ORIGINAL E n d o c r i n e

ARTICLE R e s e a r c h

Hormone Secretion by Pituitary Adenomas Is Characterized by Increased Disorderliness and Spikiness but More Regular Pulsing Ferdinand Roelfsema, Alberto M. Pereira, Nienke R. Biermasz, and Johannes D. Veldhuis. Department of Endocrinology and Metabolic Diseases (F.R., A.M.P., N.R.B.), Leiden University Medical Center, 2333ZA Leiden, The Netherlands; and Endocrine Research Unit (J.D.V.), Mayo Medical and Graduate Schools, Clinical Translational Research Center, Mayo Clinic, Rochester, Minnesota 55901

Context: Hormone secretion by functioning pituitary tumors is characterized by increased basal (nonpulsatile) secretion, enhanced pulse frequency, amplified pulse mass, and increased disorderliness. Objective: The objective of the study was to quantify (subtle) abnormalities of hormone secretion by pituitary adenomas and the influence of selective pituitary surgery and suppressive medications on these parameters. Methods: Approximate entropy (ApEn) was quantified with a refined algorithm, spikiness by a new method to evaluate sudden short-lived increases in hormone levels, and pulsing regularity, determined with a fully automated deconvolution program. These 3 distinct measures of secretory disruption were compared in untreated and treated patients with acromegaly, prolactinoma, and Cushing’s disease together with matching profiles in healthy controls. Results: ApEn and spikiness were markedly increased in all untreated patient groups and normalized after pituitary surgery in acromegaly and hypercortisolism. In contrast, hormone-suppressive medical treatment in acromegaly and prolactinoma did not normalize ApEn. Spikiness normalized in acromegalic patients but not in prolactinoma. GH and cortisol pulsing regularity was elevated in acromegaly and Cushing’s disease, respectively, and normalized after surgery. Medical treatment caused normalization of pulsing regularity in acromegaly but not in prolactinoma patients. Conclusion: This study extends the understanding of disorganized hormone secretion by hyperfunctioning pituitary adenomas. The new findings are increased spikiness in all 3 tumor groups and increased pulsing regularity in GH- and ACTH-secreting adenomas. The mechanisms behind the marked pattern irregularity and the selective normalization by surgical and medical therapies are not established yet but may include diminished feedback signaling in addition to the anatomical and functional disorganization of intrapituitary cell networks. (J Clin Endocrinol Metab 99: 3836 –3844, 2014)

ituitary adenomas comprise GH-secreting adenomas, ACTH-secreting adenomas, prolactinoma, gonadotropinomas, TSH-secreting adenoma, and a heterogeneous group of nonfunctioning adenomas. Adenoma subtypes may secrete more than one hormone, whereas some

P

clinically nonfunctioning adenomas secrete ␣-subunits and/or ␤-subunits. Rarely, pituitary hyperplasia and adenoma are caused by tumor-related secretion of hypothalamic peptides, eg, CRH- and GHRH. Surgically removed adenoma fragments continue to (hyper)secrete hormones

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received May 15, 2014. Accepted July 1, 2014. First Published Online July 11, 2014

Abbreviations: ApEn, approximate entropy; CV, coefficient of variation; IRMA, immunoradiometric assay; PRL, prolactin.

3836

jcem.endojournals.org

J Clin Endocrinol Metab, October 2014, 99(10):3836 –3844

doi: 10.1210/jc.2014-2363

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-2363

in vitro without exogenous stimuli and are therefore believed to be autonomous (1–3). Nevertheless, both in vitro and in vivo pituitary adenomas often respond, in varying degrees, to stimulatory and inhibitory hormones, peptides, and drugs. Feedback on a particular adenoma type, eg, somatotropinoma, TSH-secreting adenoma, corticotropinoma, and gonadotropinoma, by hormones as T4, cortisol, IGF-1, and sex steroids is generally diminished, and aberrant responses to stimulating (hypothalamic) hormones are often observed (4). For the clinical endocrinologist, immediate and longterm consequences of hormonal hypersecretion are especially important along with mass effects related to adenoma location and size. Additional pathophysiological aspects of tumor include the distribution of secretory pulses, regularity of the underlying secretion patterns, and the occurrence of sudden sharp increases of hormone levels. Disruption of secretory processes may point to altered signal input to or reception by the adenoma and/or to intrapituitary cellular networks (5). Recent investigations have disclosed variable degrees of disorganized hormone secretion in GH-secreting adenomas, prolactinomas, corticotropinomas, and TSH-secreting adenomas, depending on the indices quantified (6 – 10). The purpose of this retrospective analysis is to investigate unique and common secretion abnormalities of more prevalent adenomas, using all 3 distinct measures of secretory disruption, viz. pulsing regularity, disorderliness, and spikiness. A major collar issue is whether irregularity parameters respond favorably to surgical or medical treatment using improved analytical tools Specifically, 10-minute sampled 24-hour data were submitted to each of the following: 1) deconvolution analysis to quantify pulsing irregularity (11–13); 2) approximate entropy (ApEn) to quantify sample-by-sample pattern orderliness (14 –16); and 3) spikiness estimation of sudden, sharp increases in hormone levels (17).

