Serum apolipoprotein(a) correlates with growth hormone levels in Chinese patients with acromegaly

ATHEROSCLEROSIS

Atherosclerosis

104

( 1993) I83- I88

Serum apolipoprotein(a) correlates with growth hormone levels in Chinese patients with acromegaly Karen S.L. Lam*a, Richard W.C. Pangb, Edward D. Janusb, Annie W.C. Kung”, Christina C.L. Wanga “Department of Medicine, hClinical Biochemistry Unit, University of Hong Kong, Queen Mary Hospital, Po@iilam Road, Hong Kong

(Received

IO June 1993; revision

received

24 August

1993; accepted

30 August

1993)

Abstract

Untreated acromegaly is associated with an increased cardiovascular morbidity and mortality. The contribution of altered lipid metabolism remains unclear. We investigated the relationship between serum apolipoprotein(a) (ape(a)) and growth hormone (GH) levels in 15 patients with acromegaly before and during treatment with octreotide, a longacting somatostatin analogue, 288-600 &day s.c., for 6 months. Before treatment serum ape(a) was significantly elevated in acromegalic patients (geometric mean being 323 U/l vs. 142 U/l in controls (n = 92; P < 0.01)). Octreotide treatment resulted in significant reductions in serum ape(a) concentration (F= 7.22; P < 0.01; geometric mean being 232 U/I and 248 U/l at 3 months and 6 months respectively) and ape(a) concentrations on treatment were not significantly different from control values. There were significant reductions in serum GH (F = 7.30; P < 0.01) insulin growth factor 1 (IGFl) (F= 31.4, P < 0.001) and insulin (F= 4.57; P < 0.05) concentrations. Plasma glycosylated haemoglobin levels were unchanged. Ape(a) levels correlated with serum GH (r = 0.450; P < 0.01) but showed no correlation with basal insulin concentrations. Serum HDL cholesterol increased on treatment (F= 4.29; P < 0.05). Triglycerides were reduced only in the 12 patients without diabetes mellitus (F= 4.75; P < 0.05). No significant change in LDL cholesterol occurred. Our findings suggest that ape(a) may constitute another cardiovascular risk factor in untreated acromegaly and that GH may be involved in the regulation of circulating ape(a) concentration. Key words: Apolipoprotein(a);

Lipoproteins;

Growth hormone; Acromegaly; Octreotide

1. Introduction

(ape(a)), covalently linked to apolipoprotein B of low density lipoproteins (LDLs) [2,3]. It has been

Lipoprotein(a) (Lp(a)) is a macromolecular complex with structural similarity to plasminogen [l]. It consists of a glycoprotein, apolipoprotein(a)

recognized as an independent risk factor for premature atherosclerosis [2-41 and has atherogenic [2,4] and in vitro antifibrinolytic properties [5]. Studies in Chinese populations have also shown that Lp(a) is a risk factor for cerebral [6] and coronary atherosclerosis [7].

* Corresponding author,

Tel.: (852) 855 4783; Fax: (852) 872

5828.

002 l-91 50/93/$06.00 0 I993 Elsevier Ireland Ltd. All rights reserved.

