- Email: [email protected]
Gonadal Function of Patients Treated With Cisplatin Based Chemotherapy for Germ Cell Cancer
0022-534719711583-0844$03.~10
Val. 158. 844-850. September 1997 Printed In U.S.A.
THEJOVRSAL OF UROUIGY Copyright @ 1997 by AMEKIC'AY UROLO(,ICAL L~%x'i~Tios. Isc.
GONADAL FUNCTION OF PATIENTS TREATED WITH CISPLATIN BASED CHEMOTHERAPY FOR GERM CELL CANCER W. B R E N N E M A " , B. STOFFEL-WAGNER, A. HELMERS, J. MEZGER, N. JAGER D. KLINGMULLER From the Departments of Clinical Biochemistry and Internal Medicine, Uniuersity of Bonn, and the Department Kliniken, Hildesheim, Gernianv
of
AND
Urology. Stadtrsche
ABSTRACT
Purpose: The cure rate of patients with germ cell cancer of the testis has considerably improved since the introduction of cisplatin based chemotherapy. Because these patients are in their reproductive years and because some of them will be infertile after treatment, the effects of cytotoxic treatment on gonadal function are investigated by hormonal evaluations. Materials and Methods: In a transversal trial, luteinizing hormone, follicle-stimulating hormone and testosterone were determined radioimmunologically in serum samples of 232 patients with germ cell tumors after unilateral orchiectomy (patient age 18 to 64 years) up to 153 months after chemotherapy. Additionally, 51 of these patients were investigated in a longitudinal trial before and up to 5 years after chemotherapy. All patients received a t least 2 courses of different cisplatin based chemotherapy regimens: cisplatidvinblastinehleomycin, cisplatidvinblastinel bleomycidifosfamide, cisplatidetoposidehleomycin, cisplatidvinblastine/bleomycidifosfamidel etoposide. Additionally, 11 patients with germ cell tumors (age 22 to 38 years, stage I ) were investigated within the first year after orchiectomy and retroperitoneal lymphadenectomy but without chemotherapy. Results: In the transversal trial, 24 of 73 patients investigated during the first year after chemotherapy showed elevated luteinizing hormone concentrations, 5 had subnormal serum testosterone and 65 had elevated serum follicle-stimulating hormone, reflecting spermatogenesis deficits. In 28 patients studied longer than 8 years after chemotherapy (median followup 8.5 years, range 8.0 to 12.61, luteinizing hormone was elevated in 1 patient, follicle-stimulating hormone was increased in 18 and testosterone was subnormal in 1. Patients without chemotherapy treatment showed gonadotropin and testosterone within normal range and 3 patients had elevated serum follicle-stimulating hormone. In the longitudinal study, mean serum luteinizing hormone plus or minus standard deviation (3.45 ? 0.05 IUA.), follicle-stimulating hormone (7.79 2 0.13 IUA.) and testosterone (18.6 ? 0.17 nmo1.A.) were within the normal range before chemotherapy; serum follicle-stimulating hormone increased after chemotherapy and 60 months after treatment, and follicle-stimulating hormone was still significantly elevated (16.9 ? 0.71 IUA., 19 cases, p <0.001). Mean luteinizing hormone and testosterone levels were within the normal range, but 60 months after therapy the testosterone-to-luteinizinghormone ratio was still lower than before treatment ( p <0.05). Conclusions: In patients with germ cell tumors, a compensated insufficiency of the function of the Leydig cells was still observed up to 60 months after chemotherapy. Of these patients 68% showed elevated follicle-stimulating hormone levels, which reflected a functional insufficiency of the Sertoli cells with impaired spermatogenesis. This study shows that impairment of germinative functions is more severe and protracted than the impairment of the endocrine functions. K E Y WORDS:testis, drug therapy, germ cells, neoplasms
Germ cell tumors are the most frequent malignancies in men between 20 and 40 years o1d.I.' These neoplasms are classified as seminoma, nonseminoma, for example, teratoma, embryonal carcinoma, choriocarcinoma, yolk sac tumors and combined tumors." Treatment of germ cell tumors consists of orchiectomy and retroperitoneal lymphadenectomy, but in patients with disseminated disease, only cisplatin based chemotherapy is able to achieve a high cure rate.4-6 To date, the cure rate in these malignancies has increased to 90%. 1 Unfortunately, cytotoxic chemotherapy evokes acute and chronic side effects.7-") Chronic effects of cytotoxic treatment on gonadal functions require further investigation, because patients with germ cell tumors are of reproductive age Accepted for publication January 10, 1997
and some of them will be infertile after curative treatment. Leydig cells are more resistant to chemotherapy compared with germ cells, but endocrine functions (steroidogenesis, for example) are also affected in a dose-dependent manner." To our knowledge, there are at present few conclusive data about t h e duration of testicular impairment after chemotherapy or the gonadotoxic effects of different chemotherapy regimens for germ cell tumors. Therefore, the aims of the present investigation were to assess by hormonal evaluations the frequency and duration of gonadal impairment in patients after cisplatin based chemotherapy for germ cell tumors and the possibility of recovery as well as the gonadotoxic effects of different chemotherapy regimens. Luteinizing hormone and testosterone levels were measured to determine the function of the Leydig cells.Serum follicle-stimulating
844
a45
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
hormone was used a s an indirect biological indicator of Sertali cell functions. Elevated pretreatment and posttreatment follicle-stimulating hormone concentrations reflected a functional insufficiency of the Sertoli cells with an impaired spermatogenetic process.12-15 A significant inverse correlation has been demonstrated between the number of sperm and follicle-stimulating hormone concentrations.16 Additionally, the testosterone-to-luteinizing hormone ratio, which is considered to be a more sensitive marker of the function of the Leydig cells than testosterone or luteinizing hormone alone, was used to register the functions of the Leydig cells.17 MATERIAL A N D METHODS
Study design. In a transversal trial, serum concentrations of luteinizing hormone, follicle-stimulating hormone and testosterone were determined radioimmunologically in 232 patients with germ cell tumors up to 153 months after completion of cisplatin based chemotherapy. In this cross-sectional trial, hormone determinations were carried out once in each patient during posttreatment followup. At the time of the hormone evaluations disease was in complete remission, and alpha-fetoprotein and p-human chorionic gonadotropin were normal. Comparative hormone studies were performed in 73 of these patients within the first year after chemotherapy and in 28 with an interval of more than 8 years after cytotoxic treatment. No significant difference in the mean number of chemotherapy courses was observed in these 2 subsets of patients with germ cell tumors (3.41 5 0.2 versus 3.68 5 0.2 courses). Additionally, 51 of these patients were investigated in a longitudinal trial after unilateral orchiectomy before the onset of chemotherapy and during long-term followup 1, 3, 6, 12, 24, 36, 48 and 60 months after completion of chemotherapy. Eleven patients with germ cell tumors without cytotoxic treatment were studied within the first year after unilateral orchiectomy and retroperitoneal lymphadenectomy. A total of 105 healthy male volunteer, medical students and physicians of the University of Bonn (age 33.9 2 1.19 years, range 18 to 631, served a s controls. The study was approved by the ethics committee of the University of Bonn. Transversal trial. The study included 232 patients treated for germ cell tumors of the testis between 1983 and 1995 at the University of Bonn. Of these patients 213 (91.8%)underwent bilateral (138, 64.8%)or unilateral (75, 34.2%)retroperitoneal lymphadenectomy. Histology and the pathological stage of the tumors classified according to Mostofi et a13 are shown in table 1. The patients were treated with 2 to 12 cycles of different cisplatin based chemotherapy regimens: cisplatidvinblastinehleomycin (mean plus or minus standard error 3.13 5 0.06 courses, 70 cases, 30.2%),cisplatid
TABLE1. TheraDv related characteristics
of
histolom and pathological stage according to Mostofi et al' i n 232 germ cell tumors
Cisplati.inb,astine/
No. CisplatinNinblastinei Bleornycidfosfamide
Bleomycin (ri,I
(%)
No.
