| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Endocrinology, Christie Hospital, Manchester M20 4BX, United Kingdom
Address all correspondence and requests for reprints to: Dr. Stephen M. Shalet, Department of Endocrinology, Christie Hospital, Wilmslow Road, Manchester M20 4BX, United Kingdom. E-mail: stephen.m.shalet{at}man.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Study Subjects: Cortisol profiling at 20-min intervals over 24 h during the fed state was undertaken in 34 patients (10 females), aged 17–53.7 yr (median, 21.5 yr), 2–29 yr (median, 11.5 yr) after receiving conventional cranial irradiation for nonpituitary brain tumors or leukemia (n = 5) and in 33 age-, gender-, and body mass index-matched normal controls, of whom 23 patients and 17 controls were also profiled in the last 24 h of a 33-h fast.
Results: The fed profile mean cortisol concentration (mean ± SEM) was significantly increased (by 14%) in the patients compared with that in normal subjects (213 ± 6.9 vs. 187 ± 6.7 nmol/liter; P = 0.009), with all individual values above the lowest seen in normal subjects. Multiparameter deconvolution analysis revealed a parallel increase (by 20%) in cortisol secretion rates (1.8 ± 0.09 vs. 1.5 ± 0.08 nmol/liter·min; P = 0.03) due to selective augmentation of the cortisol mass released per burst with no changes in burst frequency (12/24 h) or half-life. No significant differences were observed between males and females, after short-term fasting, or between female patients and normal females. Thus, in the light of total group comparisons, male patients had even higher values than normal males, and more so during fasting (mean cortisol and cortisol secretion increased by 20 and 29% in the fed state and by 41 and 32% in the fasting state, respectively; P < 0.05).
Conclusions: This study has demonstrated that radiation-induced ACTH neurosecretory dysfunction does not exist and, thus, resolved the clinical dilemma as to whether cortisol replacement should be considered in those patients with excessive fatigue and normal stimulated cortisol responses. On the contrary, cranial irradiation causes activation of the corticotrope-adrenal axis, and in the absence of ACTH deficiency, this activation is manifested by parallel increases in circulating cortisol levels and cortisol production rates without any change in cortisol half-life. The lack of cortisol increase in female patients may be attributed to the adverse effect of their higher body mass index on cortisol secretion or may reflect a genuine gender dichotomy.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
In contrast, ACTH deficiency is quite uncommon (<5%) with radiation doses of less than 50 Gy (6) and is virtually absent with doses less than 24 Gy (4, 5). A much higher incidence of ACTH deficiency, however, is encountered after intensive irradiation (7, 8, 9, 10) and after conventional irradiation for pituitary tumors (11).
Although a robust diagnosis of clinically significant ACTH deficiency can almost always be achieved by a standard provocative test such as the insulin tolerance test (ITT), these tests can only explore the maximum h-p-adrenal reserve and are insensitive to mild functional alterations in the axis or its activity under normal circumstances (7, 12, 13). Thus, there has always been a belief that a discordance might exist between the actual corticotrope-adrenal axis status and that determined by the pattern of responses to stimulation tests. In particular, it has been suggested (14) that neurosecretory dysfunction might affect ACTH secretion in a similar fashion to that described for GH secretion in cranially irradiated children (1). The only detailed study of this nature involved 24-h cortisol and ACTH profiling of children in various stages of pubertal development who had received low-dose (18–24 Gy) cranial irradiation; intact dynamics of ACTH and cortisol secretion with no evidence of neurosecretory dysfunction were demonstrated (15).
The primary aim of the study was to describe the dynamics of cortisol secretion in adult cancer survivors who had relatively more intensive irradiation to identify any quantitative or qualitative disturbances in the secretory pattern in the presence of normal cortisol responses to the ITT.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients in this study were part of a larger cohort whose clinical details were described in a previous study (16). All had a history of whole brain irradiation for leukemia or a brain tumor anatomically distinct from the h-p region and had been shown to be free from tumor recurrence or any other medical condition that might influence their h-p function. Thirty-four adult patients with normal peak cortisol responses (>500 nmol/liter) to the ITT had undergone 24-h profiling and, thus, were suitable for this study (Table 1
). Their body mass indexes (BMIs) and ages are shown in Table 2, and the biological effective dose of radiation (BED) was calculated (median, 58.3 Gy; range, 23–106.4 Gy) for comparative purposes (16).
|
|
Two patients were receiving T4 therapy for primary hypothyroidism, and eight males had sustained variable degrees of direct testicular damage, evidenced by abnormally elevated gonadotropin levels and attributable to chemotherapy and/or testicular irradiation, of whom five had normal testosterone levels and the other three were receiving testosterone replacement therapy. None of the patients had received GH therapy for at least 2 yr before testing.