Scientific

Publishers

184

Untreated acromegaly is associated with an increased cardiovascular morbidity and mortality [8]. The known risk factors include an increased incidence of diabetes mellitus, hypertension and hypertriglyceridaemia [8,9]. Hyperinsulinism [9] and reduction of post-heparin lipase activity [lo] are probably also involved in the,development of hyperlipidaemia in these patients. Recently, growth hormone (GH) treatment of GH-deficient adults was reported to result in a marked increase in circulating Lp(a) [ 111, suggesting that GH may be involved in the regulation of Lp(a) concentrations and that an increase in Lp(a) may constitute another atherogenic risk factor in patients with GH excess. Octreotide is a long-acting somatostatin analogue which is effective in suppressing GH and normalizing insulin growth factor 1 (IGFl) levels in acromegalic patients [12]. In this study, we evaluated prospectively the changes in Lp(a) and other serum lipid levels in 15 Chinese patients with acromegaly during a 6-month course of octreotide therapy. 2. Methods 2.‘1. Subjects Fifteen patients (age 38.2 f 9.2 years, mean f S.D.; 7 female, 8 male) with active acromegaly were studied. The diagnosis was made clinically and confirmed by elevated basal serum GH levels and nonsuppressibility of serum GH to less than 5 mu/l during a 75 g oral glucose tolerance test. Three patients had diabetes mellitus requiring drug treatment and three had impaired glucose tolerance. Two patients were on testosterone replacement and two were on thyroxine replacement for hypopituitarism. There was no change in thyroxine or testosterone dosage during the study period. One woman was postmenopausal and another two had secondary amenorrhoea. No oestrogen replacement was given. 2.2. Study design All subjects were studied after informed consent according to a protocol approved by the local ethics committee. They received treatment with octreotide for 6 months, given subcutaneously as

K. S. L. Ltm ei trl. / Athrro.sc~i~~,osi.v104 ( 1993) 1X3- 188

either 2-hourly injections (10 patients; total daily dose 288-504 pg) or S-hourly injections (total dose 600 pg/day) as previously described [ 121. Biochemical assessment of the GH secretory status (hourly estimations for 13 h) and the measurement of IGFl, glycosylated haemoglobin (HbAl) and fasting lipids including Lp(a) were performed before treatment and repeated at 3-month intervals during treatment. In 10 patients, none of them on insulin, changes in basal insulin levels were also assessed. In 6 patients mean GH, IGFl and ape(a) levels were subsequently measured following transsphenoidal hypophysectomy, more than 4 weeks after stopping octreotide. Lipid and insulin levels were measured in the morning after an overnight fast. The fasting lipid levels were compared to those of 36 apparently healthy controls (12 female, 24 male; age 27.8 f 12.6 years, P < 0.01 vs. patients) recruited from the hospital staff. In addition, random ape(a) levels of 56 healthy police recruits (age 21.3 f 5.1 years; 27 female, 29 male) were also included as control values for ape(a). None of the control subjects were on any drug treatment. All control data on ape(a) and other serum lipids were measured at a single time point only. 2.3. Assays Serum GH was measured by radioimmunoassay (RIA) using reagents supplied by the National Pituitary Agency (NIDDK). IGFl was measured using a commercial RIA kit (Incstar, Stillwar, MN) with serum extraction prior to assay. Insulin was measured with an RIA kit from Amersham (Arlington Heights, IL). Serum ape(a) was measured by immunoradiometric assay (Pharmacia, Uppsala, Sweden), the intra-assay variations at low and high ape(a) levels being 2.6% and 1.4”/0respectively. Total cholesterol and triglyceride were determined enzymatically (Boehringer, Mannheim, Germany) on a Hitachi 717 analyser. HDL cholesterol was measured by the same method after precipitation of very low density lipoprotein (VLDL) and LDL with polyethylene glycol. LDL cholesterol was calculated according to the Friedewald equation. HbAl was measured using agar-gel electrophoresis (Corning Medical and Scientific, Medtield, MA). For the measurement of

