The Journal of Clinical Endocrinology & Metabolism Vol. 89, No. 10 4999-5002
Copyright © 2004 by The Endocrine Society
Adult Heights Attained by Children with Hypothalamic/Chiasmatic Glioma Treated with Growth Hormone
Roy J. Kim,
Anna Janss,
Dana Shanis,
Sarah Homan and
Thomas Moshang, Jr.
Divisions of Endocrinology (R.J.K., D.S., S.H., T.M.) and Neurooncology (A.J.), Department of Pediatrics, University of Pennsylvania School of Medicine and Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 19104
Address all correspondence and requests for reprints to: Dr. Thomas Moshang, Jr., Division of Endocrinology, Children’s Hospital of Philadelphia, 34th and Civic Center Boulevard, Philadelphia, Pennsylvania 19104. E-mail: moshang{at}email.chop.edu.
 |
Abstract
|
|---|
Hypothalamic/chiasmatic gliomas (H/CG) in children are commonly accompanied by endocrine dysfunction due to mass effects of the tumor itself or as a consequence of tumor therapy, with GH deficiency (GHD) being the most common disorder. We report the height outcomes of GH-treated H/CG patients with GHD. We reviewed the records of 14 GHD patients with H/CG who were treated with human GH. A comparison group of non-GH-treated H/CG patients was also identified. Heights were expressed as SD scores (SDS). For GH-treated patients, the mean initial height was –0.7 ± 0.3 (±SE). Their mean final height was –0.3 ± 0.3. The mean change in height SDS for the GH-treated group was +0.4. The mean initial and final height SDS for the non-GHD patients were 0.6 (SE = 0.4) and 0.0 (SE = 0.4), respectively. The mean change in height SDS was –0.6. The GHD patients had significantly lower initial height SDS compared with the non-GHD patients (P = 0.01) and had a significantly greater change in their height SDS (P = 0.04). GH treatment for H/CG patients restores much of their growth potential and improves adult height to within normal limits.
|
Introduction
|
|---|
BRAIN TUMORS ARE the most common pediatric solid organ tumor. Low grade astrocytomas occurring in the hypothalamic/chiasmatic area, also known as hypothalamic/chiasmatic gliomas (H/CGs), are a frequent form of childhood brain tumor, with a high rate of survival, but also a high rate of recurrence. This often indolent tumor may be treated with surgery, chemotherapy, or radiation. In some patients, especially those with neurofibromatosis, it is treated only expectantly. In those children with H/CG treated with cranial irradiation, GH deficiency (GHD) is the most commonly identified endocrine disorder, occurring in 32% of all H/CG patients (1). Although treatment with recombinant human GH (rhGH) is used for treatment of GHD in children surviving H/CG, there is a paucity of data on its effects on growth and final adult height in this setting.
There have been no reports on adult height outcomes in patients with H/CG. In fact, the growth response to rhGH in this specific brain tumor in children is not known. This may be related to the reluctance to give rhGH treatment to patients whose tumor has a significant chance of recurrence. The present report describes the effect of GH replacement on final height in H/CG patients seen at Children’s Hospital of Philadelphia.
|
Subjects and Methods
|
|---|
Patients
Using a clinic database, we identified all H/CG patients who received GH treatment for GHD. Patients were included in the study if they had reached adult height, determined as the height attained when the bone age reached 16 yr for males or 14 yr for girls or the sexual maturity rating was 5, and when growth velocity for the preceding year had reached a plateau (<2 cm/yr). A comparison group of H/CG patients was identified through the same database and included all H/CG patients who had reached adult height and had sufficient similar medical data as the GH-treated patients, as described below. The comparison group had normal initial growth velocity and was believed not to have GHD. For all patients, standing heights were assessed in an out-patient clinic using a fixed wall stadiometer. All study patients had their medical charts examined for relevant data using a standardized abstraction form. The data retrieved included age, sex, tumor treatment, height before GH therapy, GH dose and duration of treatment, final recorded height, Tanner stage, bone age results if available, other medications, and midparental height. Midparental height was calculated as follows: for males: (father’s height + mother’s height + 12.7 cm)/2; for females: (father’s height + mother’s height – 12.7 cm)/2. All heights were expressed as age- and sex-adjusted SD scores (SDS) based on the 2000 Growth Charts from the Centers for Disease Control and Prevention (2). The study protocol was reviewed and approved by the institutional review board at Children’s Hospital of Philadelphia.