The control group consisted of 33 healthy sex-, age-, and BMI-matched subjects (Table 2
); female patients, however, had higher BMI compared with normal females (26 ± 2.2 vs. 23.5 ± 0.5 kg/m2). All controls were previously shown to have normal ACTH and GH reserve in response to the ITT (16, 17).
Study procedures
The study was approved by the South Manchester local research ethics committee, and informed consent was obtained from all subjects. Subjects were profiled at 20-min intervals between 0900 h and 0840 h the next morning as previously described (17). Female subjects were profiled in the first half of their menstrual cycles, and none had taken any oral contraception for at least 6 months before the study.
For the fasting profiles, subjects were asked to start fasting from 0000 h (midnight) before their admission at 0800 h. Sampling was undertaken as described above (except for fasting, which was strictly supervised on the ward) until 0900 h next morning. Basal and postfasting serum and fluoride oxalate samples were taken for analysis of free fatty acid (FFA), 3-hydroxybutyrate (3-HOB), bilirubin, and glucose. Sera were then separated and immediately frozen at –80 C.
During the fasting profiles, a 24-h urine collection was successfully completed by 18 of 23 patients and by 15 of 17 normal controls who successfully undertook a fasting profile. Twenty-four-hour urine volumes were measured, and representative samples were taken and kept frozen at –80 C for urinary free cortisol (UFC) and urinary creatinine measurements.
Assays
Serum cortisol measurements were performed using the heterogeneous competitive magnetic separation assay on the Bayer Immuno 1 System (Bayer Corp., Pittsburgh, PA). The sensitivity of this method is 5.52 nmol/liter (or 0.2 µg/dl; 1 µg/dl = 27.6 nmol/liter). The inter- and the intraassay coefficients of variations are 4.7–7.9 and 3.4–5.2% for cortisol concentrations between 920 and 88 nmol/liter, respectively. All profile samples for one study subject were analyzed in the same assay run to eliminate interassay variability. UFC was measured after extraction from urine using dichloromethane. The extract was then analyzed on the Bayer Immuno 1 Analyzer. ACTH was measured by the two-site immunoradiometric assay with a detection limit of 1 pmol/liter. FFA and 3-HOB were analyzed as previously described (18, 19).
Analysis of the cortisol profiles
Multiparameter deconvolution analysis (using PULSE 4 software provided by the Center for Biomathematical Technology, University of Virginia, Charlottesville, VA) was undertaken to resolve the whole cortisol concentration-time series into its constituent secretory components and simultaneously estimate endogenous cortisol half-life (20, 21) (Fig. 1). The estimated parameters in this analysis included cortisol half-life, basal secretion, secretory burst (pulse) frequency, mean burst area, mean burst amplitude, and mean intersecretory burst interval. The burst area represents the mass of cortisol released per burst (nanomoles per liter). The daily (24-h) pulsatile cortisol mass released (daily pulsatile cortisol production rate) is the product of burst frequency and the mean burst area in the whole profile (nanomoles per liter). The average pulsatile cortisol secretion rate (nanomoles per liter per minute) is calculated as the daily pulsatile cortisol production rate per 1440 min (duration of the profile). The total cortisol secretion rate is the sum of the pulsatile and basal secretion rates.
|
Cosinor analysis was used to appraise diurnal variations in cortisol concentration profile as previously described (17) (Fig. 2).
|
Approximate entropy (ApEn) was used to appraise the orderliness or the regularity of cortisol secretion over 24 h (22). A higher ApEn value indicates a greater secretory disorderliness (23). The ApEn calculation was conducted as described previously (17).
Statistics
The data were expressed as the mean ± SEM if normally distributed or as the median and range if the data were skewed. Simple correlations to examine the relationship between variables were carried out using the Spearman rank order correlation test. Differences between groups were examined by the t test if the data were normally distributed or by the Mann-Whitney rank-sum test if the data were skewed. Paired t test was used to examine differences in the normally distributed paired (fed and fast) data. One-way ANOVA was used to compare multiple independent groups. Statistical significance was accepted at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients and normal subjects had similar peak cortisol responses to the ITT (642.4 ± 21.5 vs. 629.5 ± 14.6 nmol/liter, respectively; P = 0.6). No gender differences were observed.
Patients had significantly higher nadir (lowest), maximum, and profile mean cortisol concentrations (increased by 14%; ranges, 16–86, 392–888, and 134.2–455.3 nmol/liter, respectively) than controls (Table 2), with none of the patients having values lower than the lowest value seen in normal subjects (ranges, 10–72, 335–696, and 111.6–274.1 nmol/liter, respectively; Fig. 3).
|
).
