unit of measure Mean

6

SD

1 AUC min

?m

g/mL 115

6

21 —AUC(0-t) min

?m

g/mL 106

6

12 173

6

35C

max

mg/mL 0.4

6

0.1 0.4

6

0.1t

max

min 56

6

46 56

6

46t

1/2

min 106

6

37 —MRT min 218

6

58 —MRT(0-t) min 185

6

25 242

6

42 AUC min

?m

g/mL 83

6

8 —AUC(0-t) min

?mg/mL 80 6 4 226 6 9Cmax mg/mL 0.3 6 0.1 0.5 6 0.1tmax min 86 6 78 105 6 90t1/2 min 78 6 44 —MRT min 170 6 59 —MRT(0-t) min 151 6 28 249 6 6

3 AUC min?mg/mL 118 6 19 95 6 11AUC(0-t) min?mg/mL 115 6 18 91 6 85Cmax mg/mL 0.5 6 0.1 0.5 6 0.1tmax min 83 6 45 64 6 43t1/2 min 71 6 26 97 6 25MRT min 161 6 32 166 6 27MRT(0-t) min 151 6 25 146 6 10

4 AUC min?mg/mL 119 6 15 92 6 0.6AUC(0-t) min?mg/mL 116 6 15 87 6 4Cmax mg/mL 0.5 6 0.1 0.4 6 0.1tmax min 53 6 45 60 6 1t1/2 min 78 6 38 103 6 34MRT min 164 6 40 178 6 45MRT(0-t) min 149 6 21 151 6 22

6 H. Agersø et al./J. Pharmacol. Toxicol. 41 (1999) 1–8

Fig. 3. Mean (6SD) hGH concentration in pigs on days 1 and 22 (n54). (a) Group 1 (control). (b) Group 2 (10 mg cyclosporine/kg). (c) Group 3 (20 mgcyclosporine/kg). (d) Group 4 (40 mg cyclosporine/kg). Human growth hormone (0.5 mg/kg) was administered once daily for 22 days. (Figure 3 continues onnext page.)

H. Agersø et al./J. Pharmacol. Toxicol. 41 (1999) 1–8 7

To follow the effect on the kidneys, the serum creatinineconcentration was monitored; the measured values do notindicate any compromised renal function at the selecteddose levels. There was a tendency toward a lower serumcreatinine concentration, which may be related to a slightmalnutrition.

The disposition kinetics of hGH was described in pigs af-ter s.c. administration of hGH by use of noncompartmentalmethods. The pharmacokinetics on day 1 were comparablefor the four groups, indicating that hGH pharmacokineticsdid not seem to be affected by the immunosuppressive treat-ment after 7 days of treatment. Approximately 10 days afterinitiating the treatment the control pigs demonstrated a highlevel of anti-hGH antibody titer. Previous experiments indi-cated that these antibodies interfered with the pharmacoki-netics of hGH in both pigs and rats (unpublished observa-tions). Interference from antibodies makes it very difficultto test new sustained-release formulations of hGH in pigs(or any other species). The increase in anti-hGH antibody ti-ter prolonged the elimination of hGH, resulting in a longerhalf-life and a higher AUC. What caused this prolongedelimination is not clear, but it may be explained by the for-mation of complexes of hGH and anti-hGH antibodies inthe blood, and the elimination of this complex may be rela-tively slow. It is well known that porcine insulin comparedto semisynthetic human insulins increase the degree of im-mune response and the amount of circulating immune com-plexes in diabetic children (Dorchy et al., 1989). It may be acomparable phenomenon observed with hGH in pigs; theassay procedure may then cause the bound hGH to be re-leased and thereby demonstrate artificially high hGH con-centrations observed in nonimmunosuppressed pigs.

This model may be used in future experiments, where itwill be necessary to obtain pharmacokinetic parametersfrom compounds that are antigenic in animals.

