Acta Padiatr Scand [Suppl] 372: 47-52, 1991
Management of Growth Hormone Deficiency Through Puberty R. STANHOPE', M. URUENA', P. HINDMARSH*, A.D. LEIPER' and C.G.D. BROOK2 From the Kabi International Growth Research Centre, 'Institute of Child Health, and Middlesex Hospital, London, UK
ABSTRACT. Stanhope R.', Urena M.l, Hindmarsh P.2, Leiper A.D.' and Brook C.G.D.Z (Kabi International Growth Research Centre, 'Institute of Child Health, and 2Middlesex Hospital, London, UK). Management of growth hormone deficiency through puberty. Acta Paediatr Scand [Suppl] 372: 47-52, 1991. As a model of the growth hormone (GH) dependence of growth in prepuberty and puberty, the growth of 182 children (93 boys, 89 girls) who survived in first remission for treatment of acute lymphoblastic leukaemia was examined. Chemotherapy regimens, including intrathecal methotrexate, were similar in all patients, hut CNS treatment differed, in that one group received 2400 cGy cranial irradiation, while the other received 1800 cGy. There was a significant decrease in height SDS during prepuberty, which was equivalent in both sexes, whereas there was a much greater decrease in pubertal growth in girls than in boys. Girls treated with the lower dose regimen of cranial irradiation had their onset of pubertal maturation significantly advanced, to a mean of 9.9 years ( p < 0.001). Previous studies have indicated that the duration of puberty is shortened by GH treatment in patients with idiopathic multiple pituitary hormone deficiency or isolated GH deficiency (GHD). To determine whether an increase in the dose of GH administered during the adolescent growth spurt would improve final height, a prospective randomized trial was performed in 32 children (25 boys, 7 girls) with isolated GHD treated with a GH dose regimen of 15 IU/m2/week as daily S.C. injections. At the onset of the pubertal growth spurt, the patients were randomized either to an unchanged dose or to 30 IU/m*/week. There was no significant change in height velocity with the doubled dose of GH, but there was a trend in the advancement of pubertal maturation which was considered to be dose related. It is suggested that these findings are of relevance to the treatment of GHD in puberty, especially in girls with early or precocious puberty occurring as a consequence of low-dose cranial irradiation. It is concluded that optimum final heights may not be achieved in these patients without the therapeutic manipulation of the onset and/or duration of puberty. Key words: Isolated growth hormone deficiency, precocious puberty, duraiion of puberty, growth hormone dose.
The management of patients with idiopathic growth hormone deficiency (GHD) during puberty has recently been reviewed by Price et al. (1). Both sex steroids and growth hormone (GH) are required in puberty for an adequate growth spurt (2); however, little is known of the optimum dose regimens. Previous management regimens would now be considered inadequate, as they used suboptimal doses (3) and dose frequencies (4).GH administration more frequently than once a day may be advantageous (5). During the growth spurt of spontaneous puberty (Delemarre-van de Wad, pp. 26-31) and puberty induced by pulsatile therapy with gonadotrophin-releasing hormone (GnRH) (6), there is an increase in endogenous secretion of GH, which is coincident with the onset of the growth spurt. Indeed, other authors (7) have effectively doubled the dose of GH during puberty, because it was assumed that this was an appropriate therapeutic manoeuvre. For several years, reproductive endocrinologists have appreciated that GH has an important role both in gonadal development and in the maintenance of sexual maturation. GH acts on the gonads (8, 9), either by inducing local production of insulin-like growth factors (IGFs) or by a direct action, and augments gonadotrophin-modulated gonadal function (10, 11). GH treatment of female rhesus monkeys affects the rate of pubertal maturation without altering the onset of puberty (12). Such data has been used with the preliminary findings that GH may facilitate the induction of ovulation by gonadotrophins in human females (13).
48
R. Stanhope et al.
Acta Paedidtr Scand [suppl] 372
We have hypothesized that GH treatment may affect pubertal maturation, and that this may have important consequences for the treatment both of children with idiopathic GHD who have spontaneous puberty, and of children who have early o r precocious puberty in association with GH insufficiency resulting from low-dose cranial irradiation (14). W e have carried out a prospective study to determine whether a change in GH dose at the onset of the adolescent growth spurt would affect pubertal maturation, growth rate and, ultimately, final height in patients with idiopathic isolated GHD.
