| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Divisions of Endocrinology (A.A., M.S., S.A.O., E.O., B.R., C.J.T.), Neurosurgery (D.R.), and Clinical Chemistry (W.T.), Beaumont Hospital and Division of Radiation Oncology (S.B., C.F.), St. Lukes Hospital, Dublin 9, Ireland
Address all correspondence and requests for reprints to: Dr. Amar Agha, Department of Endocrinology, Beaumont Hospital, Beaumont Road, Dublin 9, Ireland. E-mail: amaragha{at}yahoo.com.
| Abstract |
|---|
|
|
|---|
Objective: The objective was to study the frequency of HP dysfunction in adults after RT for nonpituitary brain tumors.
Method: We studied 56 adult patients who received external beam RT for primary nonpituitary brain tumors at time intervals of 12150 months after RT. The control group consisted of 20 RT-naive patients with primary brain tumors. GH and adrenal axes were assessed using the insulin tolerance test or the glucagon stimulation test. Gonadotroph, thyrotroph, and lactotroph function were assessed using baseline blood measurements. The biological effective dose (BED) to the HP axis was calculated in the RT patients.
Results: Hypopituitarism was present in 41% of patients. The frequency of GH, ACTH, gonadotropin, and TSH deficiencies, and hyperprolactinemia was 32, 21, 27, 9, and 32%, respectively. Any degree of hypopituitarism and GH deficiency was significantly associated with longer time interval from RT and greater BED. However, gonadotropin deficiency and hyperprolactinemia were only related to BED, whereas ACTH deficiency was only significantly associated with the time interval from RT. One RT-naive patient was GH deficient.
Conclusion: Adult patients treated with cranial irradiation for primary nonpituitary brain tumors are at high risk of hypopituitarism, which is time and dose dependent. Long-term surveillance and periodic evaluation are needed. We recommend that adult late effect clinics, similar to those for children, should be established.
| Introduction |
|---|
|
|
|---|
Recent advances in the diagnostic and therapeutic interventions for primary brain tumors in adult patients are likely to result in a reduced rate of recurrence and improved survival. Therefore, more attention should be focused on late sequelae of treatment, including the potential for postirradiation hypopituitarism. In this paper, we report the results of a study of postirradiation HP function in adult survivors of primary nonpituitary brain tumors.
| Patients and Methods |
|---|
|
|
|---|
Fifty-six patients (28 males) who received cranial RT for primary brain tumors between 1991 and 2003 were included in the study. Patients were identified and recruited from the RT clinic at St. Lukes Hospital in Dublin. Patients were eligible for the study if they were older than 16 yr of age at the time of RT, if they had a primary brain tumor that was distant from the HP axis, and if at least 1 yr had elapsed since the last dose of irradiation. Exclusion criteria were as follows: patients with known HP dysfunction, patients with malignant astrocytic tumors (World Health Organization grades III or IV), patients who received glucocorticoid treatment in the preceding 6 months, and patients who were too ill to undergo testing. In all, 79 patients were initially identified to be eligible, but 19 declined to participate and four could not be contacted. Patients baseline characteristics are shown in Table 1
.
|
Twenty radiation-naive patients (12 males) with primary extrasellar brain tumors were also studied. The control patients were randomly selected from admission records at Neurosurgical Department in Beaumont Hospital. The control and RT patients were matched for their baseline characteristics (Table 1
), and similar exclusion criteria were applied.
RT
Cranial irradiation was delivered by linear accelerators of 68 MV in fractionated doses that included all or part of the hypothalamus and pituitary gland in its field. The radiation dose to the HP axis was estimated by either dose-volume histograms (using planning computerized tomography, after the year 2000) or by manual calculation from simulation films (before the year 2000). To allow for direct comparison of different RT fractionation schedules, the dose was transformed to the biological effective dose (BED), which has been shown previously to be a predictor of late adverse endocrine effects (13). The BED was calculated using the quadratic model, as follows: BED = D x [1 + d/(
/ß)], where D is the total dose, d is the fraction size,
represents is the linear nonreparable component of cell killing, and ß represents quadratic component of cell killing.
/ß = 3 for late responding tissue such as nerve tissues. Two patients with medulloblastomas also received fractionated spinal irradiation and local irradiation to the posterior fossa.