Histology: Seminoma Teratoma Embryonal Ca Chorioca Teratoca Chorioca and other germ cell tumors Seminoma and other germ cell tumors Nonclassifiable tumors Pathological stage: I IIA IIB IIC 111
vinblastinehleomycinhfosfamide (3.9 2 0.05, 62 cases, 26.74 ), cisplatidetoposidehleomycin (3.03 5 0.08, 73 cases, 3 1.4%~) and cisplatidvinblastinehleomycidifosfamide/etoposide (5.96 t 0.14, 27 cases, 11.64). Dose schedule was as follows: days 1 to 5, 20 mg./m.z cisplatin, days 1 and 2, 6.0 mgJm.2 vinblastine, days 1 to 5, 100 mg.lm.2 etoposide, days 2, 9 and 16, 30 IU bleomycin and days 1 to 5, 1.5 gm./m.2 ifosfamide daily. A total of 22 patients was treated initially according to the cisplatidvinblastinehleomycidifosfamide regimen followed by cisplatidetoposidehleomycin after vinblastine induced side effects (for example, aplasia). Another subset of 5 patients initially received cisplatidvinblastine/ bleomycidifosfamide followed by cisplatidetoposidehleomycin after a relapse of the disease which led t o cisplatid vinblastinehleomycin/ifosfamide/etoposide.Additionally, 11 patients with germ cell tumors (pathological stage I) were investigated within the first year (3.1 +- 1.16 months, range 1 to 12) after unilateral orchiectomy and retroperitoneal lymphadenectomy and were undergoing a followup protocol. Histology of the tumors, which were classified according to Mostofi e t al," were teratocarcinoma (6), embryonal carcinoma ( l ) ,choriocarcinoma (2), choriocarcinoma and other germ cell tumors (1) and nonclassifiable tumor (1). Longitudinal trial. This subset included 51 patients (age 30.1 2 0.1 year, range 18 to 64), of whom 47 had undergone unilateral retroperitoneal lymphadenectomy. The histology of the tumors in these patients was teratoma (2, 3.9%),teratocarcinomas (20,39.2%'), embryonal carcinoma (14,27.5%), choriocarcinoma combined with other germ cell tumors (1, 2.0%),seminoma (8, 15.7%),seminoma in combination with other germ cell tumors (4, 7.8%)and nonclassifiable tumors (2, 3.9%).Pathological stages of the tumors in the patients were stage I(10,19.6%),stage IIA (21,41.2%),stage IIB (12, 23.5%),stage IIC (4, 7.89) and stage I11 (4, 7.8%).The patients investigated in the longitudinal study were treated with cisplatidvinblastinehleomycin (3.25 ? 0.24 courses, range 2 to 4, 4, 7.8%),cisplatidvinblastinehleomycinhfosfamide (3.63 2 0.09, 2 to 4, 8, 15.7%),cisplatin/etoposide/ bleomycin (2.87 t 0.04, 1 to 6, 29, 56.9%)or cisplatidvinblastinehleomycidifosfamide/etoposide (6.0 5 0.25, 2 to 11, 9, 17.6%). Endocrine studies. Testicular function studies were performed during routine followup. Two blood samples were obtained in the early morning a t intervals of 20 minutes for biochemical studies, tumor marker and hormone determinations. The serum was separated, pooled and stored a t -2OC until assay. Serum concentrations of luteinizing hormone and follicle-stimulating hormone were determined in dupli-
No.
Cisplatin/Etoposide/ Bleomycin (~
No.
CisplatinA'inblastine/ Bleomycidfosfamide/ Etoposide I r/, 1
Total No 1% I
9133.31 3tll.11 3111.11
17 (27.41 5 18.11
2 (2.71 6 (8.21 22 130.21 111.41 25 134.31 1 11.4)
6 122.2, 1 (3.71
23 I991 28 t 12 11 48(2071 1 I041 63 I27 21 14 I601
9 112.91
9 114.51
8 110.9)
3 111.11
29 (12 51
4 15.71
12 (19.31
8 t10.91
2 17.51
26 Ill 21
7 121.41 10 (18.61 23 (32.9) 15 (21.41 15 (21.4)
3 (4.81 10 (16.1I 12 (19.41 13 121.0) 24 (38.71
10 113.7) 19 126.0) 27 (37.01 11 115.1) 6 18.2)
5118.51 3 (11.1) 4 114.8) 6 (22.2) 9 (33.3)
25(1081 4 2 ( 1 8 1) 66(284) 45!194) 54 (23 3)
7 11O.Ol 15 121.41 3 118.61
5 18.1) 4 16.51 10 (16.11
15 (21.41 7 l10.01
846
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
months after orchiectomy and retroperitoneal l p p h a d e n e c bmy. After chemotherapy, mean luteinizing hormone concentration in all patients was within the normal range, but it was significantly higher in patients than i n healthy controls (5.78 t 0.28 versus 2.77 i- 0.11 I U k , p <0.001) as well a s in patients without cytotoxic treatment within the first year after surgery (3.4 2 0.48 IUfl., p <0.01). Serum luteinizing hormone levels were elevated in 31 of t h e 232 patients (13.4%).Of 73 patients studied within t h e first year after chemotherapy, serum luteinizing hormone levels were elevated in 24 patients (32.9%,fig. 1, A ) . Of 28 patients studied more than 8 years after chemotherapy (median 8.5 years, range 8.0 to 12.6),only 1 (3.6%) had elevated serum luteinizing hormone levels, which reflects a significant decrease in the percentage of patients with a n impairment of the Leydig cell function ( p <0.01), depending on the interval since treatment. Mean luteinizing hormone concentrations did not differ significantly in patients treated with cisplatidvinblastinehleomycin (5.37 i- 0.29 IUA.) or cisplatidvinblastine/ bleomycidifosfamide (5.73 2 0.27 IUA.) and patients treated with cisplatidetoposidehleomycin had serum luteinizing hormone levels of 5.81 -t 0.22 IUA. Germ cell tumor patients treated with cisplatin/vinblastinehleomycin/ifosfamide/ etoposide had higher serum luteinizing hormone concentrations (6.87 -t 0.38 IUA.) than those treated with the other regimens. Mean testosterone concentration in all patients was within RESULTS normal range and did not differ significantly from healthy Transi)ersaltrial. The mean age of the 232 patients treated controls or surveillance patients after surgical treatment for germ cell tumors was 31.5 t 0.51 years (range 18 to 64). (18.1 0.43 versus 18.5 -t 0.51 and 19.0 i- 1.15 nmo1.A.). Of No significant differences were observed i n the mean age these 232 patients 12 (5.2%),including 5 of 73 patients studof patients treated with cisplatidvinblastinehleomycin ied within the first year after chemotherapy (6.8%), had (33.6 2 0.57 years), cisplatin/vinblastinehleomycidifos- subnormal testosterone concentrations (range 2.15 to 9.2 famide ( 3 2 . 