| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinological Oncology |
Service d’Endocrinologie et des Maladies de la Reproduction (P.C., J.Y., B.C., G.S.), Hôpital de Bicêtre, 94275 Le Kremlin-Bicêtre; and Département de Biologie Clinique (J.P., J.M.B.), Institut Gustave-Roussy, 94805 Villejuif, France
Address correspondence and requests for reprints to: Philippe Chanson, M.D., Service d’Endocrinologie et des Maladies de la Reproduction, CHU Bicêtre, 78 rue du Général Leclerc, F94275 Le Kremlin, Bicêtre Cedex, France.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
-subunit. It has been claimed that the paradoxical free LHß
response to TRH may be a useful clinical tool for determining the
gonadotropic nature of NFPA. We used a very specific and sensitive
immunoradiometric assay (IRMA) for free LHß measurement and another
specific IRMA to check the absence of free CGß, to study normal
subjects and 26 patients with NFPA. Basal plasma levels of LHß were
undetectable in normal men and premenopausal women in the early
follicular phase. In contrast, normal postmenopausal women had
increased basal plasma LHß, parallel to dimeric LH and -subunit
levels. In healthy subjects, stimulation with GnRH elicited an increase
in LHß while TRH was ineffective. In patients with NFPA, LHß
hypersecretion was found basally and/or after stimulation with TRH in 3
of 16 men, 3 of 5 premenopausal women, and 1 of 5 postmenopausal women,
i.e. 7 of 26 patients (26%). In 3 of these 7 cases,
-subunit and/or FSH levels were also increased. The LHß
measurement was thus truly informative on the gonadotropic nature of
NFPA in only 4 out of 26 cases (15%). In addition, increased LHß
levels and/or a positive response of free LHß to TRH was observed in
3 patients with pure prolactinomas but in no patients with GH-secreting
adenomas. Thus, using this very sensitive and specific IRMA, free LHß
measurement is rarely helpful for determining the gonadotropic nature
of NFPA. |
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
-subunit noncovalently linked to a specific ß-subunit; their
normal production requires coordinated synthesis, processing, and
dimerization of the two subunits. Particular attention has been paid to
imbalanced production of gonadotropin subunits as potential biochemical
markers of NFPA (5). Several studies have reported the clinical value
of detecting either the free -subunit (1, 6) or free LHß in
patients with NFPA (7). Furthermore, it is claimed that the paradoxical
increase in free LHß in response to TRH is frequent and might be the
most useful marker for gonadotropic adenomas in patients with NFPA (1).
However, specific detection of so-called free LHß remains difficult.
Numerous different forms of gonadotropins able to cross-react in
immunoassays circulate in plasma. Polyclonal antibodies (7) are
unsuitable for the detection of the free LHß as they cross-react at
variable degrees with dimeric LH. Monoclonal antibodies raised against
LH and capable of binding to antigenic sites present on the ß-subunit
are not specific for free LHß, but they also recognize dimeric LH. To
circumvent this problem, some authors have extracted -subunit and
intact LH before measuring free ß-subunit (8). The recent isolation
of a sulfated pituitary form of hCG possessing a ß-subunit highly
homologous to LHß also raises the possibility of interference with a
putative circulating free form of CGß (9, 10). Thus, careful
identification of epitopes is mandatory for the design of immunoassays
used in the specific detection of each molecular form of
gonadotropins. The aim of this study was to accurately determine the production of the free ß-subunit of LH in normal subjects and in patients with pituitary tumors. Two specific (i.e. non-crossreactive with the intact hormones) and sensitive immunoradiometric assays (IRMAs), one for the free form of LHß (11), the other for the free form of CGß (12, 13, 14) were used simultaneously.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Control subjects. Nine healthy men (45.8 ± 17.2 yr), nine healthy premenopausal women (32.6 ± 11.6 yr), and nine healthy postmenopausal women (52 ± 12 yr) were studied. All the premenopausal women were studied during the early follicular phase.
Patients with pituitary adenomas. Twenty-six patients (5
premenopausal women, 16 men, and 5 postmenopausal women) with
clinically nonfunctioning pituitary macroadenomas were studied. Some of
the patients had been operated on but had a residual tumor on
examination by magnetic resonance imaging, or they experienced a
recurrence at the time of the investigation. Clinical, biochemical, and
tumoral immunocytochemical characteristics of the patients are given in
Tables 1
, 2, and 3.
Twenty-one acromegalic patients and 20 patients with a PRL-secreting
adenoma were also studied [five men and 15 premenopausal women with a
macroadenoma (n = 15) or a microadenoma (n = 5)]. All the
patients gave their informed consent.
|
|
|
LH and FSH were measured using commercial kits (15, 16). Free
-subunit was measured with an immunoradiometric assay (IRMA) using
two monoclonal antibodies (Immunotech, Marseille, France).
Cross-reactivity of this immunoassay is less than 0.1% for dimeric
hormones (including CG, LH, FSH, and TSH) and 0% for the free
ß-subunit of these hormones. Results are expressed as international
units per liter (IU/L) of Medical Research Council 75/569, and the
detection limit is 0.025 IU/L. Within run and coefficients of variation
are respectively 6% and 12% at a concentration of 0.3 IU/L, and 3%
and 5% at a concentration of 2 IU/L. Normal ranges for basal plasma
free -subunit levels were 0.9–1.9 IU/L in postmenopausal women,
0–0.6 IU/L in premenopausal women during the early follicular phase,
and 0–0.3 IU/L in men.
To detect free LHß, we developed a two-site "sandwich" IRMA based on a monoclonal antibody (mAb BLH01) as the capture antibody. MAb BLH01 was selected for its specificity for free LHß, and the design of the assay has been previously reported elsewhere (11). Purified LHß (AFP-3282B), kind gift from the National Institute of Diabetes & Digestive & Kidney Diseases (NIDDK) and National Hormone and Pituitary Program (NHPP), was used as the standard. The sensitivity of this assay is defined by the least-detectable concentration, i.e. the free LHß concentration resulting in an increase (in counts per min bound) that was 2 SD higher than the mean in 20 replicates, and was found 20 ng/L. The within-run coefficient of variation, determined by assaying 20 replicates of a sample containing 500 ng/L of free LHß, was 6.5%. The between-run coefficient of variation determined by assaying a serum (800 ng/L) ten times, was 7.6%. We also performed a recovery test on one mix using equal volumes of samples and standards containing increasing levels of free LHß (0, 125, 250, 500, 1250, 2500, and 5000 ng/L). Regression analysis confirmed that the free LHß subunit concentration, when corrected for dilution, was not significantly affected by the dilution factor (r = 0.99). The relative cross-reactivity with LH (1.5%) is the result not of the recognition of LH by BLH01 but of cross-contamination of LH material by free LHß (11). However, this m-IRMA displays a cross-reactivity of about 65% with the free CGß. Thus, to distinguish between free LHß and free CGß, we used a second immunoassay specific for free CGß and based on the mAb FBT11 (17). This m-IRMA has a sensitivity of 25 ng/L for free CGß and cross-reactivity of less than 0.2% with free LHß.
Immunohistochemistry of the tumors
Tumors were fixed in Gerard’s fluid and embedded in paraffin wax. Immunocytochemical studies of the tumor samples were done as previously described (18).
Test procedures
Normal subjects and patients were all studied in the morning, recumbent, after an overnight fast. Blood was collected 15 and 0 min before, and 15, 30, 60, and 120 min after iv injection of 200 µg TRH. Plasma was stored at -20 C until use. The GnRH-test was performed on a separate day. Blood was collected 15 and 0 min before, and 15, 30, 60, and 120 min after iv injection of 100 µg GnRH.
Data analysis
Data are presented as means ± SD. A nonparametric ANOVA test was used when necessary to detect the differences between two means. P < 0.05 was considered to indicate significance.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Normal men and premenopausal women. Data for men and
premenopausal women were similar and were pooled for analysis. All the
males or premenopausal females in the early follicular phase had basal
plasma LHß and CGß levels below the detection limit of the assays.
GnRH administration produced a significant increase in -subunit and
LHß levels (peak range, 0.68–2.91 IU/L and 61–463 ng/L,
respectively), which was similar in both sexes. After TRH injection,
free -subunit levels did not change significantly (peak range,
0.20–0.68 IU/L), and the LHß level never increased above 39 ng/L.
Therefore a patient was considered to have an abnormal response to TRH
if levels of free LHß, when undetectable basally, exceeded 39 ng/L or
if free LHß increased by more than 100%. An abnormal response of
-subunit to TRH was defined by an increase above 0.68 IU/L.
Normal postmenopausal women. All the subjects had increased
basal plasma LHß levels, ranging from 29 to 148 ng/L (mean ±
SD, 76.7 ± 46 ng/L). -subunit basal levels were
also increased, ranging from 0.96 to 1.9 IU/L (mean ±
SD = 1.47 ± 0.32 IU/L). GnRH administration induced a
significant increase in -subunit and LHß in each patient. Peak
-subunit and LHß levels ranged from 2.2 to 6.3 IU/L (mean ±
SD, 4.47 ± 1.43 IU/L) and from 171 to 682 ng/L
(mean ± SD, 312 ± 164 ng/L), respectively.
Neither -subunit (peak range, 1.33–1.94 IU/L, mean ±
SD, 1.71 ± 0.32 ng/L) nor LHß levels (peak range,
30–177 ng/L, mean ± SD, 99 ± 49.6 ng/L)
significantly increase after TRH injection.
In the entire group of healthy subjects (n = 27), increments in
LHß, -subunit, and dimeric LH levels following GnRH administration
ran parallel (mean ± SD: 650 ± 440, 362 ±
203, 392 ± 216%, respectively).
Patients with nonfunctioning pituitary adenomas (NFPA)
Men with NFPA (Table 1
). Increased basal free
LHß levels were found in 3 (nos. M1, M2, M16) of the 16 men (62, 29,
and 186 ng/L, respectively). In 2 of these 3 men, basal levels of FSH
and -subunit were also moderately increased, while LH was normal
(nos. M2, M16). In 2 other cases (nos. M7, M12), free -subunit
levels were increased while LHß levels were normal. Following TRH
administration, the 3 men with supranormal basal free LHß levels
showed a further increase in free LHß to 154, 129, and 931 ng/L,
respectively, while an abnormal response of FSH or -subunit to TRH
was found in 1 (M2) and 2 patients (MM1, M2), respectively (Fig. 1).
|
). None
of the 5 premenopausal women with NFPA had an increased basal level of
free LHß. However, 3 women (nos. F2, F3, F5) with NFPA had an
abnormal response of free LHß to TRH (peak levels after TRH were 102,
111, and 160 ng/L, respectively). In one of these 3 patients, both
basal free -subunit and LH plasma levels were also increased. A
further increase in both LH and free -subunit was also observed
after TRH in this patient (no. F5, Fig. 1).
Individual responses to TRH in 6 patients with NFPA and supranormal
LHß levels. Figure 1 shows the individual responses of FSH, LH,
-subunit, and free LHß to TRH, expressed as the percentage
increase, in 6 of the 21 male and premenopausal female patients with
NFPA and supranormal LHß levels either basally or after TRH. In these
6 patients, immunocytochemical studies of the pituitary adenoma removed
at surgery proved the gonadotropic nature of the tumor (Tables 1 and 2). In all but one of the patients with NFPA, plasma free CGß was
unmeasurable either basally or after TRH. In this patient (no. M16),
free LHß hypersecretion was associated with an increase in free CGß
plasma levels both basally (34 ng/L) and after TRH (peak level, 164
ng/L); immunocytochemical studies of the excised adenoma revealed
positivity for FSHß, -subunit, and CGß.
Postmenopausal women with NFPA. None of the five
postmenopausal women with NFPA had a basal plasma free LHß level
higher than that of the normal men and premenopausal women. All the
patients had low plasma dimeric FSH and LH levels for postmenopausal
women, which may be indicative of gonadotropic failure (Table 3). Only one of these five patients had a response of
LHß to TRH, while responses of -subunit, LH, and FSH were normal
(PM5); but the peak level after TRH remained within normal
postmenopausal values.
Acromegalic patients
All the acromegalic men (n = 10) and premenopausal women (n = 6) had undetectable basal levels of free LHß and no response of free LHß to TRH. In contrast, GnRH produced an increase in free LHß level similar to that of healthy subjects (data not shown). Only one of the five postmenopausal acromegalic women had increased free LHß levels, which were in the range of healthy postmenopausal women (31 ng/L basally and 34 ng/L post-TRH). She also had increased dimeric LH and FSH levels, again in the normal postmenopausal range. The four remaining postmenopausal patients had gonadotropic failure.
Patients with prolactinoma
A single female patient, who had macroprolactinoma, had an
increased basal free LHß level (67 ng/L) and an increased response to
TRH (peak level, 161 ng/L). Her levels of -subunit, FSH, and LH were
normal both basally and after TRH. Two other female patients had normal
basal levels of free LHß but an increased response of free LHß to
TRH (peak levels 292 and 120 ng/L, respectively). In these two patients
basal -subunit levels were also supranormal (1.52 IU/L and 0.92
IU/L), while LH and FSH levels were normal. At transsphenoidal surgery,
pure PRL-secreting adenomas with negative immunostaining for
-subunit, LH, and FSH were removed.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
-subunit and, more recently, the free LHß response
to TRH have been proposed as biological markers of gonadotropic
adenomas before surgery. Indeed, increased plasma levels of free LHß
in response to TRH were the most common finding, compared with other
markers (FSH, LH, or free -subunit), in patients with NFPAs (7, 19, 20). These latter studies were based on a polyclonal assay that was not
totally specific for the free ß-subunit, as it presented a
cross-reactivity of 6.1% with intact LH. These results have recently
been challenged in a report where an indirect "two-step" method was
used to measure free LHß (8). In the first step both dimeric LH and
the free -subunit were sequestered by a monoclonal antibody directed
to the -subunit; secondly, residual LH binding activity was
attributed to free LHß (21). This assay procedure is too complex for
routine clinical use. In the present study we used a two-site
immunoassay based on a monoclonal antibody that specifically binds to
the free ß-subunit. This new assay directly measures free LHß.
However, the strong similarity (85%) between LHß and CGß means
that this assay cross-reacts with free CGß, implying simultaneous
determination of both free ß-subunits (22). To overcome this problem,
all the plasmas were tested for the absence of free CGß by using a
very specific IRMA for free CGß (12, 13, 14). Only one patient had
detectable plasma levels of both LHß and CGß. His pituitary
gonadotropic adenoma secreted , FSH, free LHß, and CGß, as shown
by immunocytochemistry. None of the other patients had circulating free
CGß.
Free LHß was only detected after stimulation with GnRH in the healthy
men and premenopausal women in early follicular phase. The amplitude of
the LHß response to GnRH was similar to that of -subunit and
dimeric LH. However, as previously reported with other assays, no
response of the LHß subunit to TRH stimulation was observed in
healthy subjects (7, 8, 19, 20, 23). Postmenopausal women had a
concomitant, parallel hypersecretion of dimeric LH, -subunit, and
free LHß subunit (8, 19), which was further stimulated by GnRH.
Dissociated secretion of the glycoprotein hormone subunits has been
observed in certain physiological and pharmacological circumstances
with the -subunit. Indeed, plasma levels of free -subunit
increase basally after the menopause (24) and are further stimulated by
GnRH (23, 25). In addition, during GnRH agonist treatment, -subunit
levels remain elevated even when dimeric LH levels are markedly
depressed (15). Thus, free LHß, like the -subunit, may be secreted
in dissociated form. This strongly argues that the synthesis of the
ß-subunit is not the limiting step in dimer secretion.
Most so-called NFPAs are gonadotropic adenomas (4, 26, 27, 28, 29, 30), but plasma
levels of gonadotropins and/or free -subunit are often normal, thus
hampering their clinical in vivo recognition. Some authors
have reported the value of measuring basal free LHß levels for the
identification of gonadotropic adenomas (19, 20). In contrast we found
that increased basal free LHß levels were rare. Three of the 21 male
and premenopausal female patients with NFPA had an increased basal free
LHß level, but -subunit and/or FSH basal levels were also above
the normal range in two cases. Thus, measurement of basal LHß levels
was truly informative in only one case. Moreover, while increased
-subunit levels were indicative of the gonadotropic nature of the
adenoma, LHß basal levels were normal in 3 other cases.
In the group of 16 male and 5 premenopausal female patients with NFPA,
an increase in plasma free LHß in response to TRH was observed in
only 6 cases (28%). Gil-del-Alamo et al. (8) reported
similar results, as only 33% of their patients with NFPA had an
abnormal response to TRH. These results contrast with reports of an
abnormal response of free LHß to TRH in 93% of men with NFPA (20).
It must also be emphasized that 3 of our 6 patients with positive LHß
responses to TRH had high basal -subunit and/or FSH levels. Thus, in
only 3 cases (14%) did LHß measurement alone identify the
gonadotropic nature of the NFPA. Daneshdoost et al. (19)
reported an abnormal response of free LHß to TRH in 68% of
postmenopausal women with NFPA. The present study does not confirm
these data, as a free LHß response to TRH was observed in only one
such patient. These discrepancies between studies may be explained by
the different immunoassays used and, particularly, to the potential
cross-reactivity of polyclonal antibodies with the various molecular
forms of gonadotropins and their subunits.
One of our patients with NFPA also had an increased plasma free CGß level, a situation previously reported by Gil-del-Alamo et al. (10) in 10% of patients with NFPA. The expression of CGß has been detected by in situ hybridization or immunocytochemistry in such pituitary adenomas (31, 32). All these data support the possibility that the biosynthesis of intact glycoproteins and free subunits by pituitary tumoral cells may be imbalanced.
As expected acromegalic patients never had increased plasma LHß
levels either basally or after LHß stimulation. In contrast, three
women with prolactinomas displayed an increased basal level and/or an
abnormal response of LHß to TRH, with a parallel increase in
-subunit levels. Their tumors, removed by surgery, showed exclusive
immunostaining for PRL. The significance of these results is not known,
but such an unexpected increase in serum-free glycoprotein subunits has
previously been reported for -subunit (33) and even for free CGß
(10) in rare patients with prolactinomas.
In conclusion, when measured with a new very sensitive and specific
immunoassay, plasma free LHß subunit levels, both basally and after
TRH stimulation, are rarely helpful in determining the gonadotropic
nature of NFPA. Its diagnostic value is similar to that of free
-subunit measurement. In addition, increased free LHß subunit
levels may be found in some patients with PRL-secreting adenomas.
Additional studies of larger series of patients with various types of
pituitary adenomas will be required to confirm the results obtained
with this new immunoassay for LHß.
|
Acknowledgments |
|---|
Received November 20, 1996.
Revised January 22, 1997.
Revised February 5, 1997.
Accepted February 12, 1997.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
-subunit hypersecretion in patients with
pituitary tumors: clinically nonfunctioning and somatotroph adenomas. J Clin Endocrinol Metab. 70:859–864.[Abstract]
-subunit in postmenopausal
women. Clin Endocrinol (Oxf). 34:477–483.[Medline]
-subunit-secreting tumors from acromegalic patients responsive to
octreotide. J Clin Endocrinol Metab. 79:1457–1464.[Abstract]
hCG, and ßhCG as measured by specific monoclonal
immunoradiometric assays. Endocrinology. 120:549–558.[Abstract]
-subunit secretion in prolactinomas and in
non-functioning adenomas: relation with the tumour size. Clin
Endocrinol (Oxf). 41:177–184.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  P. Chanson, F. Daujat, J. Young, A. Bellucci, M. Kujas, D. Doyon, and G. Schaison Normal Pituitary Hypertrophy as a Frequent Cause of Pituitary Incidentaloma: A Follow-Up Study J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 3009 - 3015. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Cedrin-Durnerin, J.-M. Bidart, P. Robert, J.-P. Wolf, M. Uzan, and J.-N. Hugues Consequences on gonadotrophin secretion of an early discontinuation of gonadotrophin-releasing hormone agonist administration in short-term protocol for in-vitro fertilization Hum. Reprod., May 1, 2000; 15(5): 1009 - 1014. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. de Roux, J. Young, S. Brailly-Tabard, M. Misrahi, E. Milgrom, and G. Schaison The Same Molecular Defects of the Gonadotropin-Releasing Hormone Receptor Determine a Variable Degree of Hypogonadism in Affected Kindred J. Clin. Endocrinol. Metab., February 1, 1999; 84(2): 567 - 572. [Abstract] [Full Text]
|
|
|
|
 |
|
 J. Pantel, P. Robert, F. Troalen, M. Kujas, D. Bellet, and J.-M. Bidart Characterization of Human Lutropin Carboxyl- Terminus Isoforms Endocrinology, February 1, 1998; 139(2): 527 - 533. [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 |
, patient M1; 
, M2; •, M16; 
, F2; 
,
F3; 
, F5). Note that the scales of the upper panel and of the lower
panel are different. Also note that in three of the six patients, basal