Materials and Methods Clinical protocol Patients and controls volunteered for and completed the sampling study without complications. Subjects originated from the same community and were evaluated in an identical sampling paradigm and hormone assays (below). Informed written consent was obtained from the subjects, and the study was approved by the Ethics Committee of the Leiden University Medical Centre. Data on number of subjects, sex, age, and body mass index are listed in Table 1. The 24-hour hormone secretion data are shown in the Supplemental Tables. Premenopausal women were studied in the early follicular phase of the menstrual cycle and postmenopausal women were not on estrogens.

jcem.endojournals.org

Table 1.

3837

Clinical Details of Patients and Controls.

Category Acromegaly Untreated Postoperative (short) Postoperative (long) SS analogs Controls Prolactinoma Untreated Dopamine analogs Controls Cushing’s disease Untreated Postoperative Controls

Number, M/F

BMI, kg/m2

Age, y

25 (16/9) 12 (5/7) 15 (7/8) 21 (16/5) 36 (17/19)

26.5 ⫾ 0.8 28.0 ⫾ 1.2 28.0 ⫾ 1.1 26.1 ⫾ 0.9 23.6 ⫾ 0.6

48.9 ⫾ 2.6 49.3 ⫾ 4.8 51.6 ⫾ 2.9 51.1 ⫾ 3.6 45.4 ⫾ 2.4

17 (6/11) 9 (5/4) 51 (28/23)

24.4 ⫾ 1.3 27.9 ⫾ 1.6 24.4 ⫾ 0.5

37.6 ⫾ 2.6 40.4 ⫾ 4.1 45.2 ⫾ 1.8

20 (7/13) 8 (3/5) 36 (19/17)

28.3 ⫾ 1.2 25.9 ⫾ 1.4 24.4 ⫾ 0.6

36.6 ⫾ 3.3 37.6 ⫾ 4.2 42.9 ⫾ 2.2

Abbreviations: BMI, body mass index; F, female; M, male; SS, somatostatin. Data are mean ⫾ SEM. Evaluation short after surgery in acromegaly took place after a median interval of 10 days (range 8 –19 d) and long after surgery, after 4.2 years (range 2.3–10.4 y). The median postsurgical time interval in Cushing patients was 7.0 years (range 4 –18 y). SS analogs were Sandostatin LAR (20 or 30 mg per 4 wk) in 13 patients and lanreotide Autogel (120 mg per 4 wk) in eight patients.

Participants maintained conventional work and sleeping patterns and reported no recent (within 10 d) transmeridian travel, weight change (⬎2 kg in 6 wk), shift work, psychosocial stress, prescription medication use, substance abuse, neuropsychiatric illness, or acute or chronic systemic disease. A complete medical history, physical examination, and screening biochemistry tests were normal. No subject had been exposed to glucocorticoids within the preceding 3 months. Volunteers were hospitalized the evening before the sampling studies. On the following morning, an iv cannula was inserted in a large vein of the forearm. Blood samples (2.0 mL) were withdrawn at 10-minute intervals for 24 hours beginning at 9:00 AM. A slow iv infusion of 0.9% NaCl and heparin (1 U/mL) was used to keep the line open. Ambulation was permitted to the lavatory only. Vigorous exercise, daytime sleep, snacks, and cigarette smoking were disallowed. Meals were provided at 8:00 AM, 12:30 PM, and 5:30 PM, and room lights were turned off between 10:00 PM and 12:00 AM, depending on individual sleeping habits. Blood was collected in prechilled siliconized tubes containing EDTA (ACTH) or heparin [GH, prolactin (PRL), and cortisol], centrifuged at 4°C to separate plasma, and frozen at ⫺20°C within 30 minutes of collection. Total blood loss was less than 360 mL. Volunteers were compensated for the time spent in the study. None of the 24-hour data has been published or presented previously or analyzed in the present manner.