9 i- 0.46 y e a r s ) a n d cisplatin/vinblastine/ nmolA., fig. 1, B ) . Of 28 patients studied longer than 8 years bleomycidifosfamide/etoposide (32.8 -t 0.47 years), respec- after chemotherapy (median 8.5 years, range 8.0 to 12.6), 2 tively, b u t p a t i e n t s t r e a t e d with cisplatin/etoposide/ (7.1%) had subnormal serum testosterone (5.34 and 7.95 bleomycin were significantly younger (27.9 -t 0.42 years, p nmo1.A. j. A recovery of endocrine functions, reflected by the <0.01) because of t h e later introduction of this chemotherapy percentage of patients with subnormal testosterone, could regimen. The interval between the completion of chemother- not be achieved. No significant differences in testosterone apy and hormone determinations did not differ significantly concentrations could be observed between patients treated between patients treated with cisplatidvinblastinehleomy- with cisplatidvinblastinehleomycin(17.5 2 0.43 nmo1.A.) or cin (45.2 t 2.28 months, range 1 to 134) and those treated cisplatin/vinblastine/bleomycidifosfamide( 18.1 t- 0.4 nmo1.i with cisplatidvinblastinehleomycidifosfamide(50.2 i- 2.39 1.). Patients treated with cisplatin/etoposidehleomycin months, 1 to 153),but this interval was significantly shorter (18.8 t 0.49 nmo1.fl.j and those treated with cisplatinl for patients treated with cisplatidetoposidehleomycin(18.8 2 vinblastinehleomycin/ifosfamide/etoposide h a d normal 1.1 months, 1 to 72, p <0.001) and cisplatidvinblastine/ mean testosterone (17.5 2 0.35 nmo1.A.). Mean folliclebleornycinhfosfamidetoposide (41.2 i- 2.24 months, 1 to 123, p stimulating hormone concentration in all patients was sig(0.05). Additionally, 11 patients with pathological stage I germ nificantly increased compared with healthy controls (18.8 -t cell tumors (age 30.2 -t 1.48 years, range 22 to 38) undergoing 0.93 versus 4.02 t 0.17 IUA., p <0.001) and to the 11 germ followup without chemotherapy were investigated 3.1 t 1.16 cell tumor patients without chemotherapy within the first year after orchiectomy and lymphadenectomy (8.5 2 1.15 * MAIAclone, Serono, Freiburg, Germany. IUA., p <0.01).Of these 11patients under surveillance 3 had t RSL, ICN Biomedicals, Inc., Costa Mesa, California
cate by immunoradiometric assay* and concentrations of testosterone by radioimmunoassay.: Normal ranges in men are 1.8 to 9.2 IUA. for luteinizing hormone, 1.6 to 9.0 IUA. for follicle-stimulating hormone, and 10.4 to 34.7 nmo1.A. for testosterone. The within-assay variation for luteinizing hormone (concentration 2.1 IUA. ), follicle-stimulating hormone (5.5 IUA.) and testosterone (12.5 nmo1.A.) was 3.5, 3.9 and 9.5%(20),respectively and the between-assay variation was 9.4, 5.3 and 9.8% (18). respectively. The testosterone-toluteinizing hormone ratio was used to describe the endocrine function of the Leydig cells more profoundly than luteinizing hormone or testosterone alone.” The normal range for the testosterone-to-luteinizing hormone ratio is 3.3 to 17.9 nmol./IU (105 men). Semen analyses. Semen analyses were performed simultaneously with serum follicle-stimulating hormone analyses in 46 patients with germ cell tumors using conventional methods to correlate sperm counts with serum follicle-stimulating hormone concentrations. I n each case, the abstinence interval was 4 days. Statistical analyses. Results were expressed as mean plus or minus standard error, as ranges and as the 5th and 95th percentile to establish reference values for the testosterone/ luteinizing hormone quotient. Statistical analyses were performed using the Mann-Whitney U test, the Wilcoxon test and a Spearman’s rank correlation (r). A p value <0.05 was considered significant.
A lo]..
T i m a(ur clumothuapy Imonth.)
Time after chamothenpy [monlhal
Time aner chemotherapy [months1
FIG. 1. Individual serum concentrations of luteinizing hormone (A), testosterone ( B , and follicle-stimulating hormone ( C Iin 232 patients with germ cell tumors u to 153 months after completion of chemotherapy. Lower and/or upper normal range is marked h y dottcd lines. L H . luteinizing hormone. F&f , follicle-stimulating hormone.
847
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
elevated follicle-stimulating hormone concentrations (27.3%, 11.0 to 16.2 IUI1.J. Serum follicle-stimulating hormone showed highest concentrations during the first year after therapy (fig. 1, C,, when 65 of 73 patients (89.0%)had elevated follicle-stimulating hormone (range 9.1 to 88.9 IUA.), then follicle-stimulating hormone gradually decreased. More than 8 years after chemotherapy (median 8.5 years, range 8.0 to 12.6),mean follicle-stimulating hormone was still elevated in 28 patients (14.3 2 2.48 IUA., median 12.5, range 3.3 to 41.5). A significantly lower percentage of patients with elevated serum follicle-stimulating hormone was evident ( p ~ 0 . 0 ),5 reflected by elevated follicle-stimulating hormone concentrations (range 10.2 to 41.5 IUA.) in 18 of these 28 patients. Still, a high percentage of patients showed elevated serum follicle-stimulating hormone as an indirect marker of Sertoli cell insufficiency (64.3%).An inverse correlation was observed between sperm count and follicle-stimulating hormone concentrations in 46 patients treated for germ cell tumors ( r = -0.77, p <0.001, fig. 2). Mean folliclestimulating hormone concentrations did not show significant differences in patients treated with cisplatin/vinblastine/ bleomycin ( 17.4 t 0.99 IU/l.) or cisplatin/vinblastine/ bleomycidifosfamide ( 18.7 ? 0.78 IUA. ). Cisplatidetoposidel bleomycin (18.6 -+ 0.86 IUA.) a s well as those treated with cisplatin/vinblastinemleomycin/ifosfamide/etoposide also showed elevated serum follicle-stimulating hormone concentrations (23.6 2 1.22 IUA.). Within the first year after completion of chemotherapy, the mean testosterone-to-luteinizing hormone ratio was within normal range but significantly lower in the patients (3.41 2 0.44 nmol./IU, p <0.001, 73) compared with the healthy controls (7.71 f 0.37 nmol./IU.) as well as the patients undergoing a followup protocol without chemotherapy (6.46 2 0.8 nmo1.A.). Of 73 patients 44 (60.3%)showed a subnormal testosterone-to-luteinizing hormone ratio (range 0.33 to 3.11 nmol./IU), whereas 1 patient in the surveillance group (9.1%) showed a subnormal testosterone-to-luteinizing hormone ratio (2.66 nmol./IU). More than 8 years after chemotherapy (median 8.5 years, range 8.0 to 12.6), the mean testosteroneto-luteinizing hormone ratio was within normal range but still significantly lower than in controls (5.01 2 0.56 nmol./ IU, p <0.01, 28). A significantly lower percentage of patients with a subnormal testosterone-to-luteinizinghormone ratio was observed (32.1%, 9 of 28 patients, range 0.67 to 2.74 nmol./IU, p <0.05) compared with patients within the first year after chemotherapy.