Chemotherapy (CT)
Five patients with gliomas received a combination of procarbazine, lomustine, and vincristine, and one patient received temozolomide. The two patients with medulloblastomas received a combination of carboplatin and etoposide (VP16).
Endocrine assessment
GH and adrenal axes (Fig. 1
).
Twenty-five RT patients (45%) and seven radiation-naive controls (35%) who had no history of seizures were studied using the insulin tolerance test (ITT) (14, 15, 16, 17). The remaining patients and controls with seizures were initially screened using the glucagon stimulation test (GST) (18, 19, 20, 21). Because the GST has occasionally been associated with false-positive (false-fail) results (18, 22), subjects who failed the test were subsequently reassessed using a second provocative test: the arginine stimulation test (AST) for GH axis (23, 24) and the standard-dose short synacthen (corticotrophin) test (SST) (25, 26, 27) for adrenal function (Fig. 1
). Basal plasma ACTH concentration was measured in all cases.
|
Other anterior pituitary hormones. Baseline morning serum samples were withdrawn for measurements of FSH, LH, free T4 (FT4), TSH, prolactin, IGF-I, and testosterone in males or estradiol levels in females. In all females, menstrual history was obtained.
In males, gonadotropin deficiency was defined by a low serum testosterone with inappropriately low (below or within normal range) gonadotropin level, in premenopausal females, by oligo-amenorrhea in the presence of low serum estradiol level without a rise in gonadotropin level, and, in postmenopausal females, by serum gonadotropin concentration in the premenopausal range (28, 29). TSH deficiency was defined by low serum FT4 level (after excluding artifactual causes) without appropriate elevation in serum TSH (28, 29). Hyperprolactinemia was defined as a basal level greater than the locally derived normal assay reference range.
Analytical methods
Serum GH was assayed using a noncompetitive immunoradiometric method (DiaSorin, Vercelli, Italy), with an intraassay coefficient of variation (CV) of 1.9, 1.9, and 3.9% at GH means of 18.4, 8.2, and 1.71 µg/liter, respectively. The interassay CV was 5.5, 4.5, and 2.3% at GH means of 17.5, 7.7, and 1.69 µg/liter, respectively. Conversion factor was 1 µg/liter = 2.5 mIU/liter. Serum cortisol was measured using a fluoroimmunoassay (AutoDELFIA; PerkinElmer, Turku, Finland) with an intraassay CV of 3.6, 2.7, and 3% and an interassay CV of 1.6, 1.1, and 1.5%, at serum cortisol concentrations of 210, 517, and 781 nmol/liter, respectively. All GH and cortisol samples from any single individual were assayed in the same batch.
Serum IGF-I was measured using HCl ethanol extraction radioimmunoassay (Nichols Institute Diagnostics, San Juan Capistrano, CA). Serum testosterone, LH, FSH, FT4, TSH, prolactin, and estradiol were measured using a fluoroimmunoassay (AutoDELFIA; PerkinElmer). Plasma ACTH was measured using a two-site immunoradiometric assay (ORIS Group, Gif-sur-Yvette, France). Plasma glucose was measured using the hexokinase method.
Normal reference ranges were as follows: testosterone, males, 10.334.5 nmol/liter (3001000 ng/dl); premenopausal estradiol, more than 100 pmol/liter (27ng/liter); FT4, 821 pmol/liter (0.621.62 ng/dl); TSH, 0.54.2 mIU/liter; prolactin, males, 83414 mIU/liter (2.311.5 ng/ml); females, 90523 mIU/liter (2.514.6 ng/ml); plasma ACTH, 1065 ng/liter.
Statistical analysis
Age, body mass index (BMI), and IGF-I SD scores (SDS) were log transformed before testing for significance using a two-sample t test. However, for descriptive purposes, data are expressed as untransformed mean ± SD. The BED and time interval from RT to testing were significantly skewed and were compared using the Wilcoxons rank-sum test for nonparametric measurements. Categorical data were compared using the Fischers exact test. Multifactorial logistic regression models were developed to assess the effect of appropriate variables in the presence of other confounding variables in the development of pituitary hormones abnormalities. The dependent variables for the models were any degree of hypopituitarism; GH, ACTH, and gonadotropin deficiencies, and hyperprolactinemia, whereas the independent variables were age, gender, BMI, BED, time interval from RT, and the presence or absence of adjuvant CT. Correlation between variables was made using the Spearmans coefficient (rs). Serum IGF-I SDS were calculated according to the following formula: IGF-I SDS = In (IGF-I) [5.92 (0.0146 x age in years)]/0.272 (30). Significance of results was determined according to the two-tailed alternative hypothesis, and results were deemed significant for P values <0.05. All of the analysis was performed using the statistical software package STATA (version 8; StataCorp, College Station, TX).