Treatment Patients with Cushing’s disease (n ⫽ 20) and acromegaly (n ⫽ 27) were treated by pituitary surgery. Acromegalic patients underwent the 24-hour sampling after a median interval of 10 days (range 8 –19 d) or long after surgery with a median period of 4.2 years (range 2.3–10.4 y). Only 3 cases underwent both preoperative and postoperative blood sampling. Postsurgical evaluation of patients with Cushing’s disease took place after a median interval of 7 years (range 4 –18 y), when they had fully recovered

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

3838

Roelfsema et al

Secretory Abnormalities of Pituitary Adenomas

and did not use hormone replacement. Five of the patients were also sampled before surgery. All patients undergoing postoperative 24-hour blood sampling were in remission and in a clinical good condition. Remission criteria for acromegaly are normal age-related serum IGF-1 concentration and a glucose-suppressed serum GH below 0.5 mU/L (0.17 ␮g/L), in line with The Endocrine Society guidelines (18). Remission criteria for Cushing’s disease are 24-hour urinary free cortisol excretion less than 220 nmol and a 1-mg dexamethasone-suppressed serum cortisol concentration below 100 nmol/L (19). Patients with prolactinoma were treated with cabergoline. Five patients underwent repeat sampling studies. Clinical control was defined as normal PRL levels and reversal of hypogonadism. One patient, surgically treated for a giant prolactinoma, remained T and cortisol deficient, requiring replacement, but his PRL levels fell well into the normal range during cabergoline treatment. One female patient had borderline PRL levels (18 ␮g/L) during treatment, but her menstrual cycle was restored, and therefore, the dose was not increased. All patients showed a considerable size reduction of the adenoma or the postsurgical remnant (one patient) during cabergoline treatment. Biochemical control of acromegaly during somatostatin analog treatment was a normal serum age-related IGF-1 concentration (18). IGF-1 is given either as absolute values or as SD score for description of the different acromegaly groups.

Assays Serum GH concentrations were measured with a sensitive time-resolved fluoroimmunoassay (Wallac). The assay is specific for 22 kDa GH. The standard was recombinant human GH (Genotropin; Pharmacia & Upjohn), which was calibrated against the World Health Organization First International Reference Preparation 80/505 (to convert milliunits per liter to micrograms per liter, divide by 2.6). The limit of detection (defined as the value 2 SD above the mean value of the zero standard) was 0.03 mU/L (0.0115 ␮g/L). The intraassay coefficient of variation (CV) varied from 1.6% to 8.4% in the assay range 0.26 – 47 mU/L, with corresponding interassay CVs of 2.0%–9.9%. Serum PRL concentrations were measured with a sensitive timeresolved fluoroimmunoassay (Wallac). The limit of detection (defined as the value 2 SD above the mean value of the zero standard) was 0.04 ␮g/L. The assay was calibrated against the third World Health Organization standard 84/500. The intraassay CV varied from 3.0% to 5.2% in the assay range of 0.1–250 ␮g/L, with corresponding interassay CVs of 3.4%– 6.2%. Plasma ACTH was measured using a two-site sandwich assay designed to detect intact ACTH molecules. The immunoradiometric assay (IRMA) consisted of a soluble 125I-labeled (indicator) monoclonal antibody directed to the N terminus of ACTH as well as a second polyclonal ACTH antibody directed to the C terminus. The second antibody was covalently conjugated to biotin to react with avidin-coated plastic beads. All incubation reagents including antibodies, human ACTH-(1–39) standard and avidin-coated beads were from Nichols Institute (Allegro IRMA). Each sample was assayed in duplicate, and all samples from any one subject were assayed in the same run. Sensitivity of the IRMA was 1.0 ng/L or 0.22 pmol/mL, and intraassay precision was 3.2%–5.8% (range of median intrasample CVs in all individuals). Cross-reactivity with ␤-endorphin, TSH, LH, FSH, GH, or PRL was less than 0.1%. The plasma cortisol concentrations were measured by RIA (Sorin Biomedica). The detection

J Clin Endocrinol Metab, October 2014, 99(10):3836 –3844

limit of this assay was 25 nmol/L. The intra- and interassay precision varied from 2% to 4%. Screening serum estradiol concentrations were assayed by RIA (Diagnostic Systems Laboratory), T by coated-tube RIA (Diagnostic Products Corp) and free T4 by electrochemoluminescence immunoassay (Elecsys 2010; Roche Diagnostics). Serum IGF-1 concentration was measured with the Immulite 2500 system (Diagnostic Products Corp). The intraassay CV was 5.0% and 7.5% at the mean serum levels of 8 and 75 nmol/L, respectively.

Deconvolution analysis Hormone concentration time series were analyzed via a recently developed automated deconvolution method, empirically validated using hypothalamo-pituitary sampling, and simulated pulsatile time series (11–13). The Matlab-based algorithm first detrends the data and normalizes concentrations to the unit interval (0, 1). Second, the program creates multiple successive potential pulse-time sets, each containing one fewer burst via a smoothing process (a nonlinear adaptation of the heat diffusion equation). Third, a maximum-likelihood expectation estimation method computes all secretion and elimination parameters simultaneously conditional on each of the multiple candidate pulse-time sets. The outcome parameters (and units) are frequency (number of bursts per total sampling period, Weibull-␭ distribution), regularity of interpulse intervals (unitless Weibull␥), slow half-life (minutes), basal and pulsatile secretion rates (concentration units per session), mass secreted per burst (concentration units), and waveform shape (mode or time delay to maximal secretion after objectively estimated burst onset, in minutes).