60
-
50
-0
-G
40
-
I
30-1.
2
2
To elucidate the influence of different chemotherapy regimens (cisplatidvinblastinehleomycin,cisplatin/vinblastine/ bleomycidifosfamide, cisplatidetoposidehleomycin, cisplatidvinblastinehleomycin/ifosfamide/etoposide)on germinative functions, we analyzed the percentage of patients with elevated follicle-stimulating hormone concentrations a t various times and intervals after chemotherapy. Within the first year a h r chemotherapy, a high percentage of patients with elevated serum follicle-stimulating hormone could be observed in all different chemotherapy regimens (cisplatidvinblastinehleomycin, 84.2% of 19 patients; cisplatidvinblastinehleomycidifosfamide, 90% of 20; cisplatidetoposidehleomycin,81.2% of 38; cisplatdvinblastinehleomycin/ifosfamide/etoposide,100% of 12). At 37 to 48 months after treatment the rate of elevated follicle-stimulating hormones was 60% in 6 of 10 patients treated with cisplatidetoposidehleomycin, 66.7% in 8 of 12 treated with cisplatidvinblastinehleomycin, 63.6% in 7 of 11 treated with cisplatidvinblastine/bleomycin/ifosfamide and 80% in 4 of 5 treated with cisplatidvinblastinehleomycidifosfamiddetoposide. At 48 to 60 months after treatment, 63.6% of the patients treated with cisplatidvinblastinehleomycin(7 of 11 patients) and 55.6% of the patients treated with cisplatid vinblastinehleomycidifosfamide (5 of 9) still had elevated follicle-stimulating hormone concentrations. Up to 72 months aRer chemotherapy, 504 of patients (5 of 10) treated with cisplatidvinblastinehleomycin still showed elevated serum follicle-stimulating hormone (fig. 3). Longitudinal trial. After orchiectomy and before chemotherapy, mean Iuteinizing hormone and testosterone concentrations were within the normal range, but mean luteinizing hormone was significantly higher than in healthy controls (3.51 ? 0.37 versus 2.77 -+ 0.11 IUA., p <0.001). One patient had supranormal serum luteinizing hormone ( 16.6 IUA.) and 4 patients had subnormal testosterone concentrations (4.4 to 7.9 nmo1.A. ). Mean serum follicle-stimulating hormone levels were significantly higher than in controls (7.79 -+ 0.13 versus 4.02 2 0.17 IUA., p <0.001), but mean follicle-stimulating hormone was still within the normal range. Before chemotherapy, 34 of these 51 patients (66.7%:) had folliclestimulating hormone concentrations within normal range, whereas 17 patients (33.3%) had elevated serum folliclestimulating hormone (range 9.4 to 26.6 IUA. ). After chemotherapy, mean luteinizing hormone and testosterone were within normal range (fig. 4, A and B However,
i
20
-
.0
1-12
OJ,
0
1
I
I
I
1
20
40
60
80
100
Sperm count [millJmL] FIG. 2. Inverse correlation of serum follicle-stimulatinghormone ( F S H )and sperm count in 46 patients treated for germ cell tumors.
13-24
25-36
37-48
49-60
61-72
Time after chemotherapy [months]
FIG. 3. Percentage of patients with elevated serum folliclestimulating hormone ( F S H ) concentrations after treatment with
different chemotherapy schedules for germ cell tumors. Small numbers on bars indicate number of patients with elevated folliclestimulating hormone concentrations compared with total number of patients.
848
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
... ...p 47
...
...
2'
12
6
0
24
36
0
60
48
12
6
24
36
48
60
Time aRer chemotherapy [months]
Time after chemotherapy [months] 301
*l
0
6
12
24
36
48
0
60
6
12
24
36
48
60
Time after chemotherapy [months]
Time after chemotherapy [months]
FIG.4. Serum concentrations of luteinizing hormone ( L H , A), testosterone ( B ) ,follicle-stimulating hormone (FSH, C) and testosteroneto-luteinizing hormone ratio tD) expressed as mean plus or minus standard error before and after chemotherapy for germ cell tumors performed in longitudinal trial. Numbers of patients are given on top of bars. *** p <0.001.** p 10.01.* p c0.05.
elevated serum luteinizing hormone was observed after 1 month of chemotherapy in 9 of 47 patients (19.1%),after 3 months in 12 of 38 (31.64), after 6 months in 5 of 39 (12.8%), after 12 months in 6 of 41 (14.6%1, after 24 months in 2 of 36 (5.55%)and after 36 months in 2 of 25 patients (7.69%).At 48
and 60 months after chemotherapy, supranormal serum luteinizing hormone was observed in 1 of 23 (4.3%)and in 2 of 19 patients (5.3%),respectively. Highest mean luteinizing hormone concentrations were observed 3 months after chemotherapy (7.17 5 0.1 I U k , 38), but up to 60 months after
TABLE2 Treatment interoal (months) and hormone concentratrons (expressed as mean plus or minus standard error, range and median) i n patients with germ cell tumors followed in a longitudinal study before and after chemotherapy ~~
Time After Chemotherapy Irnos.1 Before Range Median
~
~~
No.
47 38
38
41
24 Range Median
37
36 Range Median
27
48 Range Median 60 Rnnge Median -_ -
~~
FollicleStimulating Hormone tIU/l.,
7.792 0.13 1.0-26.6 7.8 20.9 z 0.25 1.7-59.7 19.4 23.9 2 0.3 1.O-54.1 20.8 24.4t 0.34 1.3-66.8 22.5 20.55 0.34 1.540.5 16.8 17.0 i 0.34 3.641.0 14.0 16.5t 0.42 3.8-51.7 13.9 15.2:0.41 3.S38.7 13.3 16.9I 0 71 2.1-579 14.9
6.0
12 Range Median
~~
3.45t 0.05 1.0-16.6 2.9 6.77 z 0.1 1.0-22.8
51
3 Range Median 6 Range Median
~~
(1UA.i
1 Range Median
~
Luteinizing Hormone
7.17f 0.11 1.2-17.9 5.35 5.94I0.1 1.4-21.6 4.7 5.68 f 0.09 i.sia.5 4.5 4.67t 0.49 1..5-17.1 4.1 5.29 r 0.1 1.0-13.3 4.9 4.83 0.08 2.8-9.2 4.2 5.26* 0.16 2.1-14.2 4.1 f
23 19 ~.