PATIENTS AND METHODS GHdependence of growth inprepuberry andpuberty. The growth of 182 children (93 boys, 89 girls) who were long-term survivors of acute lymphoblastic leukaemia in first remission was retrospectively examined. None had received spinal or gonadal irradiation and all had been treated with similar chemotherapeutic regimens, including intrathecal methotrexate. Ninety-three patients (5 1 boys, 42 girls; mean age at diagnosis 4.8 years) were treated with 2400 cGy cranial irradiation and 89 patients (42 boys, 47 girls; mean age at diagnosis 6.5 years) were treated with 1800 cGy cranial irradiation. None of the patients had received endocrine therapy. E’ect of GH dose during puberty. Thirty-two children (25 boys, 7 girls) with idiopathic isolated GHD were included in a prospective study in order to determine whether a change in GH dose during the adolescent growth spurt would improve final height. In late prepuberty all the patients were treated with a GH dose regimen of 15 IU/m*/week, administered as daily S.C.injections. At the onset of the growth spurt (equivalent to B2 in the girls and 8 ml testicular volume in the boys) the patients were randomized into two groups; one group continued with an identical dose regimen, while the other received 30 IU/m2/week with no change in frequency of administration. Girls commence their growth spurt with the onset of breast development, often before they have been shown to have acquired stage B2. However, for practical reasons, the acquistion of stage B2 was taken as the time for randomization of the GH dose regimen.
RESULTS
GH dependence of growth in prepuberty and puberty. The growth data for children surviving in first remission from acute lymphoblastic leukaemia showed a similar decrease in mean height SDS in prepuberty in both boys and girls (Fig. I), but there was a marked sex (a)
2400cGy
m Boys i X l Girls
I
-2L (b)
1800cGy
+2r -2L
-
-
Diagnosis
Onset of
Final
puberty
height
Fig. 1. Growth data from 182 children treated for acute lymphoblastic leukaemia with (a) 2400 cGy or (b) 1800 cGy of cranial irradiation. Height SDS values at diagnosis, at the onset of puberty, and at final height are shown. Standard deviations are shown by the horizontal bars.
Acta
Management of G f f deficiency through puberty
Paedzatr Scand [Suppl] 372
n
=
n = 10
15
I
h
2
12-
E-
10-
%
,
x
c .-
8 al
a-
>
E
0) ._
al
I
64-
2OL -0.5
I
I
0
0.5
I
I
I
1
1.5
2
1
-0.5
I
I
I
I
I
0
0.5
1
1.5
2
Years from 8 ml testicular volume Fig. 2. Two-year height velocity data for 25 boys with isolated idiopathic GHD who were randomized into either (a) an unchanged GH regimen of 15 IU/m2/week (n = 15) or (b) an increased dose of 30 IU/m2/week (n = 10) at onset of the growth spurt (8 ml testicular volume). GH was administered as daily S.C. injections.
difference during puberty. The boys had an almost equal decrease in increment of height SDS during prepuberty and puberty, while the girls had a much greater decrease in increment of height SDS during puberty than in the prepubertal period. The mean age of onset of puberty in the boys was not significantly different from normal boys (15), but there was a marked decrease in the mean age of onset of puberty in the girls. In the lower dose regimen of cranial irradiation, girls had a significant reduction in the mean age of onset of pubertal maturation, to 9.9 years ( p < 0.001), compared with normal girls (16). This decrease in the mean age of onset of pubertal maturation was less marked in the girls receiving high-dose irradiation (mean age of onset 10.6 years). The early onset of puberty in such girls with GH insufficiency has important implications for their therapy, which will be discussed below. Effect of GH dose during puberty. Height velocity at randomization was higher in the boys who were allocated to the higher dose GH regimen, but there was no significant difference in height velocity between the two groups over the first 2 years of this study ( p > 0.5) (Fig. 2). Mean height velocity at randomization for the boys who received an unaltered dose SD) and after 2 years was 8.6 2.6 cm/year. of GH was 6.9 f 1.5 cm/year (mean In the group who received the higher dose regimen, height velocity was 7.5 f 1.5 cm/year at randomization and 9.0 f 2.5 cm/year after 2 years. The response between individuals was heterogenous (Fig. 2). Data for the girls indicated a similar result, with no significant change in growth rate between the two treatment groups. However, the number of girls in this study was small (n = 7). Although the dose of GH had no significant effect on growth rate, it did influence the pattern of pubertal maturation. The mean age of onset of puberty in the boys was 12.0 years in the group with an unchanged dose regimen and 12.3 years in the group who received the higher dose of GH. In the group who received the higher dose, pubertal onset was delayed (although not significantly) compared with Tanner’s normal data (15). The rate of progress
*
*
49
50
R. Stanhope et al.
'-r,
G5[
10
11
12
13
14
15
Acta Paediatr Scand [Suppll 372
+ Boys on an unaltered
dose regimen -+
f33;:s;;;oubled
16
Age (years)
Fig. 3. Progress through puberty for the same boys as described in Fig. 2. The continuous line represents the mean age for the onset of each stage of puberty in normal boys (15). The vertical bars represent 1 standard deviation.