GHD
GHD was diagnosed based upon biochemical testing of those H/CG patients who had a declining growth velocity after tumor treatment. GHD was established if the GH peak was less than 7 µg/liter on two provocative tests or was based on the results of an overnight neurosecretory GH study (mean GH <2.5 µg /liter and no more than two GH peaks 
8 µg/liter) (3). One patient who received GH started therapy before the availability of rhGH. All other GH recipients started treatment with the recombinant form. The GH dose was 0.3 mg/kg·wk. GH therapy was given for a mean of 6 yr (SD = 3 yr).
Statistical analysis
All analyses were carried out using the heights expressed as SDS. The paired t test was used to compare initial and adult heights of subjects and controls. The nonpaired t test was used to compare the mean heights of GHD and non-GHD patients and to compare the changes in heights of GHD and non-GHD patients. The t tests were two-tailed, and a threshold of 0.05 was used to define a significant P value.
|
Results
|
|---|
There were 16 GHD patients identified through the database screen. Two GHD patients who had not reached adult height had died due to tumor recurrence and were omitted from the study. Data for the remaining 14 are reported here. There were 16 non-GHD control patients with data sufficient to be included in the analysis. The characteristics of both groups are shown in Table 1
. Twelve (86%) of 14 GHD patients received radiation as part of their tumor treatment. Of the remaining two GHD patients, one had 50% debulking surgery, and the other had neurofibromatosis and was treated only with chemotherapy. Five (31%) of 16 non-GHD patients received radiation. The median radiation dose was 54 Gy (range, 46–56 Gy) for all patients, and the mean did not vary between groups (P = 0.4). Chemotherapy was given to seven (50%) GHD patients and three (19%) non-GHD patients. Those patients who had surgical procedures most often had biopsy to define the tumor, and some were treated with debulking procedures. The brain tumors were diagnosed at a mean age of 3.3 yr in the GHD patients and 5.5 yr in non-GHD patients. The mean age at the last visit for each patient group was 17 yr.
Among GHD patients, five had precocious puberty (36%), of whom four had received GnRH agonist therapy. One GHD patient had early, but not precocious, puberty. Among controls, three had precocious puberty (19%), of whom one had received a GnRH agonist, and a fourth patient had early, but not precocious, puberty. Of the GHD patients, two were also taking T4, and one was receiving hydrocortisone replacement. Similarly, of the non-GHD patients, two were taking T4, and one was receiving hydrocortisone. One of the non-GHD patients was receiving desmopressin acetate, T4, and hydrocortisone.
For GH-treated patients, the mean initial height SDS was –0.7 (SE = 0.3), with a median of –0.7 and a range of –2.5 to 1.5 (Table 2). The mean final adult height SDS in the GH-treated patients was –0.3 (SE = 1.1), with a median of –0.5 and a range of –1.8 to 1.8. The mean change in height SDS from baseline for the GH-treated group was +0.4. This increase from initial height to final height was not statistically significant. The mean initial height SDS for the non-GHD patients was 0.6 (SE = 0.4), with a median of 0.6 and a range of –1.7 to 3.0. The mean final adult height SDS was 0.0 (SE = 0.4), with a median of –0.05and a range of –2.5 to 2.4. The mean change in height SDS for the non-GHD group was –0.6. This decline between initial and final height did not reach statistical significance (P = 0.1).
The GHD patients were significantly shorter at the start of treatment (P = 0.01; see Fig. 1). The final heights of the GHD and non-GHD patients were not significantly different (P = 0.5). The change in height SDS was significantly greater for GH-treated patients than for non-GHD patients (P = 0.04). The midparental height for both groups was 0.5 SDS (P = 0.997; Table 2
and Fig. 2) Neither patient group reached their genetic potential, as estimated by the midparental height (P = NS for GHD and non-GHD subjects; Fig. 2).
View larger version (8K):
[in this window]
[in a new window]
|
FIG. 1. Initial and adult heights in children with H/CG. Shown are the mean initial and adult heights (±SE) for non-GHD patients ( ) in comparison to GHD patients (•).
|
|
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 2. Comparison of H/CG patients’ adult heights with their midparental heights. The results for non-GHD patients are shown in A. The results for GHD patients are shown in B. Bold line, Mean. The differences between mean patient heights and midparental heights were nonsignificant (P = 0.35 and 0.06 in A and B, respectively).
|
|
To test the possibility that the radiation recipients among the patients diagnosed as non-GHD were those who had the greatest declines in height SDS (suggesting GHD of delayed onset), we performed a subgroup analysis, stratifying patients by radiation exposure. The irradiated non-GHD children had a mean height SDS decline of –1.3 vs. a decline of only –0.2 among the nonirradiated non-GHD patients (P = 0.1). The final height SDS of the irradiated and nonirradiated non-GHD patients were –0.9 and 0.4, respectively (P = 0.1). Although not statistically different, in all likelihood due to the small numbers, the difference suggests that the irradiated group did have delayed onset GHD.