~
~~
~~~
Testosterone
(nmol./l.)
Testosterone-to-Luteinizing Hormone Ratio (nmol./lU~
18.61 0.17 4.44-53.9 17.8 21.5z 0.24 0.56-69.4 20.9 19.6t 0.19 10.7-52.4 18.7 18.5 t 0.13 9.9-30.3 17.8 18.3z 0.17 8.043.1 16.8 17.8t 0.14 9.96-32.8 16.9 16.2 i 0.2 7.9-29.5 15.7 17.6I0.22 1o.R-32.9 17.3 17.5z 0.31 10.1-29.3 16.4
6.38 I0.72 1.12-16.1
~~~
3.972 0.4 0.23-12.7 3.59 4.1 t 0.73 0.73-10.1 3.3 4.12 I0.52 0.85-7.1 3.67 4.632 0.59 0.92-13.5 3.67 4.76f 0.43 0.58-12.5 4.12 3.27? 0.27 0.89-5.45 3.43 4.33I 0.52 1.76-11.9 3.59 4.73I0.59 1.91-8.17 4.84 .---
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
treatment, luteinizing hormone was still significantly higher than before chemotherapy (5.26 -+ 0.16 versus 3.45 -+ 0.05 IUA., p <0.001). Mean serum testosterone was almost within normal range and vaned between 16.5 and 21.5 nmo1.A. (table 2). A subnormal serum testosterone was observed in 2.5 to 4.2%of the patients under observation. Serum follicle-stimulating hormone increased with a significant 6-month peak (24.4 5 0.34 IUA., p <0.001, 38, fig. 4, C). Of 38 patients 37 (97.4%)had elevated folliclestimulating hormone concentrations (range 10.7 to 66.8 IU/ 1. ). Then follicle-stimulating hormone gradually decreased, but 60 months after chemotherapy, mean follicle-stimulating hormone was still significantly elevated (16.9 ? 0.71 IUA., p <0.01, 19) compared with the pretreatment values. Of 19 patients 13 (68.4%) still showed elevated serum folliclestimulating hormone concentrations (range 9.6 to 43.4 IUA.). Before chemotherapy, no significant difference could be observed in the mean testosterone-to-luteinizing hormone ratio of germ cell tumor patients and healthy controls (6.38 ? 0.72 versus 7.71 t 0.37 nmol./IU) and 12 patients (23.1%) showed a subnormal testosterone-to-luteinizing hormone ratio. One month after chemotherapy, the testosterone-toluteinizing hormone ratio significantly declined (3.97 ? 0.40 nmol./IU, p <0.05, fig. 4, D),and 3 months after treatment, a maximum of 52.6%of patients (20 of 38 patients) showed a subnormal ratio. Sixty months after therapy, this ratio still was significantly lower (4.73 -+ 0.59 nmol./IU, p <0.05) than the pretreatment ratio and the testosterone-to-luteinizing hormone ratio in healthy controls, but within the lower normal range. Of 19 patients 6 (31.6%)investigated 60 months after chemotherapy still showed a subnormal testosteroneto-luteinizing hormone ratio (1.1 to 2.6 nmol./IU).
849
Zated. Our data are supported by Stephenson et al,Z0 who -eported on oligospermia or azoospermia in 19 of 30 patients 63.370) treated with 2 t o 4 courses of chemotherapy according to the cisplatidetoposidehleomycin regimen and by Sietema et al,” who reported on a protracted luteinizing hormone and follicle-stimulating hormone elevation more than 5 years after 4 cycles of chemotherapy. In a small collective of 25 patients, after cisplatidvinblastinehleomycin treatment for metastatic testicular cancer, Hansen et a1 demonstrated that azoospermia or oligospermia was associated with elevated serum follicle-stimulating hormone within the first 2 years after treatment, followed by a gradual recovery in the further followup.18All patients showed at least partial regeneration of spermatogenesis, but 5 years after therapy, sperm density had returned to pretreatment sperm counts in only 46%. Hansen et a1 reported on higher serum folliclestimulating hormone and lower sperm density in patients treated with cisplatidvinblastinehleomycin compared with patients treated with orchiectomy alone, but this “control collective” was older than the chemotherapy-treated patients.ls To our knowledge, no conclusive data were available concerning the comparison of gonadotoxic effects- especially the influence on spermatogenesis reflected by elevated folliclestimulating hormone concentrations-of different cisplatin based chemotherapy regimens for germ cell tumors. In the first 12 months after chemotherapy, there was no difference in the percentage of patients with elevated serum folliclestimulating hormone levels in these 4 therapy groups. During the followup, the percentage of patients with elevated follicle-stimulating hormone levels decreased in the cisplatid vinblastinehleomycin, cisplatidvinblastinehleomycidifosfamide and cisplatidetoposidehleomycin groups, which reflects a gradual recovery of spermatogenesis in some of the DISCUSSION patients. Patients treated with cisplatidvinblastinehleomyIn our transversal study, we had the opportunity to screen cidifosfamide/etoposide had elevated follicle-stimulating the gonadal function of 232 patients with germ cell tumors up hormone concentrations without a clear decrease most freto 12.6 years after completion of cisplatin based chemother- quently because of the higher dose of cisplatin in this subapy. We observed a n elevation of serum luteinizing hormone group. These data are in agreement with those of Hansen et in 24 and subnormal testosterone in 5 of 73 patients studied a1.18 who found dose-dependent gonadal damage from use of within the first year after chemotherapy, which reflects a n cytostatic drugs. Now, 48 months after completion of chemoimpairment of the function of the Leydig cells. The majority therapy, no conclusion can yet be drawn as to which regimen of patients had luteinizing hormone and testosterone concen- has the least gonadotoxic side effects. The data of the cross-sectional trial were confirmed by our trations within normal range. Longer than 8 years after chemotherapy, 1 of 28 patients still had elevated luteinizing longitudinal study. After orchiectomy and before chemotherhormone and 2 patients had subnormal testosterone concen- apy, 1 of 51 patients showed elevated luteinizing hormone trations. Our results agree with those of Hansenls and and 2 patients showed subnormal testosterone concentraFossa19 et a1 who reported on luteinizing hormone elevation tions, which reflects a n impaired function of the Leydig cells. and subnormal testosterone concentrations as signs of