of puberty did not differ from normal in the group who received the low dose regimen (Fig. 3 ) . However, in the boys who received the higher dose, there was a faster progression through puberty (Fig. 3 ) . Thus, the higher dose regimen shortened the duration of puberty without altering the rate of growth, suggesting that the final height prognosis would not be improved, and indeed might even be decreased. The results for girls are not shown, as the numbers were small, but the trend for no improvement in growth rate and a faster progression through puberty with the higher dose regimen was similar to that in the boys. DISCUSSION Our data indicate an important relationship between GH secretion and the duration of puberty, which is not surprising considering the intimate relationship between GH and gonadal function (10). In order to achieve maximum height potential it would theoretically be advantageous to sustain a higher rate of growth, using high doses of exogenous GH, and at the same time slowing the rate of pubertal maturation, and thereby epiphyseal maturation, to allow a longer period for growth. However, the two factors are often interrelated, so that although low-dose cranial irradiation produces a lower incidence of GH insufficiency (17), such low-dose regimens are associated with early or precocious puberty in the girls (14), which is counterproductive. The boys, who did not have early sexual maturation, had a smaller decrease in height SDS during puberty than the girls. We hypothesize that using the 1800 cGy regimen, girls have a worse growth prognosis than boys because of the high incidence of early sexual maturation. The results of our studies of variable GH dose during puberty are preliminary, and larger numbers of patients will be required, with measurements taken to final height, before a definite conclusion can be drawn. Despite the shortcomings of our prospective study, the data suggest that an increase in the dose of GH during the pubertal growth spurt of children with idiopathic GHD may be counterproductive. By not inducing a significant increase in growth rate, while reducing the available time for pubertal growth, final height attainment may be unaltered or even reduced. Certainly GH treatment reduced the duration of puberty, and this may well be a dose-related phenomenon. With this available evidence, it seems unlikely that a change in dose of GH treatment would lead to an improved
Acta Paediatr Scand [Suppl] 372
Management of GH dejciency through puberty
final height attainment. We are unable to comment about whether this may be achievable by an alteration in dose frequency. Van der Werff ten Bosch and Bot (18) have recently described the growth of boys with multiple pituitary deficiencies prior to, and after, the availability of GH treatment. There are considerable difficulties in the interpretation of such data, because it spans a long period of time with no allowance for secular change, and also encompasses changes in treatment regimen. Twenty-four boys had pituitary hormone replacement, including androgen therapy, to induce puberty. Fourteen of the boys received GH treatment, whilst the other 10 did not. The patients who received GH treatment had a faster short-term growth rate, but progressed through ‘puberty’ in 5.5 years, compared with 9.5 years in the group who did not receive GH therapy. Because of the possibility that these results could have been an artefact of the study groups and that pubertal maturation was not spontaneous, our unit has analysed the duration of puberty in 37 girls and 97 boys with isolated GHD, during GH therapy (19). Although the timing of the onset of pubertal maturation in this group was delayed, the duration of puberty was shortened. The mean ages at onset of puberty were 12.1 years in the girls and 13.0 years in the boys, which represent significant delays compared with the values for normal children of 11.2 and 12.0 years, respectively, according to Tanner (16, 15). The mean duration of puberty (genitalia/breast stage 2 to 4) was 1.5 years in both sexes, which was considerably shorter than the corresponding normal values in girls (16) and boys (15). The dose of GH during this study was 4 IU intramuscularly three times a week, irrespective of body size, which makes the data more difficult to interpret in relation to current GH dose schedules. A more promising therapeutic approach may be to use a GnRH analogue to delay the onset, andlor increase the duration of pubertal maturation, thus allowing a longer period for pubertal growth. Certainly, the use of a GnRH analogue alone in children with GHD (20) or in girls with central precocious puberty (2 1) does not improve final height prognosis. However, the use of a GnRH analogue combined with GH treatment would certainly appear to be an appropriate therapeutic manoeuvre. This is true, particularly for the treatment of girls with acute lymphoblastic leukaemia with early sexual development. Such girls have a dual endocrinopathy, associated with low-dose cranial irradiation, GH insufficiency and earlylprecocious puberty. If treated with GH alone, the earlier onset of sexual maturation may be compounded by a shortened duration of pubertal maturation. The use of a combined regimen of GH and GnRH analogue has been reported (22), and the initial results of improvement in height prognosis are promising. It seems likely that a joint regimen of GH and GnRH analogue will theoretically yield improved results in terms of final heights of children with isolated GHD, but the optimal dose of GH to use during puberty will depend on the results of controlled prospective studies with the end point being final height attainment.