Ethics
The study was approved by the ethics section of Beaumont Hospital Medical Research Committee. The purpose of the study was explained carefully to patients who were provided with written information on the background to the study. After an interval of 1 wk, patients who agreed to participate signed written consent for inclusion in the study.
| Results |
|---|
|
|
|---|
Evidence of hypopituitarism was present in 23 patients (41%). Nine patients (16%) had single hormone deficiency, whereas 14 patients (25%) had multiple deficiencies, including four (7%) with panhypopituitarism. The breakdown pattern of anterior pituitary hormones deficiency is shown in Fig. 2
.
|
Seven patients failed the ITT (peak cortisol response, 46481 nmol/liter). Nine patients failed the GST (peak cortisol response, 33448 nmol/liter), of whom five also failed the SST (30-min cortisol response, 41476 nmol/liter) (Fig. 1
). Therefore, 12 patients (21%) were defined as glucocorticoid deficient. All 12 patients had inappropriately low basal serum ACTH concentration.
Other abnormalities. Fifteen patients (27%, seven males) had gonadotropin deficiency, which was associated in three patients with mild hyperprolactinemia [prolactin levels one to two times the upper limit of normal (ULN)] and in five patients with moderate hyperprolactinemia (prolactin level two to four times the ULN). Four male patients had evidence of primary testicular failure (three received CT); one had low testosterone with raised LH and FSH, one had low-normal testosterone with raised LH and FSH, and one had isolated raised FSH with normal testosterone and LH levels. The fourth patient who did not receive CT had isolated raised FSH with normal serum LH and testosterone concentrations.
Five patients (9%) had evidence of central hypothyroidism. One of the two patients who had spinal irradiation had subclinical primary hypothyroidism, and another patient had thyroid-peroxidase antibodies-positive primary hypothyroidism. Hyperprolactinemia was present in 18 patients (32%), eight had prolactin levels between one and two times the ULN, and 10 had prolactin levels more than two times ULN. One patient who had panhypopituitarism had low serum prolactin level.
Control subjects
All controls who had the ITT and 11 of 13 controls who had the GST achieved normal GH response more than 5 µg/liter. One control subject failed the GST (GH response, 3.25 µg/liter) but passed the AST (GH response, 7.22 µg/liter). Another, lean 33-yr-old control subject who had extensive debulking of a right-sided parietal astrocytoma failed both the GST (undetectable GH response) and the AST (GH response, 1.82 µg/liter; IGF-I SDS, 1.1). Two control subjects failed the GST but passed the SST.
All control subjects had normal gonadal and thyroid function and normal prolactin levels.
Risk factors for postradiation hypopituitarism (Tables 2
and 3
)
Logistic regression analysis showed that any degree of hypopituitarism and GHD was significantly associated with the time interval from RT and the BED (Table 2
). However, gonadotropin deficiency and hyperprolactinemia were only related to BED, whereas ACTH deficiency was only significantly associated with the time interval from RT (Table 2
). Age at time of RT, gender, BMI, or adjuvant CT were not predictors of the development of hypopituitarism.
|
|
|
| Discussion |
|---|
|
|
|---|
We have also been able to show that the insult to the HP axis is primarily caused by RT, because only one of the radiation-naive patients had GHD. This would strongly suggest that neither the original tumor nor the subsequent neurosurgery was responsible for the development of pituitary dysfunction in the vast majority of patients. It has been speculated that CT could potentiate the damaging effect of RT on the HP axis (33). We did not, however, find an association between the use of CT and HP axis dysfunction, although the small number of patients who received CT precludes definitive conclusion about this speculative relationship.