Approximate entropy ApEn was used as a scale- and model-independent regularity statistic to quantify the orderliness or regularity of consecutive plasma hormone concentration measurements over 24 hours (15, 16). Normalized ApEn parameters of m ⫽ 1 (test range) and r ⫽ 20% (threshold) of the intraseries SD were used, as described previously. The ApEn metric evaluates the consistency of recurrent subordinate (nonpulsatile) patterns in the data and thus yields information distinct from and complementary to deconvolution (pulse) analyses.

Spikiness Spikiness was defined as the ratio of the SD of the first-differenced (incremental) hormone-concentration time series to the SD of the original series (17). Spikiness quantifies the extent of sharp, brief, staccato-like unpatterned fluctuations.

Statistical analysis Group comparisons were made by ANOVA, followed by calculating specified contrasts between groups. Interrelations between variables were analyzed with regression techniques. Logarithmic transformation of the data was used where required to decrease the variation. Data are given as mean and SEM. Although several patients underwent repeat studies, groups were considered as unrelated. Analyses used Systat, version 11 (SPSS Inc). P ⬍ .05 was considered significant.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-2363

jcem.endojournals.org

3839

Results Twenty-four hour hormone secretion In untreated patients, 24-hour hormone production was increased. In untreated acromegalics mean secretion was amplified almost 50-fold, and in prolactinoma patients 36-fold (P ⬍ .0001). ACTH secretion in Cushing’s disease was increased 6-fold, whereas cortisol secretion was amplified 4-fold (P ⬍ .0001) (Supplemental Tables 1-3). Shortly after surgery (median 10 d), GH secretion in acromegalic patients was still elevated (P ⫽ .01) but normalized in the long-term (median 4.2 y) after surgery (P ⫽ .46). IGF-1 levels returned to normal levels both shortly and long after pituitary surgery (Supplemental Table 1). PRL secretion in prolactinoma patients under dopaminergic treatment was normalized compared with healthy controls (P ⫽ .35). In contrast, acromegalic patients under somatostatin analog treatment had still raised GH and IGF-1 levels (P ⫽ .001 and P ⬍ .0001, respectively). Details are reported below. Secretory patterns in untreated patients (Supplemental Tables 1–3) ApEn of the hypersecreted hormones were greatly increased, irrespective of the underlying condition (all values P ⬍ .0001). The mean values are depicted in Figure 1. Spikiness was also increased in all patient groups (acromegaly, P ⫽ .002; prolactinoma, P ⬍ .0001; in Cushing’s disease ACTH, P ⫽ .037 and cortisol, P ⬍ .0001) with the mean values shown in Figure 2. Pulse regularity (Weibull-␥) was greatly increased in acromegaly, pointing to a more regular pulsing, but in prolactinoma patients, the tendency for more regular pulsing was not significant (P ⫽ .11). In Cushing’s disease cortisol pulsing was more regular (P ⫽ .048), but that of ACTH was not different from that of controls (P ⫽ .14, Figure 3). Secretory patterns after pituitary surgery Patients with acromegaly were studied either shortly after surgery (median 10 d) or later after surgery (median 4.2 y). Shortly after surgery, ApEn was lower, decreasing from a mean level of 1.323– 0.942 but still greatly elevated compared with controls (P ⬍ .001) (Figure 1). In contrast, spikiness and pulse regularity were already normalized shortly after surgery (Figures 2 and 3 and Supplemental Table 1). Long after surgery, spikiness and pulsing regularity remained normal, and ApEn was no longer different from controls. ApEn, spikiness, and pulsing regularity of both ACTH and cortisol secretion normalized after successful pituitary surgery in patients with Cushing’s disease (Figures 1–3 and Supplemental Table 3).

Figure 1. ApEn of the 24-hour serum (plasma) secretion profiles in patients and controls. Statistical differences with the control group are shown. C, healthy control subjects; MT, medically treated patients; TS, surgically treated patients; TSl, long after surgery; TSs, shortly after surgery; U, untreated patients.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

3840

Roelfsema et al

Secretory Abnormalities of Pituitary Adenomas

Figure 2. Spikiness of the 24-hour serum (plasma) secretion profiles in patients and controls. Statistical differences with the control group are shown. C, healthy control subjects; MT, medically treated patients; TS, surgically treated patients; TSl, long after surgery; TSs, shortly after surgery; U, untreated patients.