REFERENCES 1. Price DA, Shalet SM, Clayton PE. Management of idiopathic growth hormone deficient patients during puberty. Acta Paediatr Scand [Suppl] 1988; 347: 44-51.
2. Aynsley-Green A, Zachmann M, Prader A. Interrelation of the therapeutic effects of growth hormone and testosterone on growth in hypopituitarism. J Pediatr 1976; 89: 992-9. 3. Brook CGD, Hindmarsh PC, Smith PJ. The management of short stature. In: Brook CGD, ed. Clinical paediatric endocrinology. 2nd ed., Oxford: Blackwell Scientific Publications 1989; 118-27. 4. Albertsson-Wikland K. The effect of human growth hormone injection frequency on linear growth rate. Acta Paediatr Scand [Suppl] 1987; 337: 110-16. 5 . Hindmarsh PC, Stanhope R, Preece MA, Brook CGD. Frequency of administration of growth hormone - an important factor in determining growth response to exogenous growth hormone. Horm Res [Suppl 41 1990; 33: 83-9. 6. Stanhope R, Pringle PJ, Brook CGD. The mechanism of the adolescent growth spurt induced by low dose pulsatile GnRH treatment. Clin Endocrinol 1988; 28: 83-91.
51
52
R. Stanhope et al.
Acta
Paediatr Scand [Suppl] 372
7. Vanderschueren-Lodeweyckx M, van den Broeck J, Wolter R, Malvaux P. Early initiation of growth hormone treatment: influence on final height. Acta Paediatr Scand [Suppl] 1987; 337: 4-11. 8. Davoren JB, Hsueh JW. Growth hormone increases ovarian levels of immunoreactive somatomedin C/ insulin-like growth factor 1 in vivo. Endocrinology 1986; 118: 888-90. 9. Tres LL, Smith E, van Wyk JJ, et al. Somatomedin-C in rat sertoli-spermatogenic cell cultures. J Cell Biol 1983; 97: 18a. 10. Adashi EY, Resnick CE, D’Ercole AJ, et al. Insulin-like growth factors as intraovarian regulators of granulosa cell growth and function. Endocr Rev 1985; 6: 400-19. 1 1 . Mason HD, Martikainen H, Beard RW, Anyaoku V, Franks S . Direct gonadotrophic effect of growth hormone on oestradiol production by human granulosa cells in vitro. J Endocrinol 1990; 126: R1-4. 12. Wilson ME, Gordon TP, Rudman CG, Tanner JM. Effects of growth hormone on the temporosexual maturation in female rhesus monkeys. J Clin Endocrinol Metab 1989; 68: 29-38. 13. Homburg R, Eshel A, Abdalla HI, et al. Growth hormone facilitates ovulation induction by gonadotrophins. Clin Endocrinol 1988; 29: 113-17. 14. Leiper AD, Stanhope R, Kitching P, Chessells JM. Precocious and premature puberty associated with treatment of acute lymphoblastic leukaemia. Arch Dis Child 1987; 62: 1107-12. 15. Marshall WA, Tanner JM. Variations in the patterns of pubertal changes in boys. Arch Dis Child 1970; 45: 13-23. 16. Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in girls. Arch Dis Child 1969; 44: 291-303. 17. Shalet SM. Irradiation induced growth failure. Clin Endocrinol Metab 1986; 15: 591-606. 18. Van der Werff ten Bosch JJ, Bot A. Does hGH treatment promote adult height of hypopituitary children? Neth J Med 1988; 32: 217-25. 19. Darendeliler F, Hindmarsh PC, Preece MA, Cox L, Brook CGD. Growth hormone increases rate of pubertal maturation. Acta Endocrinol (Copenh) 1990; 122: 414-16. 20. Stanhope R, Brook CGD. The effect of gonadotrophin releasing hormone analogue on height prognosis in growth hormone deficiency and normal puberty. Eur J Pediatr 1988; 148: 200-2. 21. Stanhope R, Pringle PJ, Brook CGD. Growth, growth hormone and sex steroid secretion in girls with central precocious puberty treated with a GnRH analogue. Acta Paediatr Scand 1988; 77: 525-30. 22. Leiper AD, Stanhope R, Preece MA, Grant DB, Chessells JM. Precocious and early puberty and growth failure in girls treated for acute lymphoblastic leukaemia. Horm Res 1988; 30: 72-6. (R.S.) The Institute of Child Health 30 Guilford Street London WClN IEH UK
Management of Growth Hormone Deficiency Through Puberty R. STANHOPE', M. URUENA', P. HINDMARSH*, A.D. LEIPER' and C.G.D. BROOK2 From the Kabi International Growth Research Centre, 'Institute of Child Health, and Middlesex Hospital, London, UK