5. Conclusion

The formation of anti-hGH antibodies influences thepharmacokinetics of hGH in pigs. The formation of antibod-ies may be prevented by immunosuppression. From thepresent experiment, a dose of 20 mg cyclosporine, 2 mgazatioprine, and 2 mg prednisolone/kg/day administeredp.o. was able to prevent the pigs from producing anti-hGHantibodies without incurring severe side effects. This modelmay by useful in future experiments evaluating sustained-release formulations of hGH, in addition to other com-pounds that may induce an immune response in pigs.

References

[1] Andreoni, K.A., Odonnell, C.P., Burdick, J.F., & Robotham, J.L. (1994).Hepatic and renal blood-flow responses to a clinical dose of intrave-nous cyclosporine in the pig. Immunopharmacology 28, 87–94.

[2] de Muinck Keizer-Schrama, S., & Rikken, B. (1992). Growth hor-mone deficiency: Groeihormoondeficientie: diagnostiek en behandel-ing. Tijdschrift Voor Kindergeneeskunde 60, 147–154.

[3] Dorchy, H., Duchateau, J., Bosson, D., & D’Hooge, D. (1989). Trans-fer from purified porcine insulins to semisynthetic human insulins de-creases insulin antibodies and circulating immune complexes in dia-betic children and adolescents. A two-year follow- up. Diabete Metab15, 107–110.

[4] Emlen, W., & Mannik, M. (1980). Effect of preformed immune com-plexes on the clearance and tissue localization of single-strandedDNA in mice. Clin Exp Immunol 40, 264–272.

[5] Gottschalk, U., Garnett, M.C., Ward, R.K., Maibucher, A., & Kohn-lein, W. (1991). Increased serum stability and prolonged biologicalhalf-life of neocarzinostatin covalently bound to monoclonal antibod-ies. J Antibiot 44, 1148–1154.

[6] Johnson, O.L., Cleland, J.L., Lee, H.J., Charnis, M., Duenas, E., Ja-worowicz, W., Shepard, D., Shahzamani, A., Jones, A.J., & Putney,S.D. (1996). A month-long effect from a single injection of microen-capsulated human growth hormone. Nature Med 2, 795–799.

[7] Johnson, O.L., Jaworowicz, W., Cleland, J.L., Bailey, L., Charnis,M., Duenas, E., Wu, C., Shepard, D., Magil, S., Last, T., Jones, A.J.,

Fig. 3. continued from previous page

8 H. Agersø et al./J. Pharmacol. Toxicol. 41 (1999) 1–8

& Putney, S.D. (1997). The stabilization and encapsulation of humangrowth hormone into biodegradable microspheres. Pharm Res 14,730–735.

[8] Lee, H.J., Riley, G., Johnson, O., Cleland, J.L., Kim, N., Charnis, M.,Bailey, L., Duenas, E., Shahzamani, A., Marian, M., Jones, A.J., & Put-ney, S.D. (1997). In vivo characterization of sustained-release formula-tions of human growth hormone. J Pharmacol Exp Ther 281, 1431–1439.

[9] Rehlaender, B.N., & Cho, M.J. (1998). Antibodies as carrier proteins.Pharm Res 15, 1652–1656.

[10] Tsao, P.W., Ito, S., Wong, P.Y., Radde, I.C., Bryson, S., Young, D.S.,Caspi, J., Diaz, R.J., Martell, M.F., Augustine, J.M., Wilson, G.J., &Coles, J.G. (1992). Pharmacodynamics and pharmacokinetics of cy-closporine-a in the newborn pig. Dev Pharmacol Ther18, 20–25.

[11] Van Haeften, T.W., Bolli, G.B., Dimitriadis, G.D., Gottesman, I.S.,Horwitz, D.L., & Gerich, J.E. (1986). Effect of insulin antibodies andtheir kinetic characteristics on plasma free insulin dynamics in pa-tients with diabetes mellitus. Metabolism 35, 649–656.

plasma concentration of hgh and anti-hgh antibodies after subcutaneous administration of hgh for 3...

Documents