None of the patients had basal, pulsatile, or total cortisol secretion rate (0.03–0.53, 0.9–2.53, and 1.13–2.84 nmol/liter·min, respectively) lower than the lowest value seen in normal subjects (0.004–0.2, 0.9–2.5, and 0.98–2.65 nmol/liter·min, respectively; Fig. 3).
Fasting was confirmed objectively in each individual study subject and in the whole group by a significant fall in blood glucose in conjunction with increases in serum bilirubin, FFA, and 3-HOB (Table 3).
|
).
|
). The mean UFC in the fasting state was insignificantly higher in the patients (119 ± 16 vs. 104 ± 11 nmol/24 h; P = 0.5). The UFC/urinary creatinine ratio, which was used to correct for a suboptimal urine collection, was also increased in the patients compared with the normal subjects, and the difference almost reached statistical significance (12.72 ± 1.4 vs. 9.46 ± 0.9 nmol/mmol; P = 0.08). However, the fact that only 18 of the 23 patients had UFC estimation in the fasting state may have limited the significance level demonstrated in these comparisons.
No differences between the single measurements (median, ranges) of ACTH levels of the patients (17.8, 1.8–47.5 pmol/liter) and normal subjects (18, 12.2–51.3 pmol/liter) were seen at baseline (P = 0.3) or at the end of the fast (21.1, 7–46.6 vs. 21.5, 13.8–49.6; P = 0.6). In each group, the fasting ACTH distribution was similar to that at baseline (P > 0.05).
Effects of gender on cortisol secretion
No significant gender differences were observed in cortisol secretory dynamics in the normal subjects or the patient cohort (Table 5).
|
Thus, the differences (in the fed state) between male patients (n = 24) and normal males (n = 24) were marked, and even more so during fasting (n = 18 and n = 15, respectively); the mean cortisol concentration and total cortisol secretion rate were increased by 20 and 29% above the normal in the fed state and by 41 and 32% in the fasting state, respectively (Table 6).
|
Patients with normal GH status had higher mean cortisol concentrations and total cortisol secretion rates compared with those who had partial or severe GH deficiency (Fig. 4). The BED was slightly, but insignificantly, higher in GH-deficient patients (57 ± 5.3 vs. 48 ± 7.1 Gy; P = 0.3). The differences were amplified when normal females and female patients were excluded from the analysis (Fig. 5). Again, the BED was slightly, but insignificantly, higher in the GH-deficient male patients (56.8 ± 5.1 vs. 46.6 ± 5.9 Gy; P = 0.2) The BMI, age, age at irradiation, and the periods of follow-up were similar in all groups (P > 0.5).
|
|
Patients whose BED was less than 50 (35.2 ± 2.6) Gy and those with higher BED (64.5 ± 2.8 Gy) had similar mean cortisol concentrations (212.4 ± 10.5 vs. 212.3 ± 9.4 nmol/liter; P = 0.99) and total cortisol secretion rates (1.8 ± 0.11 vs. 1.76 ± 0.11 nmol/liter·min). The former group, however, had much longer postirradiation follow-up (15.8 ± 1.9 vs. 10.6 ± 1.6 yr; P = 0.04). The large span of the postirradiation interval (2–29 yr) limited the capacity of the study to examine the impact of BED alone.
Diurnal variation and secretory orderliness
The cortisol circadian rhythm was preserved in all patients during the fed and fasting states with no gender differences. The acrophase occurred between 0800 and 0900 h in almost all patients and all normal subjects. Fasting had no effect on the timing or amplitude of the diurnal rhythm (data not presented). ApEn values were similar in all groups, and no gender differences were observed in the fed or fasting states.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Contrary to the initial premise that radiation might cause CRH deficiency and, consequently, ACTH neurosecretory dysfunction with reduced cortisol production and cortisol levels, cranially irradiated patients who are not ACTH deficient, as defined by a pharmacological test, have, in fact, significant elevations in circulating cortisol production rates (by 20%) mirrored by significant increases in circulating cortisol levels (by 14%); this is a novel finding that has not been reported previously. In addition, all individual patients’ values were above the minimum values in normal subjects, thus excluding ACTH neurosecretory dysfunction.
Qualitatively, the pattern of cortisol secretion did not change; similar circadian rhythmicity, secretory burst frequency, intersecretory pulse interval, and secretory orderliness (ApEn values) were observed in both groups and all other subgroups during the fed and fasting states.