J Clin Endocrinol Metab, October 2014, 99(10):3836 –3844

Figure 3. Regularity of hormone pulses (Weibull-␥ distribution) detected by deconvolution of the 24-hour serum (plasma) secretion profiles in patients and controls. Statistical differences with the control group are shown. C, healthy control subjects; MT, medically treated patients; TS, surgically treated patients; TSl, long after surgery; TSs, shortly after surgery; U, untreated patients.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-2363

jcem.endojournals.org

3841

the IGF-1 SD score as a parameter, for Cushing’s disease the 24 hours cortisol secretion, and for prolactinoma also the 24-hour secretion. The latter two were logarithmically transformed before applying regression analyses. No significant relations were found between the severity of the illness and secretory pattern parameters (data not shown). Influence of tumor size All acromegalic patients had a macroadenoma (Hardy classification II to IIb but no microadenoma, thus making such an analysis impossible. The Cushing group contained six patients with a macroadenoma and 14 with a microadenoma. Eight patients had a macroprolactinoma and nine a microprolactinoma. Figure 4. Linear relations between ApEn and spikiness in patients with acromegaly (GH), patients with a prolactinoma (PRL), and patients with Cushing’s disease (ACTH and cortisol). Comparing ApEn, spikiness, and Weibull-␥ in these cohorts, no differSecretory patterns during medical treatment ences were found between patients with a macroadenoma Acromegaly and microadenoma (data not shown). Nine patients were controlled by somatostatin analog treatment. Their IGF-1 SD score was 0.37 (range ⫺0.45– Influence of gender 1.31). The remaining 12 patients were not controlled. The We explored the influence of gender on the three seIGF-1 SD score was 6.24 (range 3.0 –13.8). Seven of the cretion parameters. Only GH ApEn in the controls controlled patients had previous pituitary surgery, but showed a significant difference, being higher in women only one of the noncontrolled group did. Nevertheless, than men (0.547 ⫾ 0.037 vs 0.364 ⫾ 0.023, P ⫽ .003), but ApEn, spikiness, and Weibull-␥ were statistically similar not in untreated acromegalic patients (1.428 ⫾ 0.072 vs in both groups (P values between .62 and .86). Compared 1.263 ⫾ 0.098, P ⫽ .26) or in untreated Cushing’s disease with the control group, the ApEn of the combined groups and prolactinoma. Similarly, in all treated patient groups, of medically treated patients was greatly increased no ApEn gender differences were present. In addition, (1.017 ⫾ 0.076 vs 0.465 ⫾ 0.033, P ⬍ .0001), whereas spikiness and Weibull-␥ was sex independent in all patient spikiness and Weibull-␥ were not different (see Figures groups. 1–3 and Supplemental Table 1). Finally, IGF-1 SD score, ApEn, spikiness, and Weibull-␥ did not differ between Correlations between secretory parameters patients treated with octreotide LAR and lanreotide AuThere were highly significant relationships between togel (P values between .10 and .72). ApEn and spikiness in untreated patients, with squared correlation coefficients ranging from 0.48 to 0.81 (Figure Prolactinoma 4). In normal controls a positive correlation was present Patients with a prolactinoma and treated with dopa- only for prolactin (R2 ⫽ 0.57, P ⫽ .0001). minergic agonists had no previous treatment, except one patient. Patients on dopaminergic treatment had no improvement of ApEn, spikiness, and pulsing regularity (see Discussion Figures 1–3 and Supplemental Table 2), although PRL Key findings in this exploratory investigation were inlevels normalized. creased ApEn in all untreated pituitary adenomas and increased spikiness. Pituitary surgery in acromegaly deInfluence of disease severity We explored whether disease severity impacts on creased ApEn of GH in the short term and normalized it ApEn, spikiness, and Weibull-␥. For acromegaly we used in the long term. Comparable results were obtained in

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

3842

Roelfsema et al

Secretory Abnormalities of Pituitary Adenomas

patients surgically treated for Cushing’s disease. However, medical treatment with long-acting somatostatin analogs in acromegaly, even when the disease was controlled, and dopaminergic drugs in prolactinoma, leading to normal PRL levels, failed to normalize ApEn and spikiness. In the normally regulated hypothalamo-pituitary-end organ system, increased ApEn, denoting a less regular secretion pattern, can be induced by diminished feedback restraint, for instance, by blocking T secretion or diminishing IGF-1 secretion, leading to increased irregularity of LH and GH output (20). Other conditions with increased ApEn include aging and the well-known gender difference of GH ApEn (14, 21–25). On the other hand, artificially stimulating pituitary hormone secretion with infusions of GHRH, GH secretagogues and TRH also induces more irregular hormone output (26). Spikiness quantifies sudden and abrupt short-lived increases in hormone level, not attributable to assay artifacts, possibly reflecting pituitary, nonhypothalamic events. Increased spikiness has been described for ACTH in aging, especially in women, GH secretion in women compared with men, and PRL in men older than 50 years (27–31). As inferred from ApEn, feedback is diminished in hormone-secreting pituitary adenomas, and this phenomenon is exploited in the clinical diagnosis. However, faulty feedback is probably not the (only) explanation for the raised ApEn. Suppressing hormone secretion by increasing negative feedback, thus by somatostatin analogs in GH- or TSH-secreting tumors or dopamine agonists in prolactinoma does not restore secretion disorderliness. In contrast, curative pituitary surgery did restore the abnormal secretion patterns. Our patient and control group consisted typically of middle-aged subjects. At this age, there is a striking difference in GH ApEn between men and women (22, 32). Interestingly, this gender difference was lost in patients with acromegaly. Collectively, the present results suggest that the subtle hormone secretory abnormalities are tumor-related events and not the result of abnormal external (from hypothalamic and or peripheral target organs origin) feedforward or feedback signaling acting on the adenoma. Spikiness is less well studied than ApEn. Recently we described increased ApEn and spikiness in patients harboring a TSH-secreting pituitary tumor. The present work emphasizes that this is a general phenomenon in other hormone-secreting pituitary adenomas as well, which normalizes after surgery in patients with Cushing’ disease and those with acromegaly. In contrast, medical therapy resulted in divergent results. In prolactinoma spikiness was unchanged, whereas in acromegaly normalization was noted. For unknown reasons, spikiness in active acromeg-

J Clin Endocrinol Metab, October 2014, 99(10):3836 –3844

aly is less elevated than in other adenoma types, and this may have facilitated normalization. Although ApEn and spikiness are conceptually different measures, there were highly significant correlations between the two variables in untreated patients. In controls such a relationship was found for prolactin but not for the other hormones. The significance of these findings at this time is not known. Pulsing regularity was increased in active acromegaly and for cortisol, but not ACTH, secretion in Cushing’s disease. At present, there is not a full explanation for these findings and divergent results, but it may be related to different extent of system isolation, leading to diminished input from extrapituitary systems. We also explored whether tumor size, severity of disease, or gender influenced any of the three secretory measures, but we could not demonstrate any relationship. Recently ACTH secretion was reported to be higher in men than women with Cushing’s disease, and in the present cohort such a gender difference was also noted for ACTH but not for cortisol (data not shown). Nevertheless, ApEn and the other secretory parameters were identical in men and women suffering from Cushing’s disease (33). The independence of the secretory characteristics from disease severity, adenoma size, and gender are consistent with the hypothesis that these measures in pituitary adenomas are driven by the adenoma per se. During the last decade, ideas on the pituitary cell organization have been revolutionized. The application of multiphoton imaging and long-working distance microscopy in vivo in mice has disclosed homotypic pituitary cell networks in close spatial association with an extensive fenestrated capillary network and a supporting folliculostellate cell network (34 –37). These pituitary cell organizations fulfill the principles of networks, showing robustness, modularity, connectivity, and coordination (38). Connectivity and coordination are realized, first, by cellcell contacts via intercellular gap junctions, cell surface proteins, and autocrine and paracrine interactions; second, by folliculostellate cells, which form an excitable network capable of transmitting long-range calcium signals; and third, by vascular activity, which not only regulates the oxygen supply to the metabolic active cells during the secretion process but also regulates the diffusion of the secreted hormones from the perivascular space into the circulation in a temporal and spatially coordinated way, resulting in distinct pulses required for proper functioning of the target organs. It is likely that the cellular organization in pituitary adenomas is different from the intermingled networks of the normal pituitary, although this has not yet been studied in animals harboring pituitary tu-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/jc.2014-2363

mors. However, the vasculature is different, being less dense and the capillaries less fenestrated (39). A potential drawback of this study is its retrospective design. Nevertheless, we believe that this approach is acceptable, first, considering this otherwise extremely costly project; second, the fact that the sampling procedure was similar during the whole research period; and third, that the same robust hormone assays were used in patients and healthy volunteers. Other limitations of the study are the absence of data on successfully operated prolactinoma patients, the potential influence of a nonsecreting pituitary adenoma on secretory parameters in the absence of pituitary failure, data on cyclic Cushing’s disease, and the influence of medical treatment with pasireotide in Cushing’s disease. In conclusion, the most striking features of pituitary adenomas are the increased basal (nonpulsatile) secretion and the presently described increased disorderliness and spikiness. In a less well-organized, hormone-secreting network, these characteristics are to be expected. Interestingly, they were still present in medically treated patients. At present it is not known whether such (subtle) abnormalities are harmful in the long term. Recently we established that acromegalic patients on somatostatin drugs were less well off than surgically treated patients with respect to progression of acromegalic arthropathy, irrespective of the IGF-1 level (40). These observations exemplify the necessity of unraveling the pathophysiology of hormone secretion characteristics in pituitary adenomas.

Acknowledgments Address all correspondence and requests for reprints to: Ferdinand Roelfsema MD, PhD, Department of Endocrinology and Metabolic Diseases, Leiden University Medical Center, Albinusdreef 2, 2333ZA Leiden, The Netherlands. E-mail: [email protected]. This work was supported in part by the National Institutes of Health Grants AG19596 and AG29362 (to J.D.V.). Disclosure Summary: The authors have nothing to disclose.

References 1. Tulipano G, Faggi L, Losa M, et al. Effects of AMPK activation and combined treatment with AMPK activators and somatostatin on hormone secretion and cell growth in cultured GH-secreting pituitary tumor cells. Mol Cell Endocrinol. 2013;365(2):197–206. 2. Pecori GF, Bucciarelli LG, Saccani A, et al. Ghrelin stimulates adrenocorticotrophic hormone (ACTH) secretion by human ACTHsecreting pituitary adenomas in vitro. J Neuroendocrinol. 2007; 19(3):208 –212. 3. Hanson PL, Aylwin SJ, Monson JP, Burrin JM. FSH secretion predominates in vivo and in vitro in patients with non-functioning pituitary adenomas. Eur J Endocrinol. 2005;152(3):363–370.

jcem.endojournals.org

3843

4. Melmed S. The Pituitary. 3rd ed. London: Academic Press; 2013. 5. Hodson DJ, Molino F, Fontanaud P, Bonnefont X, Mollard P. Investigating and modelling pituitary endocrine network function. J Neuroendocrinol. 2010;22(12):1217–1225. 6. Veldman RG, Frolich M, Pincus SM, Veldhuis JD, Roelfsema F. Basal, pulsatile, entropic, and 24-hour rhythmic features of secondary hyperprolactinemia due to functional pituitary stalk disconnection mimic tumoral (primary) hyperprolactinemia. J Clin Endocrinol Metab. 2001;86(4):1562–1567. 7. van den Berg G, Frolich M, Veldhuis JD, Roelfsema F. Combined amplification of the pulsatile and basal modes of adrenocorticotropin and cortisol secretion in patients with Cushing’s disease: evidence for decreased responsiveness of the adrenal glands. J Clin Endocrinol Metab. 1995;80(12):3750 –3757. 8. van den Berg G, Pincus SM, Veldhuis JD, Frolich M, Roelfsema F. Greater disorderliness of ACTH and cortisol release accompanies pituitary-dependent Cushing’s disease. Eur J Endocrinol. 1997; 136(4):394 – 400. 9. Roelfsema F, Kok S, Kok P, et al. Pituitary-hormone secretion by thyrotropinomas. Pituitary. 2009;12(3):200 –210. 10. Roelfsema F, Keenan DM, Veldhuis JD. Endogenous ACTH concentration-cortisol secretion dose analysis unmasks decreased ACTH potency in Cushing’s disease with restoration after successful pituitary adenomectomy. J Clin Endocrinol Metab. 2011;96(12): 3768 –3774. 11. Liu PY, Keenan DM, Kok P, Padmanabhan V, O’Byrne KT, Veldhuis JD. Sensitivity and specificity of pulse detection using a new deconvolution method. Am J Physiol Endocrinol Metab. 2009; 297(2):E538 –E544. 12. Keenan DM, Chattopadhyay S, Veldhuis JD. Composite model of time-varying appearance and disappearance of neurohormone pulse signals in blood. J Theor Biol. 2005;236(3):242–255. 13. Keenan DM, Roelfsema F, Biermasz N, Veldhuis JD. Physiological control of pituitary hormone secretory-burst mass, frequency, and waveform: a statistical formulation and analysis. Am J Physiol Regul Integr Comp Physiol. 2003;285(3):R664 –R673. 14. Pincus SM. Orderliness of hormone release. Novartis Found Symp. 2000;227:82–96. 15. Pincus SM, Keefe DL. Quantification of hormone pulsatility via an approximate entropy algorithm. Am J Physiol. 1992;262(5 Pt 1): E741–E754. 16. Pincus SM. Approximate entropy as a measure of system complexity. Proc Natl Acad Sci USA. 1991;88(6):2297–2301. 17. Pincus SM, Schmidt PJ, Palladino-Negro P, Rubinow DR. Differentiation of women with premenstrual dysphoric disorder, recurrent brief depression, and healthy controls by daily mood rating dynamics. J Psychiatr Res. 2008;42(5):337–347. 18. Giustina A, Chanson P, Bronstein MD, et al. A consensus on criteria for cure of acromegaly. J Clin Endocrinol Metab. 2010;95(7):3141– 3148. 19. Tiemensma J, Kokshoorn NE, Biermasz NR, et al. Subtle cognitive impairments in patients with long-term cure of Cushing’s disease. J Clin Endocrinol Metab. 2010;95(6):2699 –2714. 20. Veldhuis JD, Straume M, Iranmanesh A, et al. Secretory process regularity monitors neuroendocrine feedback and feedforward signaling strength in humans. Am J Physiol Regul Integr Comp Physiol. 2001;280(3):R721–R729. 21. Pincus SM, Mulligan T, Iranmanesh A, Gheorghiu S, Godschalk M, Veldhuis JD. Older males secrete luteinizing hormone and testosterone more irregularly, and jointly more asynchronously, than younger males. Proc Natl Acad Sci USA. 1996;93(24):14100 – 14105. 22. van den Berg G, Veldhuis JD, Frolich M, Roelfsema F. An amplitude-specific divergence in the pulsatile mode of growth hormone (GH) secretion underlies the gender difference in mean GH concentrations in men and premenopausal women. J Clin Endocrinol Metab. 1996;81(7):2460 –2467. 23. Veldhuis JD. Altered pulsatile and coordinate secretion of pituitary

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.

3844

24.

25.

26.

27.

28.

29.

30.

31.

Roelfsema et al

Secretory Abnormalities of Pituitary Adenomas

hormones in aging: evidence of feedback disruption. Aging (Milano). 1997;9(suppl 4):19 –20. Veldhuis JD, Bowers CY. Three-peptide control of pulsatile and entropic feedback-sensitive modes of growth hormone secretion: modulation by estrogen and aromatizable androgen. J Pediatr Endocrinol Metab. 2003;16(suppl 3):587– 605. Veldhuis JD, Keenan DM, Liu PY, Iranmanesh A, Takahashi PY, Nehra AX. The aging male hypothalamic-pituitary-gonadal axis: pulsatility and feedback. Mol Cell Endocrinol. 2009;299(1):14 –22. Van den Berghe G, de Zegher F, Baxter RC, et al. Neuroendocrinology of prolonged critical illness: effects of exogenous thyrotropin-releasing hormone and its combination with growth hormone secretagogues. J Clin Endocrinol Metab. 1998;83(2):309 –319. Roelfsema F, Pijl H, Keenan DM, Veldhuis JD. Prolactin secretion in healthy adults is determined by gender, age and body mass index. PLoS One. 2012;7(2):e31305. Veldhuis JD. Aging and hormones of the hypothalamo-pituitary axis: gonadotropic axis in men and somatotropic axes in men and women. Ageing Res Rev. 2008;7(3):189 –208. Veldhuis JD, Roelfsema F, Iranmanesh A, Carroll BJ, Keenan DM, Pincus SM. Basal, pulsatile, entropic (patterned), and spiky (staccato-like) properties of ACTH secretion: impact of age, gender, and body mass index. J Clin Endocrinol Metab. 2009;94(10):4045– 4052. Veldhuis JD, Roelfsema F, Keenan DM, Pincus S. Gender, age, body mass index, and IGF-I individually and jointly determine distinct GH dynamics: analyses in one hundred healthy adults. J Clin Endocrinol Metab. 2011;96(1):115–121. Veldhuis JD, Iranmanesh A, Roelfsema F, et al. Tripartite control of

32.

33.

34.

35.

36.

37.

38. 39. 40.

J Clin Endocrinol Metab, October 2014, 99(10):3836 –3844

dynamic ACTH-cortisol dose responsiveness by age, body mass index, and gender in 111 healthy adults. J Clin Endocrinol Metab. 2011;96(9):2874 –2881. Pincus SM, Gevers EF, Robinson IC, et al. Females secrete growth hormone with more process irregularity than males in both humans and rats. Am J Physiol. 1996;270(1 Pt 1):E107–E115. Zilio M, Barbot M, Ceccato F, et al. Diagnosis and complications of Cushing’s disease: gender-related differences. Clin Endocrinol (Oxf). 2014;80(3):403– 410. Le Tissier PR, Hodson DJ, Lafont C, Fontanaud P, Schaeffer M, Mollard P. Anterior pituitary cell networks. Front Neuroendocrinol. 2012;33(3):252–266. Mollard P, Hodson DJ, Lafont C, Rizzoti K, Drouin J. A tridimensional view of pituitary development and function. Trends Endocrinol Metab. 2012;23(6):261–269. Fauquier T, Guerineau NC, McKinney RA, Bauer K, Mollard P. Folliculostellate cell network: a route for long-distance communication in the anterior pituitary. Proc Natl Acad Sci USA. 2001; 98(15):8891– 8896. Schaeffer M, Hodson DJ, Lafont C, Mollard P. Endocrine cells and blood vessels work in tandem to generate hormone pulses. J Mol Endocrinol. 2011;47(2):R59 –R66. Alon U. Network motifs: theory and experimental approaches. Nat Rev Genet. 2007;8(6):450 – 461. Turner HE, Harris AL, Melmed S, Wass JA. Angiogenesis in endocrine tumors. Endocr Rev. 2003;24(5):600 – 632. Claessen KM, Ramautar SR, Pereira AM, et al. Progression of acromegalic arthropathy despite long-term biochemical control: a prospective, radiological study. Eur J Endocrinol. 2012;167(2):235– 244.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 January 2017. at 07:10 For personal use only. No other uses without permission. . All rights reserved.