ABSTRACT. Stanhope R.', Urena M.l, Hindmarsh P.2, Leiper A.D.' and Brook C.G.D.Z (Kabi International Growth Research Centre, 'Institute of Child Health, and 2Middlesex Hospital, London, UK). Management of growth hormone deficiency through puberty. Acta Paediatr Scand [Suppl] 372: 47-52, 1991. As a model of the growth hormone (GH) dependence of growth in prepuberty and puberty, the growth of 182 children (93 boys, 89 girls) who survived in first remission for treatment of acute lymphoblastic leukaemia was examined. Chemotherapy regimens, including intrathecal methotrexate, were similar in all patients, hut CNS treatment differed, in that one group received 2400 cGy cranial irradiation, while the other received 1800 cGy. There was a significant decrease in height SDS during prepuberty, which was equivalent in both sexes, whereas there was a much greater decrease in pubertal growth in girls than in boys. Girls treated with the lower dose regimen of cranial irradiation had their onset of pubertal maturation significantly advanced, to a mean of 9.9 years ( p < 0.001). Previous studies have indicated that the duration of puberty is shortened by GH treatment in patients with idiopathic multiple pituitary hormone deficiency or isolated GH deficiency (GHD). To determine whether an increase in the dose of GH administered during the adolescent growth spurt would improve final height, a prospective randomized trial was performed in 32 children (25 boys, 7 girls) with isolated GHD treated with a GH dose regimen of 15 IU/m2/week as daily S.C. injections. At the onset of the pubertal growth spurt, the patients were randomized either to an unchanged dose or to 30 IU/m*/week. There was no significant change in height velocity with the doubled dose of GH, but there was a trend in the advancement of pubertal maturation which was considered to be dose related. It is suggested that these findings are of relevance to the treatment of GHD in puberty, especially in girls with early or precocious puberty occurring as a consequence of low-dose cranial irradiation. It is concluded that optimum final heights may not be achieved in these patients without the therapeutic manipulation of the onset and/or duration of puberty. Key words: Isolated growth hormone deficiency, precocious puberty, duraiion of puberty, growth hormone dose.
The management of patients with idiopathic growth hormone deficiency (GHD) during puberty has recently been reviewed by Price et al. (1). Both sex steroids and growth hormone (GH) are required in puberty for an adequate growth spurt (2); however, little is known of the optimum dose regimens. Previous management regimens would now be considered inadequate, as they used suboptimal doses (3) and dose frequencies (4).GH administration more frequently than once a day may be advantageous (5). During the growth spurt of spontaneous puberty (Delemarre-van de Wad, pp. 26-31) and puberty induced by pulsatile therapy with gonadotrophin-releasing hormone (GnRH) (6), there is an increase in endogenous secretion of GH, which is coincident with the onset of the growth spurt. Indeed, other authors (7) have effectively doubled the dose of GH during puberty, because it was assumed that this was an appropriate therapeutic manoeuvre. For several years, reproductive endocrinologists have appreciated that GH has an important role both in gonadal development and in the maintenance of sexual maturation. GH acts on the gonads (8, 9), either by inducing local production of insulin-like growth factors (IGFs) or by a direct action, and augments gonadotrophin-modulated gonadal function (10, 11). GH treatment of female rhesus monkeys affects the rate of pubertal maturation without altering the onset of puberty (12). Such data has been used with the preliminary findings that GH may facilitate the induction of ovulation by gonadotrophins in human females (13).
48
R. Stanhope et al.
Acta Paedidtr Scand [suppl] 372
We have hypothesized that GH treatment may affect pubertal maturation, and that this may have important consequences for the treatment both of children with idiopathic GHD who have spontaneous puberty, and of children who have early o r precocious puberty in association with GH insufficiency resulting from low-dose cranial irradiation (14). W e have carried out a prospective study to determine whether a change in GH dose at the onset of the adolescent growth spurt would affect pubertal maturation, growth rate and, ultimately, final height in patients with idiopathic isolated GHD.
PATIENTS AND METHODS GHdependence of growth inprepuberry andpuberty. The growth of 182 children (93 boys, 89 girls) who were long-term survivors of acute lymphoblastic leukaemia in first remission was retrospectively examined. None had received spinal or gonadal irradiation and all had been treated with similar chemotherapeutic regimens, including intrathecal methotrexate. Ninety-three patients (5 1 boys, 42 girls; mean age at diagnosis 4.8 years) were treated with 2400 cGy cranial irradiation and 89 patients (42 boys, 47 girls; mean age at diagnosis 6.5 years) were treated with 1800 cGy cranial irradiation. None of the patients had received endocrine therapy. E’ect of GH dose during puberty. Thirty-two children (25 boys, 7 girls) with idiopathic isolated GHD were included in a prospective study in order to determine whether a change in GH dose during the adolescent growth spurt would improve final height. In late prepuberty all the patients were treated with a GH dose regimen of 15 IU/m*/week, administered as daily S.C.injections. At the onset of the growth spurt (equivalent to B2 in the girls and 8 ml testicular volume in the boys) the patients were randomized into two groups; one group continued with an identical dose regimen, while the other received 30 IU/m2/week with no change in frequency of administration. Girls commence their growth spurt with the onset of breast development, often before they have been shown to have acquired stage B2. However, for practical reasons, the acquistion of stage B2 was taken as the time for randomization of the GH dose regimen.
RESULTS
GH dependence of growth in prepuberty and puberty. The growth data for children surviving in first remission from acute lymphoblastic leukaemia showed a similar decrease in mean height SDS in prepuberty in both boys and girls (Fig. I), but there was a marked sex (a)
2400cGy
m Boys i X l Girls
I
-2L (b)
1800cGy
+2r -2L
-
-
Diagnosis
Onset of
Final
puberty
height
Fig. 1. Growth data from 182 children treated for acute lymphoblastic leukaemia with (a) 2400 cGy or (b) 1800 cGy of cranial irradiation. Height SDS values at diagnosis, at the onset of puberty, and at final height are shown. Standard deviations are shown by the horizontal bars.
Acta
Management of G f f deficiency through puberty
Paedzatr Scand [Suppl] 372
n
=
n = 10
15
I
h
2
12-
E-
10-
%
,
x
c .-
8 al
a-
>
E
0) ._
al
I
64-
2OL -0.5
I
I
0
0.5
I
I
I
1
1.5
2
1
-0.5
I
I
I
I
I
0
0.5
1
1.5
2
Years from 8 ml testicular volume Fig. 2. Two-year height velocity data for 25 boys with isolated idiopathic GHD who were randomized into either (a) an unchanged GH regimen of 15 IU/m2/week (n = 15) or (b) an increased dose of 30 IU/m2/week (n = 10) at onset of the growth spurt (8 ml testicular volume). GH was administered as daily S.C. injections.
difference during puberty. The boys had an almost equal decrease in increment of height SDS during prepuberty and puberty, while the girls had a much greater decrease in increment of height SDS during puberty than in the prepubertal period. The mean age of onset of puberty in the boys was not significantly different from normal boys (15), but there was a marked decrease in the mean age of onset of puberty in the girls. In the lower dose regimen of cranial irradiation, girls had a significant reduction in the mean age of onset of pubertal maturation, to 9.9 years ( p < 0.001), compared with normal girls (16). This decrease in the mean age of onset of pubertal maturation was less marked in the girls receiving high-dose irradiation (mean age of onset 10.6 years). The early onset of puberty in such girls with GH insufficiency has important implications for their therapy, which will be discussed below. Effect of GH dose during puberty. Height velocity at randomization was higher in the boys who were allocated to the higher dose GH regimen, but there was no significant difference in height velocity between the two groups over the first 2 years of this study ( p > 0.5) (Fig. 2). Mean height velocity at randomization for the boys who received an unaltered dose SD) and after 2 years was 8.6 2.6 cm/year. of GH was 6.9 f 1.5 cm/year (mean In the group who received the higher dose regimen, height velocity was 7.5 f 1.5 cm/year at randomization and 9.0 f 2.5 cm/year after 2 years. The response between individuals was heterogenous (Fig. 2). Data for the girls indicated a similar result, with no significant change in growth rate between the two treatment groups. However, the number of girls in this study was small (n = 7). Although the dose of GH had no significant effect on growth rate, it did influence the pattern of pubertal maturation. The mean age of onset of puberty in the boys was 12.0 years in the group with an unchanged dose regimen and 12.3 years in the group who received the higher dose of GH. In the group who received the higher dose, pubertal onset was delayed (although not significantly) compared with Tanner’s normal data (15). The rate of progress
*
*
49
50
R. Stanhope et al.
'-r,
G5[
10
11
12
13
14
15
Acta Paediatr Scand [Suppll 372
+ Boys on an unaltered
dose regimen -+
f33;:s;;;oubled
16
Age (years)
Fig. 3. Progress through puberty for the same boys as described in Fig. 2. The continuous line represents the mean age for the onset of each stage of puberty in normal boys (15). The vertical bars represent 1 standard deviation.
of puberty did not differ from normal in the group who received the low dose regimen (Fig. 3 ) . However, in the boys who received the higher dose, there was a faster progression through puberty (Fig. 3 ) . Thus, the higher dose regimen shortened the duration of puberty without altering the rate of growth, suggesting that the final height prognosis would not be improved, and indeed might even be decreased. The results for girls are not shown, as the numbers were small, but the trend for no improvement in growth rate and a faster progression through puberty with the higher dose regimen was similar to that in the boys. DISCUSSION Our data indicate an important relationship between GH secretion and the duration of puberty, which is not surprising considering the intimate relationship between GH and gonadal function (10). In order to achieve maximum height potential it would theoretically be advantageous to sustain a higher rate of growth, using high doses of exogenous GH, and at the same time slowing the rate of pubertal maturation, and thereby epiphyseal maturation, to allow a longer period for growth. However, the two factors are often interrelated, so that although low-dose cranial irradiation produces a lower incidence of GH insufficiency (17), such low-dose regimens are associated with early or precocious puberty in the girls (14), which is counterproductive. The boys, who did not have early sexual maturation, had a smaller decrease in height SDS during puberty than the girls. We hypothesize that using the 1800 cGy regimen, girls have a worse growth prognosis than boys because of the high incidence of early sexual maturation. The results of our studies of variable GH dose during puberty are preliminary, and larger numbers of patients will be required, with measurements taken to final height, before a definite conclusion can be drawn. Despite the shortcomings of our prospective study, the data suggest that an increase in the dose of GH during the pubertal growth spurt of children with idiopathic GHD may be counterproductive. By not inducing a significant increase in growth rate, while reducing the available time for pubertal growth, final height attainment may be unaltered or even reduced. Certainly GH treatment reduced the duration of puberty, and this may well be a dose-related phenomenon. With this available evidence, it seems unlikely that a change in dose of GH treatment would lead to an improved
Acta Paediatr Scand [Suppl] 372
Management of GH dejciency through puberty
final height attainment. We are unable to comment about whether this may be achievable by an alteration in dose frequency. Van der Werff ten Bosch and Bot (18) have recently described the growth of boys with multiple pituitary deficiencies prior to, and after, the availability of GH treatment. There are considerable difficulties in the interpretation of such data, because it spans a long period of time with no allowance for secular change, and also encompasses changes in treatment regimen. Twenty-four boys had pituitary hormone replacement, including androgen therapy, to induce puberty. Fourteen of the boys received GH treatment, whilst the other 10 did not. The patients who received GH treatment had a faster short-term growth rate, but progressed through ‘puberty’ in 5.5 years, compared with 9.5 years in the group who did not receive GH therapy. Because of the possibility that these results could have been an artefact of the study groups and that pubertal maturation was not spontaneous, our unit has analysed the duration of puberty in 37 girls and 97 boys with isolated GHD, during GH therapy (19). Although the timing of the onset of pubertal maturation in this group was delayed, the duration of puberty was shortened. The mean ages at onset of puberty were 12.1 years in the girls and 13.0 years in the boys, which represent significant delays compared with the values for normal children of 11.2 and 12.0 years, respectively, according to Tanner (16, 15). The mean duration of puberty (genitalia/breast stage 2 to 4) was 1.5 years in both sexes, which was considerably shorter than the corresponding normal values in girls (16) and boys (15). The dose of GH during this study was 4 IU intramuscularly three times a week, irrespective of body size, which makes the data more difficult to interpret in relation to current GH dose schedules. A more promising therapeutic approach may be to use a GnRH analogue to delay the onset, andlor increase the duration of pubertal maturation, thus allowing a longer period for pubertal growth. Certainly, the use of a GnRH analogue alone in children with GHD (20) or in girls with central precocious puberty (2 1) does not improve final height prognosis. However, the use of a GnRH analogue combined with GH treatment would certainly appear to be an appropriate therapeutic manoeuvre. This is true, particularly for the treatment of girls with acute lymphoblastic leukaemia with early sexual development. Such girls have a dual endocrinopathy, associated with low-dose cranial irradiation, GH insufficiency and earlylprecocious puberty. If treated with GH alone, the earlier onset of sexual maturation may be compounded by a shortened duration of pubertal maturation. The use of a combined regimen of GH and GnRH analogue has been reported (22), and the initial results of improvement in height prognosis are promising. It seems likely that a joint regimen of GH and GnRH analogue will theoretically yield improved results in terms of final heights of children with isolated GHD, but the optimal dose of GH to use during puberty will depend on the results of controlled prospective studies with the end point being final height attainment.