dis- In accordance with several investigators, mean luteinizing turbed Leydig cell function in some patients treated for germ hormone and testosterone concentrations were within norcell tumors with cisplatin based chemotherapy. Levels of mal range after chemotherapy.18.19 However, mean luteinizSerum follicle-stimulating hormone-a sensitive parameter ing hormone was still significantly higher years after comof germinative function12-1s (fig. 2 t w e r e significantly ele- pletion of chemotherapy compared with pretreatment values, vated after chemotherapy. In the first year after treatment, which reflects a compensated insufficiency of the Leydig 65 of 73 patients (89%)showed elevated follicle-stimulating cells. In addition, we used the testosterone-to-luteinizing horhormone concentrations that reflected spermatogenesis impairment and follicle-stimulating hormone then gradually mone ratio, which is considered to be a more sensitive marker decreased. More than 8 years after chemotherapy, however, of the Leydig cell function than testosterone or luteinizing 18 of 28 patients (64.3%)still had elevated serum follicle- hormone alone.17 Before chemotherapy and after orchiecstimulating hormone. Thus, a significant disturbance of sper- tomy, the mean testosterone-to-luteinizinghormone ratio did matogenesis reflected by elevated follicle-stimulating hor- not differ significantly among patients, healthy controls and mone concentrations was still evident in the majority of the germ cell tumor patients without chemotherapy, excluding patients investigated. Our data contrast with the observa- an impairment of the function of the Leydig cells caused tions of Fossa et a1,19 who reported on median follicle- by orchiectomy. Twelve patients (23.1%) had subnormal stimulating hormone levels within the upper normal range in testosterone-to-luteinizing hormone ratios and 3 months af47 germ cell tumor patients with a median followup of 9 ter treatment, a maximum of 52.6%of the patients showed a Years after chemotherapy generally treated according t o the subnormal ratio. For up to 60 months in the postclsphtidvinblastinehleomycin schedule, which may imply chemotherapy period, the testosterone-to-luteinizing horthat many of these patients had supranormal follicle- mone ratio declined and was significantly lower in patients stimulating hormone levels. Unfortunately, the authors didI than in controls. However, 60 months after completion of not show individual hormone data of the patients investi- chemotherapy 31.6% still had a subnormal testosterone-to-
850
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
luteinizing hormone ratio, which reflects a g r a d u a l recovery of Leydig cell function. This finding suggests that y e a r s after chemotherapy, higher endogenous luteinizing hormone secretion i s still necessary t o s t i m u l a t e testosterone synthesis and/or secretion t o t h e s a m e degree as in healthy subjects. These findings confirm that a f t e r chemotherapy, a n impairm e n t of testosterone synthesis/secretion-although generally still p r e s e n t in a compensated form-is based on a primary testicular defect in p a t i e n t s treated for g e r m cell tumors. This i m p a i r m e n t c a n be caused directly by chemotherapy a n d by reduced testicular blood flow a f t e r treatment w i t h cytotoxic agents.22 The clinical relevance of compensated Leydig cell disturbance in p a t i e n t s treated for g e r m cell tumors m u s t be illuminated by sexual a n a m n e s i s that details reduction of libido a n d erectile dysfunction. Before chemotherapy, 17 of 51 p a t i e n t s with germ cell t u m o r s (33.3%) had elevated follicle-stimulating hormone concentrations. which reflects a n insufficiency of Sertoli cell function. After completion of chemotherapy, s e r u m folliclestimulating hormone levels increased, with a significant peak a t 6 months, t h e n gradually decreased. Even 60 m o n t h s after chemotherapy, follicle-stimulating hormone w a s significantly higher t h a n before t r e a t m e n t and an elevation of s e r u m follicle-stimulating hormone was still obvious in m o r e t h a n 68%of the patients under observation. CONCLUSIONS
A compensated insufficiency of Leydig cell function reflected b y p e a k levels of s e r u m luteinizing hormone and a low testosterone-to-luteinizinghormone ratio occurs i n u p to 528 of patients with g e r m cell t u m o r s within the first m o n t h s after chemotherapy, but t h i s generally compensated impairment of the Leydig cells w a s still evident i n a b o u t 31%of the patients 60 m o n t h s a f t e r completion of treatment. The clinical relevance of t h i s compensated endocrine disturbance still needs to be investigated. The elevation of s e r u m folliclestimulating hormone, which reflects an i m p a i r m e n t of germinative functions, i s more severe a n d protracted 6 m o n t h s after chemotherapy. At 60 m o n t h s after chemotherapy, more t h a n 68% of the p a t i e n t s still showed elevated folliclestimulating hormone concentrations. Because some of these patients will be infertile after curative t r e a t m e n t , the option of s p e r m cryopreservation should be considered for patients with g e r m cell t u m o r s before chemotherapy. To date i n our study, i t could not be decided which chemotherapy regimen investigated had the l e a s t gonadotoxic side effects.
Mrs. M. Lafosse, I. Horn a n d U. Bartholomay provided technical assistance. REFERENCES
1. Fossa, S. D., Aass, S. D. and Kaalhus, 0.: Testicular cancer in young Norwegians. J . Surg. Oncol., 3 9 43, 1988. 2. Mackay, E. N. and Sellers, A. H.: A statistical review of malignant testicular tumours based on the experience of the Ontario Cancer Foundation Clinics, 1938-1961. Canad. Med. Ass. J., %k889, 1966. 3. Mostofi, F. K. and Sellers, A. H.: Histology typing of testis tumours. Geneva, Switzerland: World Health Organization, 1977. 4. Wheeler, J. E.: The history of teratomas. In: The Teratomas.