185

K.S.L. Lum et (I/. / A~hrrosclerosi.~ 104 (1993) 183-188

3 months and 6 months (geometric mean 232 U/l and 248 U/l respectively) were not significantly different from those observed at baseline in controls. This was accompanied by significant reductions in mean serum GH (mean 13-h GH level; F = 7.30; P < 0.01) IGFl (F= 31.4, P < 0.001) and fasting insulin (F= 4.57; P < 0.05) levels (Fig. 1). There was no significant change in HbAl (Table 2). following levels also decreased Ape(a) transsphenoidal hypophysectomy (geometric mean 373 U/l vs. 543 U/l before octreotide therapy, P < 0.05, n = 6) accompanied by corresponding changes in serum GH (51.4 f 30.5 mUi1 vs. 241 f 108 mUi1 before octreotide, P > 0.05) and IGFl (1.92 * 0.29 mu/l vs. pre-treatment 3.48 f 0.40 mu/l, P < 0.05). In the 10 patients who had serial measurement of fasting insulin levels during octreotide therapy, the relationship between ape(a) and hormonal changes was assessed. When the data before and during octreotide treatment were pooled for analysis, ape(a) showed a positive correlation with mean GH levels (r = 0.450, P < 0.01, n = 30) but no correlation with fasting insulin or IGFl. Even when the data from all 15 patients were analysed, no significant correlation could be demonstrated between ape(a) and IGFl (r = 0.283; P > 0.05; n = 45) while the positive correlation between ape(a) and mean GH remained unchanged (r = 0.401; P < 0.01; n = 45). IGFl levels correlated significantly with both GH (r = 0.756, P < 0.001) and fasting insulin (r = 0.422, P < 0.05). The serum lipids levels before and during octreotide therapy are summarised in Table 2. Treat-

IGFl, insulin, ape(a) and other lipid levels, all samples of each patient were included in the same assay. All samples were stored at -20°C until assay. Samples for serum lipids had been previously thawed and refrozen.

2.4. Statistics Statistical analyses were performed using the MINITAB (Minitab Inc., State College, PA) statistics package. Because of their skewed distribution, ape(a) levels were normalized by logarithmic transformation prior to analysis. Correlations between variables were tested by the Pearson correlation test. Two-way analysis of variance was used to evaluate the serial changes during treatment. Student’s t-test was used as appropriate. 3. Results The serum ape(a) and lipid levels in controls and untreated patients are shown in Table 1. Ape(a) levels in the control population (including 56 healthy police recruits) showed a skewed distribution with a range of 17- 1613 U/l and a geometric mean of 142 U/l. Before treatment, ape(a) concentrations in the acromegalic patients were significantly elevated compared to controls (P < 0.01; range 90- 1164 U/l; geometric mean 323 U/l). Triglyceride levels were also higher (P < O.OOOl),while HDL cholesterol levels were lower than controls (P < 0.001). Treatment with octreotide resulted in a significant reduction in mean serum ape(a) concentration (Fig. 1; F = 7.22; P < 0.01) so that values at

Table

1

Serum ape(a) and fasting lipid levels in controls and untreated

acromegalic

patients

APO(a)

Triglyceride

HDL

LDL

W/l)

(mmol/l)

cholesterol

cholesterol

(mmolil)

(mmol/l)

Patients (n = 15)

323’

2.04 * 0.31***

0.65 zt 0.05**

3.58 * 0.31

Controls (n = 36)”

142

0.93 f

0.92 f

3.33 * 0.14

Mean f a

SE.

For ape(a)

except for ape(a)

(geometric

0.06

mean).

only n = 92.

“P < 0.01; **p

< 0.001; ***P

< 0.0001 vs. control.

0.05

186

I

01

BEFORE

I

3 MONTHS

1

6 MONTHS

120

I

&FORE

I 3 MONTHS

I 6 MOh ITI

Fig. I. Changes in circulating ape(a) (top) and hormonal (bottom) levels before and during octreotide treatment. *P < U.05, **P < 0.01. ***P < 0.001. vs. before treatment; n = 15 except for insulin (n = IO). Data are expressed as mean f S.E.M. except for ape(a) (-. geometric mean).

ment with octreotide was associated with a significant increase in HDL cholesterol (F = 4.29, P < 0.05). There was no significant change in LDL cholesterol. A significant reduction in triglyceride concentration was demonstrated only after excluding the three patients with diabetes mellitus (F = 4.75, P < 0.05). Fasting triglyceride

in the non-diabetic patients decreased from 1.70 f 0.19 mmol/l pre-treatment to 1.12 f 0.12 mmol/l at 3 months (P < 0.05) and 1.33 f 0.13 mmol/l at 6 months (P > 0.05). A negative correlation was found between serum triglyceride and HDL cholesterol concentrations (r = -0.425, n = 45, P < 0.01). Neither serum triglyceride nor

K. S. L. Lam ef ul. / Alhl,ro.sc,ii,rosis 104 IIYY3)

Table 2 Changes in metabolic

Before 3 months 6 months

parameters

IX7

lN3-IN8

before and during

octreotide

treatment

Triglyceride (mmolil)

HDL cholesterol (mmol/l)

LDL cholesterol (mmol/l)

HbA I (I%,)

2.04 f 0.31 1.74 f 0.44 2.32 * 0.79

0.65 zt 0.05 0.79 f 0.06* 0.75 f 0.07

3.58 +z 0.31 3.37 f 0.20 3.19 f 0.25

7.8 ?? 0.4 7.7 f 0.9 7.7 f 0.9

Normal range: HbA I, < 8%. Mean f S.E.M.; n = 15. *P < 0.05 vs. before treatment,

paired

I-test.

HDL cholesterol showed any correlation with fasting insulin concentrations. 4. Discussion We have shown in this study an increase in ape(a) concentration in patients with active acromegaly which was reduced when the excessive GH secretion was ameliorated following octreotide therapy or transsphenoidal surgery. A positive correlation was demonstrated between serum ape(a) and mean GH concentrations, giving further support for the involvement of GH in the regulation of ape(a) concentrations in these patients, This, together with the absence of a demonstrable correlation between serum ape(a) and IGFl levels, suggests that GH regulates ape(a) concentration via a direct cellular action, such as by regulating hepatic apolipoprotein synthesis. Since an inverse correlation is known to exist between ape(a) size and plasma Lp(a) concentration [ 131,it is also possible that the high level of ape(a) in patients with acromegaly may be explained by an over-representation of the low molecular weight ape(a) phenotypes among the patients vs. the control population. The reduction in ape(a) levels following octreotide or surgery would then suggest an effect of GH on the phenotypic expression of ape(a). This possibility should be addressed by studying the effect of GH-lowering therapy on ape(a) phenotype in these patients. The increase in serum triglycerides [9] in acromegaly and their normalization following octreotide therapy has been previously reported [14].

Our data also agree with those of Cohen et al. [ 141 which suggest that diabetes mellitus is associated with a lack of triglyceride normalization during octreotide therapy. An 8% elevation in HDL cholesterol, which failed to reach statistical significance, was also found in the same study [14]. The reduced HDL cholesterol levels and their increase during octreotide therapy were not unexpected in view of the inverse relationship between serum triglyceride and HDL cholesterol concentrations, as demonstrated in this study. Our data also suggest that the changes in ape(a) and serum lipids during octreotide treatment resulted from the changes in GH secretory status and were independent of changes in insulin secretion or glycaemic control. The levels of HDL cholesterol in the present study were relatively low compared to those reported in previous studies among Hong Kong Chinese [6,15]. This discrepancy may, in part, be attributed to differences in methodology and sex distribution of the subjects studied. Whether the use of stored sera which had been previously thawed and refrozen also contributed to this discrepancy remains to be ascertained. It is well known that Lp(a) levels are under strong genetic influence [7]. Recent studies suggest that other factors such as gonadal steroids [16], thyroid hormones [17] and growth hormone [l l] also participate in the regulation of Lp(a) concentration. Our findings provide further support to the involvement GH in Lp(a) regulation and suggest that the increased cardiovascular morbidity and mortality in patients with acromegaly may, at least in part, be attributable to an increase in circulating Lp(a) concentration.

K. S. L. Ltntr e/ t/l. / Arl~c~ro.vi~l~~,o.si.v 104 i IYY3) 1X3- 1X8

188

5. Acknowledgements

We thank the staff of the Metabolic Ward, Queen Mary Hospital, Hong Kong for carrying out the hormonal and metabolic investigations, Mr. D. Robinson of the Clinical Biochemistry Unit for assistance in obtaining samples from healthy controls, Mr. K.S. Lau for technical assistance, MS Venus Yuen for secretarial assistance and Sandoz, Hong Kong for their generous sup‘port for this study.