REFERENCES 1. Price DA, Shalet SM, Clayton PE. Management of idiopathic growth hormone deficient patients during puberty. Acta Paediatr Scand [Suppl] 1988; 347: 44-51.
2. Aynsley-Green A, Zachmann M, Prader A. Interrelation of the therapeutic effects of growth hormone and testosterone on growth in hypopituitarism. J Pediatr 1976; 89: 992-9. 3. Brook CGD, Hindmarsh PC, Smith PJ. The management of short stature. In: Brook CGD, ed. Clinical paediatric endocrinology. 2nd ed., Oxford: Blackwell Scientific Publications 1989; 118-27. 4. Albertsson-Wikland K. The effect of human growth hormone injection frequency on linear growth rate. Acta Paediatr Scand [Suppl] 1987; 337: 110-16. 5 . Hindmarsh PC, Stanhope R, Preece MA, Brook CGD. Frequency of administration of growth hormone - an important factor in determining growth response to exogenous growth hormone. Horm Res [Suppl 41 1990; 33: 83-9. 6. Stanhope R, Pringle PJ, Brook CGD. The mechanism of the adolescent growth spurt induced by low dose pulsatile GnRH treatment. Clin Endocrinol 1988; 28: 83-91.
51
52
R. Stanhope et al.
Acta
Paediatr Scand [Suppl] 372
7. Vanderschueren-Lodeweyckx M, van den Broeck J, Wolter R, Malvaux P. Early initiation of growth hormone treatment: influence on final height. Acta Paediatr Scand [Suppl] 1987; 337: 4-11. 8. Davoren JB, Hsueh JW. Growth hormone increases ovarian levels of immunoreactive somatomedin C/ insulin-like growth factor 1 in vivo. Endocrinology 1986; 118: 888-90. 9. Tres LL, Smith E, van Wyk JJ, et al. Somatomedin-C in rat sertoli-spermatogenic cell cultures. J Cell Biol 1983; 97: 18a. 10. Adashi EY, Resnick CE, D’Ercole AJ, et al. Insulin-like growth factors as intraovarian regulators of granulosa cell growth and function. Endocr Rev 1985; 6: 400-19. 1 1 . Mason HD, Martikainen H, Beard RW, Anyaoku V, Franks S . Direct gonadotrophic effect of growth hormone on oestradiol production by human granulosa cells in vitro. J Endocrinol 1990; 126: R1-4. 12. Wilson ME, Gordon TP, Rudman CG, Tanner JM. Effects of growth hormone on the temporosexual maturation in female rhesus monkeys. J Clin Endocrinol Metab 1989; 68: 29-38. 13. Homburg R, Eshel A, Abdalla HI, et al. Growth hormone facilitates ovulation induction by gonadotrophins. Clin Endocrinol 1988; 29: 113-17. 14. Leiper AD, Stanhope R, Kitching P, Chessells JM. Precocious and premature puberty associated with treatment of acute lymphoblastic leukaemia. Arch Dis Child 1987; 62: 1107-12. 15. Marshall WA, Tanner JM. Variations in the patterns of pubertal changes in boys. Arch Dis Child 1970; 45: 13-23. 16. Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in girls. Arch Dis Child 1969; 44: 291-303. 17. Shalet SM. Irradiation induced growth failure. Clin Endocrinol Metab 1986; 15: 591-606. 18. Van der Werff ten Bosch JJ, Bot A. Does hGH treatment promote adult height of hypopituitary children? Neth J Med 1988; 32: 217-25. 19. Darendeliler F, Hindmarsh PC, Preece MA, Cox L, Brook CGD. Growth hormone increases rate of pubertal maturation. Acta Endocrinol (Copenh) 1990; 122: 414-16. 20. Stanhope R, Brook CGD. The effect of gonadotrophin releasing hormone analogue on height prognosis in growth hormone deficiency and normal puberty. Eur J Pediatr 1988; 148: 200-2. 21. Stanhope R, Pringle PJ, Brook CGD. Growth, growth hormone and sex steroid secretion in girls with central precocious puberty treated with a GnRH analogue. Acta Paediatr Scand 1988; 77: 525-30. 22. Leiper AD, Stanhope R, Preece MA, Grant DB, Chessells JM. Precocious and early puberty and growth failure in girls treated for acute lymphoblastic leukaemia. Horm Res 1988; 30: 72-6. (R.S.) The Institute of Child Health 30 Guilford Street London WClN IEH UK