A substantial increase in cortisol secretion after more than 3 d of fasting has been described (24, 25, 26). In this study, short-term fasting did not appear to influence cortisol secretion significantly; however, it augmented the differences in mean cortisol concentrations and total cortisol secretion rates observed in the fed state to much higher values (35 and 27%, respectively), indicating a more responsive (proactive) h-p-adrenal axis in the patients.
The augmentation of cortisol levels and secretion appeared to mostly involve male patients. This apparent sexual dichotomy may be due to the attenuating effect of higher BMI in female patients on cortisol concentrations and production rates (27, 28). We could not reliably examine the impact of fasting in female subjects because of the small number of fasted female patients (n = 5) and fasted normal women (n = 2).
Although pulsatile ACTH secretion was not measured in this study, it would be reasonable to assume that the increase in cortisol production reflects actual activation and upward shift in the feedback control settings of the CRH-ACTH axis. Although cortisol secretion is not exclusively regulated by ACTH input, e.g. local neuropeptides can influence cortisol secretion as well as direct neuronal input by a multisynaptic pathway between the suprachiasmatic nucleus and the adrenal cortex (29), any increase in cortisol production and cortisol levels in our patients, related to possible changes in these mechanisms leading to increased adrenal responsiveness to ACTH or even non-ACTH-dependent stimulation of the adrenal cortex, cannot be sustained in the presence of normal CRH-ACTH axis activity and feedback control (settings). The best examples for dissociation between ACTH and cortisol secretion come from studies in women (30, 31, 32) and narcoleptic men (33), in whom increased adrenal responsiveness to ACTH is counterbalanced by a reduction in ACTH levels and, hence, maintenance of equivalent cortisol levels and secretion rates to those seen in normal men. Therefore, in the presence of normal (nonautonomous) adrenocortical function, increased cortisol secretion due to non-ACTH-dependent mechanisms is unsustainable without a proactive (stimulated) CRH-ACTH axis. In concert with this, all studies that demonstrated increased cortisol production in a variety of pathophysiological states have implicated underlying activation of the CRH-ACTH axis (34, 35, 36).
Although no differences were observed in single 0900 h ACTH measurements, the latter does not necessarily correlate with the actual 24-h ACTH secretion; augmentation of CRH-ACTH activity may well be related to changes in overall pulsatility pattern and possibly tonic secretion without significant changes in single random measurements. In addition, the apparent discrepancy between spontaneous and stimulated (during the ITT) cortisol secretion reflects the fact that the ITT is a ceiling test that usually explores maximum h-p-adrenal reserve and lacks sensitivity in characterizing or measuring (subtle) changes in axis activity under normal physiological circumstances.
Regulation of CRH secretion is complicated and poorly understood (34). Activation of the CRH-ACTH-adrenal axis has been implicated in the body response and adaptation to chronic stress of any nature: biological, physical, or psychological (36). Cancer survivors might be expected to be exposed to a much higher level of stress of a varying nature compared with their healthy peers.
Alternatively, activation of the h-p-adrenal axis could be the result of the molecular and cellular changes produced by radiation injury to the brain. Enhanced stimulatory input from suprahypothalamic centers is a plausible mechanism; however, enhanced CRH secretion is more likely to be mediated by a variety of cytokines and lipid mediators of some mild and chronic radiation-induced inflammatory process locally in the hypothalamus or elsewhere in the central nervous system. Although there is no direct evidence for radiation causing morphological inflammatory changes in the hypothalamus, radiation has been shown to produce inflammatory changes in other areas of the brain (37, 38, 39) or at least result in increased gene expression and production of a variety of inflammatory and proinflammatory cytokines, such as ILs, interferons, and TNFs (40, 41, 42, 43). Many of these inflammatory cytokines generated locally or delivered through the bloodstream, especially TNF-
and IL-6, have been shown to enhance CRH secretion (44, 45, 46, 47, 48). The activation of the CRH-ACTH-adrenal axis in the presence of local or systemic inflammation is believed to play a crucial role in preventing the overshoot of the inflammatory process and resulting tissue damage (44, 49, 50).
In addition to radiation causing direct neuronal cell death and degeneration, there is a belief that radiation may cause delayed brain tissue damage and dysfunction through chronic inflammation and/or enhanced release of proinflammatory cytokines (41, 51). In this context, corticotrope-adrenal axis activation could be quite beneficial in limiting the adverse effects of radiation on central nervous system tissues as well as the h-p, thus reducing the incidence and severity of long-term neuronal complications and hormonal deficits. In fact, the finding of higher cortisol secretion in patients without GH deficiency, i.e. without definitive h-p damage, may suggest that those who mount a stronger corticotrope-adrenal response are less likely to develop h-p damage. Alternatively, the lower cortisol secretion in patients with GH deficiency may suggest that the activation of the CRH-ACTH-adrenal axis is counterbalanced by some time-dependent damage in ACTH secretory status. Obviously, with more intensive irradiation and longer postirradiation periods, the extent of the damage to the CRH neurons and/or ACTH-secreting cells and the degree of the axis activation will determine the final outcome of the CRH-ACTH-cortisol secretory status.
There is ample evidence in animal as well as human studies of the inhibitory influence of the neurotransmitter 
-aminobutyric acid on CRH expression and release (52, 53, 54). Thus, radiation-induced reduction in -aminobutyric acidergic tone (55) leading to disinhibition of CRH release may provide an alternative or an additional mechanism for the long-term activation of the CRH-ACTH-adrenal axis.
The other key question raised by our findings relates to the impact of prolonged exposure to higher levels of cortisol on a number of biological end points. In particular, changes in body composition and the development of the metabolic syndrome and osteoporosis (35) cannot be ruled out, abnormalities that are frequently seen in cancer survivors (56, 57, 58).
From a clinical practice perspective, the findings of this study argue strongly against the suggestion that subnormal cortisol secretion under ordinary circumstances (ACTH neurosecretory dysfunction) might contribute to the excessive fatigue suffered by some cancer survivors and, thus, resolve the clinical dilemma as to whether cortisol replacement should be considered in such patients with normal cortisol responses to an ITT.
In conclusion, radiation-induced ACTH neurosecretory dysfunction does not exist, and activation of the corticotrope-adrenal axis occurs in adults who received irradiation of the h-p axis during treatment of nonpituitary brain tumors. In the absence of ACTH deficiency, this activation is manifested by a parallel increase in circulating cortisol levels and cortisol production rates without any change in cortisol half-life, overall secretory pattern, or diurnal rhythmicity. The chronic stress associated with the poor quality of life and long-term disabilities from cancer treatment may play a role in this phenomenon, in addition to the direct effects of radiation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online August 9, 2005
Abbreviations: ApEn, Approximate entropy; BED, biological effective dose of radiation; BMI, body mass index; CBG, cortisol-binding globulin; FFA, free fatty acid; 3-HOB, 3-hydroxybutyrate; h-p, hypothalamic-pituitary; ITT, insulin tolerance test; UFC, urinary free cortisol.
Received April 15, 2005.
Accepted June 30, 2005.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
injection in the common marmoset (Callithrix jacchus jacchus). Life Sci 68:1657–1665[CrossRef][Medline]
and IL-1ß mediate intercellular adhesion molecule-1 induction via microglia-astrocyte interaction in CNS radiation injury. J Neuroimmunol 95:95–106[CrossRef][Medline]
-aminobutyric acid on the secretion of corticotrophin-releasing factor-41 and arginine vasopressin from the rat hypothalamus in vitro. J Endocrinol 122:719–723-Aminobutyric acid inhibits ß-endorphin secretion from the anterior pituitary but not the neurointermediate lobe in the rat. Endocrinology 118:360–366-aminobutyric acid-ergic neurotransmission as a putative mechanism of radiation induced activation of the gonadotropin releasing-hormone-pulse generator leading to precocious puberty in female rats. Neurosci Lett 297:45–48[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  J. D. Veldhuis, F. Roelfsema, A. Iranmanesh, B. J. Carroll, D. M. Keenan, and S. M. Pincus Basal, Pulsatile, Entropic (Patterned), and Spiky (Staccato-like) Properties of ACTH Secretion: Impact of Age, Gender, and Body Mass Index J. Clin. Endocrinol. Metab., October 1, 2009; 94(10): 4045 - 4052. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Johannsson, R. Bergthorsdottir, A. G Nilsson, H. Lennernas, T. Hedner, and S. Skrtic Improving glucocorticoid replacement therapy using a novel modified-release hydrocortisone tablet: a pharmacokinetic study Eur. J. Endocrinol., July 1, 2009; 161(1): 119 - 130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Debono, R. J Ross, and J. Newell-Price Inadequacies of glucocorticoid replacement and improvements by physiological circadian therapy Eur. J. Endocrinol., May 1, 2009; 160(5): 719 - 729. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Debono, C. Ghobadi, A. Rostami-Hodjegan, H. Huatan, M. J. Campbell, J. Newell-Price, K. Darzy, D. P. Merke, W. Arlt, and R. J. Ross Modified-Release Hydrocortisone to Provide Circadian Cortisol Profiles J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1548 - 1554. [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 |
), normal subjects fed (
), and normal subjects fasted (
). Note the shift to the right and upward in the patients’ values.