|
Discussion
|
|---|
Although there are a number of reports that describe the early effects of GH therapy on the growth of children with different neoplasms, there are few reports documenting adult height outcomes. In a retrospective cohort study of children with a variety of brain tumors, growth velocity increased after the first year of GH treatment, although final height was much less than midparental height and was dependent on type of treatment: patients receiving craniospinal irradiation were the shortest, and chemotherapy had an additive negative effect (4). GH treatment in patients with GHD from craniospinal irradiation for brain tumors distant from the hypothalamic-pituitary region resulted in less height loss compared with controls. The adult height SDS attained, however, was only –3.4 for GH recipients and –2.6 for untreated control patients who had normal initial growth (5). These low adult heights and slow growth associated with craniospinal irradiation are probably due to radiation-induced damage of the vertebral column. An increase in growth velocity for the first 2 yr of treatment was observed in one study of craniopharyngioma patients, although no final height data were presented in that report (6). In a study of irradiated patients who had acute lymphoblastic leukemia, face and neck tumors, or medulloblastoma, GH therapy seemed to attenuate the loss of height from GHD, but the effects did not reach statistical significance (7). Another study of acute lymphoblastic leukemia patients showed that GH replacement increased final height by 1 SD compared with GHD subjects who were not treated with GH (8). A 1985 study of 27 patients with a variety of brain tumors, excluding craniopharyngioma, showed an improvement in height of about 1 SD with GH therapy; however, the final height was only –2.1 SDS (9).
In our study, GHD patients were significantly shorter initially compared with the non-GHD subjects. Over time, GH treatment was associated with a significantly greater gain in SDS compared with non-GHD patients. The average net impact of GH therapy was an increase of 0.4 SD or 2.9 cm for an adult male and 2.6 cm for an adult female. At the end of the study period, the final heights of the GHD and non-GHD patients did not differ significantly. This suggests that treatment with GH in deficient patients restores much of their height potential relative to non-GHD tumor patients. Notably, however, both patient groups had mean adult height SD scores that fell short of their midparental height.
Although GH treatment was associated with height gains, the patients in the non-GHD group had declines in their height SDS over time. The average loss was more than 0.5 SD. Although the fall in SDS did not reach statistical significance, the negative trend raises the question of whether some of these patients developed GHD over the years following their tumor treatment. We showed that among our non-GHD patients, those receiving radiation therapy showed a tendency toward greater height SDS loss with time and a trend toward a lower adult height compared with nonirradiated non-GHD patients. Furthermore, two of the non-GHD patients were receiving replacement therapy for other hormone deficiencies; they were also radiation recipients. These two patients showed declines in height SDS of –2 and –1.3. In addition, there were two other patients in the control group who had precocious puberty, which was untreated. They did not receive radiation. These patients had lost –0.6 and –1.7 in height SDS by the time they reached adult height. A retrospective report evaluating adult height outcomes in brain tumor patients indicated that height improvement over the recent years was due to the use of GnRH agonist treatment (10). A final factor was whether surgical intervention or tumor location per se could be contributing to the occurrence of GHD and poor growth, such as in the two nonirradiated GHD patients, one of whom was treated with debulking surgery and the other with neurofibromatosis treated only with chemotherapy. Collectively, these analyses suggest that the use of irradiation and untreated precocious puberty are early and independent predictors for compromised adult height, even in the presence of a normal initial growth velocity.
Although the patient population differed, our findings were similar to those reported by Clayton et al. (5), who found that GH therapy for radiation-induced GHD attenuated loss of height relative to that in untreated subjects. They also observed, as we did, that patients with initially normal growth velocity who did not receive GH subsequently had a decline in their height SDS. In their series, the untreated patients were all demonstrated to be GHD, and all had received craniospinal irradiation. Their height loss was 1.7 SD. In our series, in which only a minority of the untreated patients received radiation, and none of it included the spine, the height loss was 0.6 SD. We suspect that some of the non-GHD patients in our study may have developed GHD in the years subsequent to their initial tumor treatment. Our subgroup analysis of the irradiated non-GHD patients suggests that these patients may have experienced greater declines in their height SDS because of later onset GHD. This subgroup analysis suggests that all H/CG patients be followed expectantly for the development of potential GHD even in the face of early normal growth velocity.
That GHD can arise several years after radiation therapy has been described in other patient populations. Clayton and Shalet (11) found that abnormal insulin-stimulated GH responses (GH peak, <15 µg /liter) occurred with increasing frequency with each passing year in patients who received cranial radiation for a variety of neoplasms. All seven patients in their study who received more than 35 Gy had GHD by 5 yr postradiation. In a report of 25 children irradiated for ependymoma, low grade astrocytoma, or high grade astrocytoma, 11 had developed GHD after 6 months, and 20 after 12 months (12).
In summary, the high radiation doses used in the treatment of brain tumors are associated with compromised adult height, possibly due to GHD. Because this can occur even if initial growth velocity is normal, testing and later retesting of GH secretion capacity may be warranted in patients who received radiation. GH therapy in this setting results in improvement in final height, although not full genetic potential. Deficiencies of other hormones may be a marker for hypothalamic or pituitary injury and should also raise clinical suspicion of GHD. Premature closure of growth plates associated with precocious puberty is an additional source of adult height loss in these patients and should be aggressively suppressed with a GnRH agonist (10). These findings illustrate that the development of endocrinopathy after H/CG treatment can be a dynamic process and underscore the need for careful serial monitoring of growth and development. Clinicians should maintain a low threshold for evaluating and treating GHD and precocious puberty in these patients, because the prompt treatment of these conditions preserves growth potential.
|
Footnotes
|
|---|
This work was supported by a T32 institutional grant from the National Institutes of Health (to R.J.K.).
This work was presented in part at the 84th Annual Meeting of The Endocrine Society, Philadelphia, PA, June 18–22, 2003.
Abbreviations: GHD, GH deficiency; H/CG, hypothalamic/chiasmatic glioma; rhGH, recombinant human GH; SDS, SD score.
Received March 12, 2004.
Accepted June 30, 2004.
|
References
|
|---|
- Collett-Solberg PF, Sernyak H, Satin-Smith M, Katz LL, Sutton L, Molloy P, Moshang Jr T 1997 Endocrine outcome in long-term survivors of low-grade hypothalamic/chiasmatic glioma. Clin Endocrinol (Oxf) 47:79–85[CrossRef][Medline]
- 2000 Growth charts. Webpage of Centers for Disease Control and Prevention; reviewed December 30, 2002; cited December 3, 2003; available at: http://www.cdc.gov/growthcharts
- Rose SR, Ross JL, Uriarte M, Barnes KM, Cassorla FG, Cutler Jr GB 1988 The advantage of measuring stimulated as compared with spontaneous growth hormone levels in the diagnosis of growth hormone deficiency. N Engl J Med 319:201–207[Abstract]
- Ogilvy-Stuart AL, Shalet S 1995 Growth and puberty after growth hormone treatment after irradiation for brain tumours. Arch Dis Child 73:141–146[Abstract/Free Full Text]
- Clayton PE, Shalet SM, Price DA 1988 Growth response to growth hormone therapy following craniospinal irradiation. Eur J Pediatr 147:597–601[CrossRef][Medline]
- Hogeveen M, Noordam C, Otten B, Wit J, Massa G 1997 Growth before and during growth hormone treatment in children operated for craniopharyngioma. Horm Res 48:258–262[Medline]
- Sulmont V, Brauner R, Fontoura M, Rappaport R 1990 Response to growth hormone treatment and final height after cranial or craniospinal irradiation. Acta Paediatr Scand 79:542–549[Medline]
- Adan L, Souberbielle J, Blanche S, Leverger G, Schaison G, Brauner R 1996 Adult height after cranial irradiation with 24 Gy: factors and markers of height loss. Acta Paediatr 85:1096–1101[Medline]
- Herber SM, Dunsmore IR, Milner RD 1985 Final stature in brain tumours other than craniopharyngioma: effect of growth hormone. Horm Res 22:63–67[Medline]
- Gleeson HK, Stoeter R, Ogilvy-Stuart AL, Gattamaneni HR, Brennan BM, Shalet SM 2003 Improvements in final height over 25 years in growth hormone (GH)-deficient childhood survivors of brain tumors receiving GH replacement. J Clin Endocrinol Metab 88:3682–3689[Abstract/Free Full Text]
- Clayton PE, Shalet SM 1991 Dose dependency of time of onset of radiation-induced growth hormone deficiency. J Pediatr 118:226–228[CrossRef][Medline]
- Merchant TE, Goloubeva O, Pritchard DL, Gaber MW, Xiong X, Danish RK, Lustig RH 2002 Radiation dose-volume effects on growth hormone secretion. Int J Radiat Oncol Biol Physics 52:1264–1270[CrossRef][Medline]
Top
Abstract
Introduction