The effect of cranial irradiation on HP function in adults treated for nonpituitary disease has only received limited attention. Harrop et al. (34) studied nine patients who were irradiated for primary brain tumors that were remote from the HP axis, six of whom were older than 16 yr at the time of RT. Five patients showed impaired GH, two impaired gonadotropin, and one impaired TSH secretion. Adrenal function appeared normal in all patients. Littley et al. (35) evaluated pituitary function in seven adults who were studied 513 yr after RT for gliomas (30 Gy in eight fractions) and reported GH, gonadotropin, ACTH, and TSH deficiencies in six, five, three, and two patients, respectively. The higher rate of pituitary dysfunction in this study probably reflected the longer interval between RT and evaluation. In contrast to these studies, Constine et al. (36) evaluated 32 patients aged 665 yr who received cranial RT for primary brain tumors at 213 yr before the study and found a strikingly high frequency of central hypothyroidism in 65% of patients, but the frequency of hypothyroidism among children and adults was not separately reported. Sixty-one percent of the postpubertal patients had hypogonadism, but there was inconsistency in the reported frequency of adrenal insufficiency between the metyrapone test (35% fail) and the CRH test (all normal). GH axis was not assessed in that study.
One particular difficulty in assessing GH and ACTH reserves in adult patients with brain tumors is the high frequency of seizure disorders associated with tumors of astrocytic origin, which precludes the use of the gold-standard ITT (14, 15, 16, 17). The alternative test, the GST, assesses the entire GH and adrenal axes and has been shown to be reliable (18, 19, 20, 21), although it is occasionally associated with false-positive results (18, 22). For this reason, we reassessed patients who failed the GST with another stimulus and only accepted the patients to be GH or ACTH deficient if they failed the second test also. Therefore, we feel confident that we did not overestimate the prevalence of GH and ACTH deficiency in our cohort. It is, however, possible that some of our RT patients who showed normal stimulated GH response may have GH neurosecretory dysfunction, a state that is characterized by diminished spontaneous or physiological GH secretion but normal response to provocative stimuli (37).
The insult causing pituitary dysfunction after cranial irradiation is believed to be primarily hypothalamic in origin because the hypothalamus is more sensitive to the effect of radiation than the pituitary (33, 38). The exact mechanism of radiation damage to the hypothalamus is not known, but both vascular etiology and direct damage to the cell nuclei have been proposed. The latter hypothesis may explain the delayed onset of hormone deficiency until the cells of the slowly dividing tissue die during mitosis (38).
Our findings have important implications for clinicians looking after this group of patients. Adult patients who received cranial irradiation for nonpituitary brain tumors should be screened for the late effects of hypopituitarism. Replacement therapy with sex steroids, glucocorticoids, and T4 have well-established value, but the benefit (39) and safety (40) of GH replacement in this subgroup of hypopituitary patients is less well established and requires additional studies.
Current 5-yr overall survival rates for patients who were treated for low-grade gliomas are estimated to be between 60 and 70% (41), whereas treatments for other more benign tumors such as meningiomas are highly effective. Therefore, more attention should be focused on the slow effect of cranial RT on the HP axis, necessitating long-term surveillance and periodic clinical and biochemical evaluation to detect evolving hypopituitarism. We recommend that adult late effects clinics, similar to those for children, should be established for adult RT patients. In this context, prospective longitudinal studies are needed to establish the appropriate timing and frequency of post-RT pituitary assessment.
| Footnotes |
|---|
First Published Online September 6, 2005
Abbreviations: AST, Arginine stimulation test; BED, biological effective dose; BMI, body mass index; CT, chemotherapy; CV, coefficient of variation; FT4, free T4; GHD, GH deficiency; GST, glucagon stimulation test; HP, hypothalamic-pituitary; ITT, insulin tolerance test; RT, radiotherapy; SDS, SD scores; SST, short synacthen (corticotropin) test; TBI, total body irradiation; ULN, upper limit of normal.
Received July 12, 2005.
Accepted August 31, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Fernandez, M. Brada, L. Zabuliene, N. Karavitaki, and J. A H Wass Radiation-induced hypopituitarism Endocr. Relat. Cancer, September 1, 2009; 16(3): 733 - 772. [Abstract] [Full Text] [PDF] |
||||
![]() |
H J Schneider, S Rovere, G Corneli, C G Croce, V Gasco, R Ruda, S Grottoli, G K Stalla, R Soffietti, E Ghigo, et al. Endocrine dysfunction in patients operated on for non-pituitary intracranial tumors. Eur. J. Endocrinol., October 1, 2006; 155(4): 559 - 566. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |