| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Articles: Hormones and Reproductive Health |
Department of Growth and Reproduction (M.S., A.-M.A., N.E.S., J.M.) and Pediatric Clinic II (M.S., K.S., H.H.), Juliane Marie Centre; Section of Radiation Biology, Department of Oncology (M.S., H.S.P.); and Department of Radiotherapy Physics (S.L.), Finsen Centre, The National University Hospital, Rigshospitalet, Copenhagen, Denmark DK-2100
Address all correspondence and requests for reprints to: Marianne Schmiegelow, M.D., Department of Growth and Reproduction 5064, Juliane Marie Centre, The National University Hospital, Rigshospitalet, Blegdamsvej 9, DK-2100, Copenhagen, Denmark. E-mail: rh04642{at}rh.dk
Abstract
The effect of radiotherapy (RT) and chemotherapy (CT) on gonadal function was assessed in males treated for a childhood brain tumor not directly involving the hypothalamus/pituitary (HP) axis in a population-based study with a long follow-up time. All males <15 yr at the time of diagnosis (median: 9.0 yr, range: 0.8 to14.9 yr) and diagnosed from January 1970 through February 1997 in the eastern part of Denmark and [gte]18 yr at the time of follow-up (median: 25.8 yr, range:18.5 to 39.3 yr) were included. Thirty males fulfilled the criteria. The median age at time of RT was 9.0 yr (range: 0.8 to 14.9 yr) and the median length of follow-up was 18 yr (range: 2.0 to 28.0 yr). The biological effective dose of RT was determined to the HP region and to the spine and expressed in gray because the biological effective dose gives a means of expressing the biological effect on normal tissue of different dosage schedules in a uniform way. Levels of serum FSH, luteinizing hormone (LH), sexual hormone-binding globulin, testosterone, and inhibin B were measured and compared with healthy age-matched male controls (n = 347), and the patients had a GnRH stimulation test performed with determination of peak FSH and LH.
Patients treated with RT + CT (n = 13), compared with patients treated with RT only (n = 17), had significantly higher median peak FSH (8.33 vs. 3.79 IU/L, P = 0.03) and median peak LH (20.0 vs. 12.8 IU/L, P = 0.03), and significantly lower median inhibin B (86.0 vs. 270 pg/ml, P = 0.03), and median inhibin B/FSH ratio (12.8 vs. 107.9, P = 0.04), which indicates gonadal damage. Inhibin B and inhibin B/FSH ratio were also significantly lower in the RT + CT group, compared with controls (median: 86.0 vs. 215 pg/ml, P = 0.02), (median:12.8 vs. 67; P = 0.01), respectively. We found a significantly inverse correlation between basal FSH and inhibin B and FSH and total testicular volume (rs = -0.83; P < 0.0001), (rs = -0.67; P < 0.0001), respectively, and a significant correlation between inhibin B and total testicular volume (rs = 0.63; P < 0.0001). Stepwise backward multiple linear regression analysis showed the best-fit model to predict inhibin B levels included total testicular volume (P = 0.002) and CT (P = 0.09). Median basal LH in the RT-only group was significantly lower, compared with controls (3.44 vs. 2.45 IU/L; P = 0.0001) indicating secondary hypogonadism, and in both the RT + CT group and the RT-only group, levels of testosterone were significantly lower, compared with our reference population (12.8 vs. 21.9 nmol/L; P = 0.001, and 14.7 vs. 21.9 nmol/L; P = 0.0003), respectively.
In conclusion these data suggest that cranial irradiation for a childhood brain tumor may affect the HP axis, and adjuvant CT can reduce inhibin B indicating primary gonadal damage. Thus, such patients may have normal or even low levels of FSH despite damage to the seminiferous epithelium, and because the fertility status by a semen analysis for psychological reasons can be difficult to obtain in this group of patients, we suggest inhibin B as the most useful direct serum marker of spermatogenesis in the follow-up of individuals who have received both cranial irradiation and gonadotoxic chemotherapy. However, because the number of patients with RT + CT and RT only are small, these data must be confirmed in further studies.
THE CURE RATE in childhood brain tumors has reached a level of about 60%. However, the improvement in prognosis has been achieved at the expense of serious late effects including the potential for severe gonadal damage (1). In previous studies of adults who suffered from childhood brain tumors, radiation doses >35 Gy to the hypothalamus/pituitary (HP) axis resulted in deficiency of GnRH because of damage of the hypothalamus or deficiency of FSH and luteinizing hormone (LH) because of damage of pituitary gonadotrophs resulting in secondary hypogonadism (2, 3).
Direct irradiation of the gonads and chemotherapy (CT) may result in primary gonadal dysfunction. Irradiation of the testes prepubertally with 24 Gy for a testicular relapse in acute lymphoblastic leukemia may lead to both germ cell depletion as well as delayed sexual maturation associated with decreased testosterone levels and increased LH levels (4, 5). The effect of irradiation of the testes of adult males has been shown to be dose dependent with exposures to the testes of as little as 0.75 to 6.0 Gy resulting in reduced sperm production with raised FSH and LH levels (6, 7). However, craniospinal (CSI) irradiation itself, as in children treated for a medulloblastoma, has been reported not to damage testicular function (8).
The gonadal function after treatment with CT has especially been investigated in relation to the chemoregimens for Hodgkin’s disease, in which alkylating agents have been associated with damage of the germinal cells of the testes with oligospermia or azoospermia and the effect of CT on Leydig cell function with lowered testosterone levels (9, 10, 11).
Generally, the evaluation of gonadal status with regard to spermatogenesis has been assessed either directly by semen samples or indirectly by determination of the levels of gonadotrophins and a clinical evaluation of possible reduced testicular volume. However, recently it has become possible to estimate the gonadal function of male survivors directly by measuring the serum levels of the bioactive gonadal peptide hormone inhibin B by means of a new enzyme-linked immunosorbent assay (12). Inhibin B is supposed to be produced by the Sertoli cells and the germ cells of the testes and directly reflects the degree of seminiferous tubular damage (13, 14). Inhibin B exerts negative feedback regulation of pituitary production and secretion of FSH (15).
To assess the effect of radiation therapy (RT) and CT on gonadal function in a population-based follow-up study of males treated for a brain tumor in childhood, we have measured inhibin B, gonadotrophins, testosterone, and total testicular volume.
Materials and Methods
Patients
The study included all male patients that were 1) diagnosed and treated for a brain tumor in the eastern part of Denmark including all counties of Sealand, the Faroe Islands, and Greenland, diagnosed from January 1970 through February 1997, 2) aged <15 yr at the time of diagnosis (median: 9.0 yr, range: 0.8 to14.9 yr) of a primary brain tumor not directly involving the HP axis, 3) treated with cranial or CSI RT and with or without CT, 4) >1 yr from end of treatment at time of follow-up, 5) [gte]18 yr at time of follow-up (median: 25.8 yr, range:18.5 to 39.3 yr), and 6) in remission at time of follow-up. A total of 31 males fulfilled these criteria. However, one of two patients, who was on testosterone treatment when included in the study, did not discontinue his medication, which is why he was excluded. Thus, we ended up with 30 patients. The median age at the time of RT was 9.0 yr (range: 0.8 to 14.9 yr) and the median length of follow-up was 18 yr (range: 2.0 to 28.0 yr).
The tumor diagnoses were astrocytoma (n = 12), medulloblastoma (n = 11), ependymoma (n = 4), glioma (n = 2), or nonhistologically verified diagnosis (n = 1). This latter patient was assumed to have an opticus glioma. According to the World Health Organization classification, 16 patients had a benign and 13 patients a malignant tumor (16). Ten patients had hydrocephalus at the time of diagnosis to such a degree that they had a shunt operation performed. Eight patients were taking other hormones, which included thyroxine (n = 7), hydrocortisone (n = 1), and GH (n = 4).
Treatment
Surgery. Twenty-nine children had a biopsy or a total or partial resection of the tumor performed in addition to RT.
Radiotherapy. Seven children had been treated on
60Co units and 23 children had been treated with
external conventional 4, 6, or 8 megavoltage RT delivered by a linear
accelerator. Fifteen children were treated with CSI, 7 with whole brain
irradiation only, and 8 children were treated with focal cranial
irradiation (Table 1
). The overall
treatment time differed according to different treatment schedules used
from 1970 to 1997 from 1 fraction every second day to 1 fraction per
day, five times a week, and the dose per fraction differed from an
increment strategy starting at very low doses increasing over the
treatment period to the administration of uniform fractions during the
whole treatment period.
|
Methods
Hormone analyses. Between 0800 h and 1000 h, blood
samples were drawn from an antecubital vein for measurement of levels
of serum of basal FSH, LH, testosterone, inhibin B, and sexual
hormone-binding globulin (SHBG). A GnRH stimulation test was carried
out according to a standard protocol: 0.1 mg GnRH (Relefact) was
administered iv and serum samples for FSH and LH were drawn at 0 and at
30 min and measured by time-resolved fluoroimmunometric assay
(AutoDelfia, Wallac, Inc., Turku, Finland) (18, 19). The sensitivities of the FSH and LH assays were 0.06 and
0.05 IU/L, respectively. The intra- and interassay coefficients of
variation were both <5% in the FSH and LH assays. SHBG was measured
by a two-sided fluoroimmunometric assay (AutoDelfia, Wallac, Inc.), and the detection limit for SHBG was 0.23 nmol/L, and the
intra- and interassay coefficients of variation was <6%. Testosterone
was measured by RIA (Coat-A-Count, Diagnostic Products,
Los Angeles, CA). The detection limit for testosterone was 0.23 nmol/L,
and the intra- and interassay coefficients of variation was 
10%.
Serum inhibin B was measured by an immunometric assay as previously
described (19, 20). The sensitivity of the inhibin B assay
was 18 pg/mL, and the intra- and interassay coefficients of variation
were 13% and 19%, respectively.
Biological effective dose. To get a uniform comparison of the different treatment regimens, we used the biologic effective dose (BED) of RT, which gives a means of comparing different RT schedules (21). The BED received by the HP region and the spine was determined by the linear quadratic model as previously described (22).
Statistics. Because the data for the patient population had a skewed distribution, we used median and range. The normative data used are based on an age range of 22 to 44 yr, which is similar to the cohort with an age range of 18.5 to 39.3 yr. The normal range for inhibin B, gonadotrophins, testosterone, SHBG, and total testicular volume is unchanged for individuals above 18 yr of age, which is why we did not use SD scores. Correlations between variables were tested with the nonparametric Spearman’s correlation analysis (rS = correlation coefficient) (23). We performed the Kruskal-Wallis test to analyze the variance by ranks comparing three unpaired groups.
For comparing data from two independent groups, the Mann-Whitney U test was used. The relationship between the independent variables total testicular volume, age at RT, time elapsed since RT and CT (given = 1, not given = 0), levels of testosterone, and spinal irradiation (given = 1, not given = 0), and the dependent variable inhibin B were analyzed by stepwise backward multiple regression analysis. The validity of the regression model was checked using standard tests. These included assessing the distribution of the residuals and testing for normality and checking the linearity assumptions in the model by means of standard scatter plots. Data analyses were done with the SPSS, Inc. (Chicago, IL) statistical software (24). Two-sided P values of <0.05 were regarded as significant.
Ethics. The patients gave their informed consent. The study is in accordance with the Helsinki II declaration and approved by the local ethical committee of Copenhagen, Denmark (approval 01–339/96.).
Results
The median and range of BED and endocrinological end points for
the 30 patients are summarized in Table 2.
For comparison, serum inhibin B levels, the inhibin B/FSH ratio, basal FSH
and LH levels, and levels of testosterone and SHBG were measured in normal
age-matched male volunteers (n = 347) (median age: 31 yr, range: 22
to 44 yr) and are also included in the table. The normative data for total
testicular volumes for adults are in the range of 40–50 mL (25, 26).
|
|
There was no significant difference between the group of patients who had been treated with RT + CT (n = 13), compared with the group of patients who had been treated with RT only (n = 17) with respect to median BED to the HP axis (80 vs. 78 Gy) but a significant difference with respect to median BED to the spine (51 vs. 50 Gy) (P = 0.02). However, there was no significant difference with respect to median age at RT (8.4 vs. 9.1 yr), or median time of follow-up (16.0 vs. 19.0 yr).
Patients who had received RT + CT, compared with those who had received
RT only, had no significant difference in median basal LH
(3.54 vs. 2.45 IU/L). The RT + CT group corresponded to
normal or near-normal levels in our reference population (3.54
vs. 3.44 IU/L); however, regarding the RT-only group, there
was a significantly lower basal LH, compared with controls (2.45
vs. 3.44 IU/L; P = 0.0001; Fig. 5). In addition, patients treated with RT
+ CT, compared with those treated with RT only, had a significantly
higher median peak LH response to the GnRH test (20.0 IU/L
vs. 12.8 IU/L; P = 0.03; Fig. 6). Furthermore, there was no significant
difference between patients treated with RT + CT, compared with
patients treated with RT only, in median testosterone (12.8
vs. 14.7 nmol/L); however, both the RT + CT group and the
RT-only group had significantly lower testosterone levels, compared
with our reference population (12.8 vs. 21.9 nmol/L;
P = 0.001, and 14.7 vs. 21.9 nmol/L;
P = 0.0003, respectively) (Fig. 7). Regarding median SHBG, we found a
significant difference in the RT + CT group, compared with the RT-only
group (21.0 vs. 37.0 nmol/L; P = 0.002), and
in the RT + CT group, compared with the reference population (21.0
vs. 30.0 nmol/L; P = 0.001) (data not
shown). There was no significant difference in median total testicular
volume (27.0 vs. 40.0 mL; P = 0.16) in the
RT + CT group, compared with the RT-only group, probably due to the
small sample size.
|
|
|
We found a significant inverse correlation between basal FSH and
inhibin B (rs = -0.83; P <
0.0001) (data not shown) and peak FSH and inhibin B
(rs = -0.84; P < 0.0001; Fig. 8). Furthermore, there was a significant
inverse correlation between FSH and total testicular volume
(rs = -0.67; P < 0.0001) (data
not shown) and a significant correlation between inhibin B and total
testicular volume (rs = 0.63; P
< 0.0001; Fig. 9). There was no
significant inverse relation between LH and testosterone
(rs = -0.19; P = 0.3) or between
peak LH and testosterone (rs = -0.31;
P = 0.09; Fig. 10).
|
|
|
).
|
In the present study, we found that the BED to the HP axis was
without significant difference in the RT + CT and the RT-only group,
indicating that the same degree of RT-induced central hypogonadism
would be expected in the two groups. Because no strict normative data
for the peak FSH and peak LH response to a GnRH provocative test exist,
we compared the two groups. We found a significantly higher peak FSH
response to the GnRH test in the group of patients who had been treated
with both RT and CT, compared with the group of patients who had been
treated with RT only. The results indicate a primary gonadal
dysfunction caused by damage of the Sertoli cells and germ cells of the
testes by CT. So even if a scatter dose from the spinal fields reached
the testes, the amount of irradiation was probably of minor importance.
This is in accordance with a former study with a small number of
patients with medulloblastoma, who had all been treated with either
CCNU or carmustine + CSI or CSI alone in which it was shown that CT
resulted in a primary gonadal dysfunction, but spinal irradiation had
no influence (8). Furthermore, the present study
demonstrated a lower inhibin B level and inhibin B/FSH ratio in the
group of patients who had been treated with RT + CT, compared with
those treated with RT only, and the stepwise backward multiple linear
regression analysis showed that the best-fit model to predict inhibin B
included CT in the model in contrast to spinal irradiation. Because
inhibin B is known to be produced mainly in the germinal epithelium of
the testis in the adult male (12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27), our results further
underline the fact that adjuvant CT causes damage of the Sertoli cells
and/or the germ cells resulting in reduced spermatogenesis and reduced
total testicular volume (Fig. 9
). Inhibin B gave us as a direct serum
marker the opportunity to assess the testicular damage without semen
analyses, which was not found to be feasible in this cohort of patients
because of the psychological impairment of the individuals. Several
studies have shown that inhibin B correlates with spermatogenesis
(27, 28, 29, 30). In a Danish study of 349 men (13),
it was concluded that the predictive power of detecting sperm counts
below 20 million/mL among normal men whose inhibin B and FSH both were
below 80 pg/mL and above 10 IU/L, respectively, was 100%
(14).
We found in the group that had been treated with RT + CT (n = 13),
a median inhibin B of 86 pg/mL and a median peak FSH of 8.33 IU/L, and
of the 13 patients, 6 had a serum inhibin B level < 80 pg/mL and
a peak FSH 
10 IU/L, which is why it is reasonable to assume a reduced
sperm count in our study population. However, further studies are
needed to determine a threshold level of inhibin B signifying permanent
infertility.
There was no significant difference between the RT + CT group and the RT-only group regarding the levels of testosterone. Because the levels of testosterone in both groups were significantly lower, compared with the controls, this probably indicated RT-induced central damage leading to decreased LH and thereby low serum testosterone. The fact that the GnRH test provoked significantly elevated levels of peak LH in the group of patients treated with RT + CT, compared with the group of patients who had been treated with RT only, may suggest subtle damage to the Leydig cells, and we speculate that the Leydig cells are sensitive but less sensitive to CT, compared with Sertoli cells and germ cells. This is consistent with other reports that CT has an influence on Leydig cell dysfunction (11, 31, 32). On the other hand, because SHBG levels were lower in the RT + CT group, it cannot be excluded that biologically active testosterone was higher in this group.
Little is known about the effects of CT on the HP axis, but some reports indicate that CT potentiates the deleterious effects of radiation on growth (33, 34). However, whether this suggested effect on GH-releasing hormone cells of the hypothalamus and/or the somatotrophs of the pituitary could also be expected of GnRH cells of the hypothalamus and/or the gonadotrophs of the pituitary we can only speculate. The fact that inhibin B is produced only in the testes underlines its superiority as measurement of FSH and LH in the evaluation of gonadal status in patients who have been treated with both RT and CT.
To our knowledge, this is the first population-based study of the gonadal status of survivors of a childhood brain tumor employing inhibin B as a useful marker for testicular damage. Inhibin B is of particular value because FSH may be lowered by the concomitant damage of the HP region by RT. FSH and LH have previously been used as indirect markers of spermatogenesis and Leydig cell function, respectively. However, when both the HP function and the primary gonadal function have been disturbed, the degree of central hypogonadism caused by cranial irradiation or CT is very difficult to interpret. The interaction between central and primary hypogonadism in the present study is probably also why BED was not significantly correlated to the endocrinological end points because other reports have shown significant dose-response correlations (22).
The present data showed strong significant correlations between testicular size and FSH and inhibin B, which confirms that inhibin B could be used as a direct marker of male gonadal status. In another report in which serum inhibin B was used in detecting male gonadal dysfunction following different treatment schedules for different childhood cancer survivors, it was indicated that inhibin B could be used earlier than testicular volume or semen analyses in the estimation of gonadal function (35).
In conclusion, these data suggest that cranial irradiation for a childhood brain tumor may affect the HP axis resulting in secondary hypogonadism, and adjuvant CT may have a deleterious affect upon the seminiferous epithelium of the testes leading to reduction in inhibin B and inhibin B/FSH ratio indicating primary gonadal damage. Thus, such patients may have normal or even low levels of FSH despite damage to the seminiferous epithelium, and because the fertility status by a semen analysis for psychological reasons can be difficult to obtain in this group of patients, we suggest inhibin B as the most useful direct serum marker of spermatogenesis in the follow-up of individuals who have received both cranial irradiation and gonadotoxic CT. However, because the number of patients with RT + CT and RT only are small, these data must be confirmed in further studies.
|
|
|
|
1 This work has financially been supported by The Danish Children
Cancer Foundation, The Gangsted Rasmussen Foundation (Grant 8436), The
Lundbeck Foundation (Grant 39/96), The Olga Boel Foundation, The
Rosalie Petersen Foundation (Grant 62036-3), The Ville Heise Foundation
(Grant M1-96), The Haensch Foundation, and The Fraenkel
Foundation. 
Received January 3, 2001.
Revised February 26, 2001.
Accepted March 12, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  H K Gleeson and S M Shalet The impact of cancer therapy on the endocrine system in survivors of childhood brain tumours Endocr. Relat. Cancer, December 1, 2004; 11(4): 589 - 602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Schmiegelow, U. Feldt-Rasmussen, A. K. Rasmussen, M. Lange, H. S. Poulsen, and J. Muller Assessment of the Hypothalamo-Pituitary-Adrenal Axis in Patients Treated with Radiotherapy and Chemotherapy for Childhood Brain Tumor J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 3149 - 3154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Schmiegelow, U. Feldt-Rasmussen, A. K. Rasmussen, H. S. Poulsen, and J. Muller A Population-Based Study of Thyroid Function after Radiotherapy and Chemotherapy for a Childhood Brain Tumor J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 136 - 140. [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 |