Edited by I. Damjanov, B. B. Knowles and D. Solter. Clifton, New Jersey: Humana Press, pp. 1-22. 5. Nichols, C. R. and Roth, J. B.: Management of metastatic nonseminoma, development of effective chemotherapy. In: Testicular Cancer: Investigation and Management. Edited by A. Horwich. New York: Chapman and Hall Medical, chapt. 14, pp. 185-203. 6. Einhorn, L. H. and Williams, S. D.: Chemotherapy of disseminated testicular cancer. A random prospective study. Cancer, 4 6 1339, 1980. 7. Ellis, M. and Sikora, K.: The current management of testicular cancer. Brit. J. Urol., 5 9 2, 1987. 8. Hartlapp, J. H.. Weissbach, L. and Bussar-Maatz, R.: Adjuvant chemotherapy in nonseminomatous testicular tumour stage 11. Int. J. Androl., 10: 277, 1987. 9. Klingmiiller, D., Brenneman, W., Brensing, K. A., Deitenbeck, R., Stoffel-Wagner, B., Hartlapp, J. H., Jager, N. and Mumperow, E.: Hodenfunktionsstorungen bei Patienten mit Hodentumoren. Urologe LBI, 32: 9, 1992. 10. Brenneman, W., Stoffel-Wagner, B., Bidlingmaier, F. and Klingmiiller, D.: Immunoreactive plasma inhibin levels in men after polyvalent chemotherapy of germinal cell cancer. Acta Endocr., 126 224, 1992. 11. Grigor, K. M. and Donohue, J. P.: Reproductive aspects of testicular germ cell cancer: general discussion. Eur. Urol., 23: 177, 1993. 12. Palmieri, G., Lotrecchiano, G . , Ricci, G., Spiezia, R., Lombardi, G., Bianco, A. R. and Torino, G.: Gonadal function after multimodality treatment in men with testicular germ cell cancer. Eur. J. Endocr., 134: 431, 1996. 13. Kreuser, E. D., Harsch, U., Hetzel, W. D. and Schreml, W.: Chronic gonadal toxicity in patients with testicular cancer after chemotherapy. Eur. J. Cancer Clin. Oncol., 22: 289,1986. 14. de Kretser, D. M., Burger, H. G. and Hudson, B.: The relationship between germinal cells and serum FSH levels in males with infertility. J . Clin. Endocr. Metab., 3 8 787, 1974. 15. de Kretser, D. M., Burger, H. G. and Bremner, W. J.: Control of FSH and LH secretion. Monogr. Endocr., 3 5 12, 1983. 16. Franchimot, O., Millet, D., Vendrely, E., Letawe, J., Legros, J. J. and Netter, A,: Relationship between spermatogenesis and serum gonadotropin levels in azoospermia and oligospermia. J. Clin. Endocr. Metab., 3 4 1003, 1972. 17. Giagulli, V. A. and Vermeulen, A,: Leydig cell function in infertile men with idiopathic oligospermic infertility. J. Clin. Endocr. Metab., 66:62, 1988. 18. Hansen, P. V., Trykker, H., Helkjoer, P. E. and Andersen, J.: Testicular function in patients with testicular cancer treated with orchiectomy alone or orchiectomy plus cisplatin-based chemotherapy. J. Natl. Cancer Inst., 81: 1246. 1989. 19. Fossa, S. D., Lehne, G., Heimdal, K. and Theodorsen, L.: Chemical and biochemical long-term toxicity after postoperative cisplatin-based chemotherapy in patients with low-stage testicular cancer. Oncology, 53: 300, 1995. 20. Stephenson, W. T., Poirier, S. M., Rubin, L. and Einhorn, L. H.: Evaluation of reproductive capacity in germ cell tumor patients following treatment with cisplatin, etoposide and bleomycin. J. Clin. Oncol., 13: 2278, 1995. 21. Gietema, J. A., Sleijfer, D. T., Willemse, P. H., Schraffordt Koops, H., van Ittersum, E., Verschuren, W. M., Kromhout, D., Sluiter, W. J., Mulder, N. H. and de Vries, E. G.: Long-term follow-up of cardiovascular risk factors in patients given chemotherapy for disseminated nonseminomatous testicular cancer. Ann. Intern. Med., 116 709, 1992. 22. Huane. H. F.. Poaach. - L. M.. Nathan, E. and Gielio. W.: Acute and chronic affects of cisplatinium upon testicular function in the rat. J. Androl., 11: 436, 1990. I
Y
Val. 158. 844-850. September 1997 Printed In U.S.A.
THEJOVRSAL OF UROUIGY Copyright @ 1997 by AMEKIC'AY UROLO(,ICAL L~%x'i~Tios. Isc.
GONADAL FUNCTION OF PATIENTS TREATED WITH CISPLATIN BASED CHEMOTHERAPY FOR GERM CELL CANCER W. B R E N N E M A " , B. STOFFEL-WAGNER, A. HELMERS, J. MEZGER, N. JAGER D. KLINGMULLER From the Departments of Clinical Biochemistry and Internal Medicine, Uniuersity of Bonn, and the Department Kliniken, Hildesheim, Gernianv
of
AND
Urology. Stadtrsche
ABSTRACT
Purpose: The cure rate of patients with germ cell cancer of the testis has considerably improved since the introduction of cisplatin based chemotherapy. Because these patients are in their reproductive years and because some of them will be infertile after treatment, the effects of cytotoxic treatment on gonadal function are investigated by hormonal evaluations. Materials and Methods: In a transversal trial, luteinizing hormone, follicle-stimulating hormone and testosterone were determined radioimmunologically in serum samples of 232 patients with germ cell tumors after unilateral orchiectomy (patient age 18 to 64 years) up to 153 months after chemotherapy. Additionally, 51 of these patients were investigated in a longitudinal trial before and up to 5 years after chemotherapy. All patients received a t least 2 courses of different cisplatin based chemotherapy regimens: cisplatidvinblastinehleomycin, cisplatidvinblastinel bleomycidifosfamide, cisplatidetoposidehleomycin, cisplatidvinblastine/bleomycidifosfamidel etoposide. Additionally, 11 patients with germ cell tumors (age 22 to 38 years, stage I ) were investigated within the first year after orchiectomy and retroperitoneal lymphadenectomy but without chemotherapy. Results: In the transversal trial, 24 of 73 patients investigated during the first year after chemotherapy showed elevated luteinizing hormone concentrations, 5 had subnormal serum testosterone and 65 had elevated serum follicle-stimulating hormone, reflecting spermatogenesis deficits. In 28 patients studied longer than 8 years after chemotherapy (median followup 8.5 years, range 8.0 to 12.61, luteinizing hormone was elevated in 1 patient, follicle-stimulating hormone was increased in 18 and testosterone was subnormal in 1. Patients without chemotherapy treatment showed gonadotropin and testosterone within normal range and 3 patients had elevated serum follicle-stimulating hormone. In the longitudinal study, mean serum luteinizing hormone plus or minus standard deviation (3.45 ? 0.05 IUA.), follicle-stimulating hormone (7.79 2 0.13 IUA.) and testosterone (18.6 ? 0.17 nmo1.A.) were within the normal range before chemotherapy; serum follicle-stimulating hormone increased after chemotherapy and 60 months after treatment, and follicle-stimulating hormone was still significantly elevated (16.9 ? 0.71 IUA., 19 cases, p <0.001). Mean luteinizing hormone and testosterone levels were within the normal range, but 60 months after therapy the testosterone-to-luteinizinghormone ratio was still lower than before treatment ( p <0.05). Conclusions: In patients with germ cell tumors, a compensated insufficiency of the function of the Leydig cells was still observed up to 60 months after chemotherapy. Of these patients 68% showed elevated follicle-stimulating hormone levels, which reflected a functional insufficiency of the Sertoli cells with impaired spermatogenesis. This study shows that impairment of germinative functions is more severe and protracted than the impairment of the endocrine functions. K E Y WORDS:testis, drug therapy, germ cells, neoplasms