8 9 IO

II

12

6. References McLean, J.W., Tomlinson, J.T.. Kuang. W.J. cDNA sequence of human apolipoprotein homologous to plasminogen, Nature, 300 (1987) Berg, K.. Genetics of atherogenic lipoprotein (a) relation to other atherosclerotic risk factors, Med., IS (1991) 209. Utermann, G., The mysteries of lipoprotein 246 (1989) 904.

et al., (a) is 132. and its

13

Ann. Intern. 14 (a), Science,

the role of the ape(a) gene in coronary heart disease, J. Clin. Invest., 89 (1992) 1040. Nabarro. J.D.N., Acromegaly - a review, Clin. Endocrinol., 26 (1987) 481. Nikkila, E.A. and Pelkonen, R., Serum lipids in acromegaly, Metabolism, 24 (1975) 829. Murase, T., Yamada, N., Ohsawa, K.. Kosaka, K. and Morita, S., Decline of post-heparin plasma lipoprotein lipase in acromegalic patients. Metabolism. 29 (1980) 666. Eden, S., Wiklund, O., Oscarsson, J. and Rosen. T.. Treatment of growth-hormone-deficient adults results in a marked increase in Lp(a) and HDL-cholesterol concentrations. Arteriosclerosis Thromb.. 13 (1993) 296. Wang, C., Lam, K.S.L., Arceo, E., and Chan, F.L., Comparison of the effectiveness of 2-hourly vs. 8-hourly subcutaneous injections of a somatostatin analog (SMS 201-995) in the treatment of acromegaly, J. Clin. Endocrinol. Metab., 69 (1989) 670. Utermann, G., Menzel, H.J., Kraft, H.G.. Duba. H.C., Kemmler, H.G. and Seitz. C., Lp(a) glycoprotein phenotypes: inheritance and relation to Ip(a)-lipoprotein concentrations in plasma, J. Clin. Invest.. 80 (1987) 458. Cohen, R., Chanson, P., Bruckert, E. et al., Effects of octreotide on lipid metabolism in acromegaly, Horm. Metab. Res., 24 (1992) 397.

Rhoads, C.G., Dahlen. G.. Berg, K.. Morton, N.E. and Dannenberg, A.L., Lp(a) lipoprotein as a risk factor for myocardial infarction, J. Am. Med. Assoc., 256 (1986) 2540.

15

Lam, K.S.L., Chat-r, M.K. and Yeung, R.T.T. High density lipoprotein cholesterol, hepatic lipase and lipoprotein lipase activities in thyroid dysfunction - effects of treatment, Q. J. Med., 59 (1986) 513.

Hajjar. K.A.. Gavish, D., Breslow. J.L. and Nachman, R.L., Lipoprotein (a) modulation of endothelial tell surface tibrinolysis and its potential role in atherosclerosis, Nature. 339 (1989) 303.

16

Woo, J., Lau, E., Lam, C.W.K. et al., Hypertension, lipoprotein (a), and apolipoprotein A-l as risk factors for stroke in the Chinese, Stroke, 22 (1991) 203. Sandholzer. C.H., Boerwinkle, E.. Saha, N., Tong, M.C. and Utermann, G.. Apolipoprotein (a) phenotypes, Lp(a) concentration and plasma lipid levels in relation to coronary heart disease in a Chinese population: evidence for

17

Henriksson, P., Angelin, B. and Berglund, L.. Hormonal regulation of serum Lp(a) levels: Opposite effects after oestrogen treatment and orchidectomy in males with prostatic carcinoma, J. Clin. Invest., 89 (1992) 1161. De Bruin, T.W.A., van Barlingen, H., van Linde-Sibenius Trip, M., van Vuurst de Vries, A.-R., Akveld, M.J. and Erkelens, W., Lipoprotein (a) and apolipoprotein B plasma concentrations in hypothyroid, euthyroid and hyperthyroid subjects, J. Clin. Endocrinol. Metab., 76 (1993) 121.