Germ cell tumors are the most frequent malignancies in men between 20 and 40 years o1d.I.' These neoplasms are classified as seminoma, nonseminoma, for example, teratoma, embryonal carcinoma, choriocarcinoma, yolk sac tumors and combined tumors." Treatment of germ cell tumors consists of orchiectomy and retroperitoneal lymphadenectomy, but in patients with disseminated disease, only cisplatin based chemotherapy is able to achieve a high cure rate.4-6 To date, the cure rate in these malignancies has increased to 90%. 1 Unfortunately, cytotoxic chemotherapy evokes acute and chronic side effects.7-") Chronic effects of cytotoxic treatment on gonadal functions require further investigation, because patients with germ cell tumors are of reproductive age Accepted for publication January 10, 1997
and some of them will be infertile after curative treatment. Leydig cells are more resistant to chemotherapy compared with germ cells, but endocrine functions (steroidogenesis, for example) are also affected in a dose-dependent manner." To our knowledge, there are at present few conclusive data about t h e duration of testicular impairment after chemotherapy or the gonadotoxic effects of different chemotherapy regimens for germ cell tumors. Therefore, the aims of the present investigation were to assess by hormonal evaluations the frequency and duration of gonadal impairment in patients after cisplatin based chemotherapy for germ cell tumors and the possibility of recovery as well as the gonadotoxic effects of different chemotherapy regimens. Luteinizing hormone and testosterone levels were measured to determine the function of the Leydig cells.Serum follicle-stimulating
844
a45
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
hormone was used a s an indirect biological indicator of Sertali cell functions. Elevated pretreatment and posttreatment follicle-stimulating hormone concentrations reflected a functional insufficiency of the Sertoli cells with an impaired spermatogenetic process.12-15 A significant inverse correlation has been demonstrated between the number of sperm and follicle-stimulating hormone concentrations.16 Additionally, the testosterone-to-luteinizing hormone ratio, which is considered to be a more sensitive marker of the function of the Leydig cells than testosterone or luteinizing hormone alone, was used to register the functions of the Leydig cells.17 MATERIAL A N D METHODS
Study design. In a transversal trial, serum concentrations of luteinizing hormone, follicle-stimulating hormone and testosterone were determined radioimmunologically in 232 patients with germ cell tumors up to 153 months after completion of cisplatin based chemotherapy. In this cross-sectional trial, hormone determinations were carried out once in each patient during posttreatment followup. At the time of the hormone evaluations disease was in complete remission, and alpha-fetoprotein and p-human chorionic gonadotropin were normal. Comparative hormone studies were performed in 73 of these patients within the first year after chemotherapy and in 28 with an interval of more than 8 years after cytotoxic treatment. No significant difference in the mean number of chemotherapy courses was observed in these 2 subsets of patients with germ cell tumors (3.41 5 0.2 versus 3.68 5 0.2 courses). Additionally, 51 of these patients were investigated in a longitudinal trial after unilateral orchiectomy before the onset of chemotherapy and during long-term followup 1, 3, 6, 12, 24, 36, 48 and 60 months after completion of chemotherapy. Eleven patients with germ cell tumors without cytotoxic treatment were studied within the first year after unilateral orchiectomy and retroperitoneal lymphadenectomy. A total of 105 healthy male volunteer, medical students and physicians of the University of Bonn (age 33.9 2 1.19 years, range 18 to 631, served a s controls. The study was approved by the ethics committee of the University of Bonn. Transversal trial. The study included 232 patients treated for germ cell tumors of the testis between 1983 and 1995 at the University of Bonn. Of these patients 213 (91.8%)underwent bilateral (138, 64.8%)or unilateral (75, 34.2%)retroperitoneal lymphadenectomy. Histology and the pathological stage of the tumors classified according to Mostofi et a13 are shown in table 1. The patients were treated with 2 to 12 cycles of different cisplatin based chemotherapy regimens: cisplatidvinblastinehleomycin (mean plus or minus standard error 3.13 5 0.06 courses, 70 cases, 30.2%),cisplatid
TABLE1. TheraDv related characteristics
of
histolom and pathological stage according to Mostofi et al' i n 232 germ cell tumors
Cisplati.inb,astine/
No. CisplatinNinblastinei Bleornycidfosfamide
Bleomycin (ri,I
(%)
No.
Histology: Seminoma Teratoma Embryonal Ca Chorioca Teratoca Chorioca and other germ cell tumors Seminoma and other germ cell tumors Nonclassifiable tumors Pathological stage: I IIA IIB IIC 111
vinblastinehleomycinhfosfamide (3.9 2 0.05, 62 cases, 26.74 ), cisplatidetoposidehleomycin (3.03 5 0.08, 73 cases, 3 1.4%~) and cisplatidvinblastinehleomycidifosfamide/etoposide (5.96 t 0.14, 27 cases, 11.64). Dose schedule was as follows: days 1 to 5, 20 mg./m.z cisplatin, days 1 and 2, 6.0 mgJm.2 vinblastine, days 1 to 5, 100 mg.lm.2 etoposide, days 2, 9 and 16, 30 IU bleomycin and days 1 to 5, 1.5 gm./m.2 ifosfamide daily. A total of 22 patients was treated initially according to the cisplatidvinblastinehleomycidifosfamide regimen followed by cisplatidetoposidehleomycin after vinblastine induced side effects (for example, aplasia). Another subset of 5 patients initially received cisplatidvinblastine/ bleomycidifosfamide followed by cisplatidetoposidehleomycin after a relapse of the disease which led t o cisplatid vinblastinehleomycin/ifosfamide/etoposide.Additionally, 11 patients with germ cell tumors (pathological stage I) were investigated within the first year (3.1 +- 1.16 months, range 1 to 12) after unilateral orchiectomy and retroperitoneal lymphadenectomy and were undergoing a followup protocol. Histology of the tumors, which were classified according to Mostofi e t al," were teratocarcinoma (6), embryonal carcinoma ( l ) ,choriocarcinoma (2), choriocarcinoma and other germ cell tumors (1) and nonclassifiable tumor (1). Longitudinal trial. This subset included 51 patients (age 30.1 2 0.1 year, range 18 to 64), of whom 47 had undergone unilateral retroperitoneal lymphadenectomy. The histology of the tumors in these patients was teratoma (2, 3.9%),teratocarcinomas (20,39.2%'), embryonal carcinoma (14,27.5%), choriocarcinoma combined with other germ cell tumors (1, 2.0%),seminoma (8, 15.7%),seminoma in combination with other germ cell tumors (4, 7.8%)and nonclassifiable tumors (2, 3.9%).Pathological stages of the tumors in the patients were stage I(10,19.6%),stage IIA (21,41.2%),stage IIB (12, 23.5%),stage IIC (4, 7.89) and stage I11 (4, 7.8%).The patients investigated in the longitudinal study were treated with cisplatidvinblastinehleomycin (3.25 ? 0.24 courses, range 2 to 4, 4, 7.8%),cisplatidvinblastinehleomycinhfosfamide (3.63 2 0.09, 2 to 4, 8, 15.7%),cisplatin/etoposide/ bleomycin (2.87 t 0.04, 1 to 6, 29, 56.9%)or cisplatidvinblastinehleomycidifosfamide/etoposide (6.0 5 0.25, 2 to 11, 9, 17.6%). Endocrine studies. Testicular function studies were performed during routine followup. Two blood samples were obtained in the early morning a t intervals of 20 minutes for biochemical studies, tumor marker and hormone determinations. The serum was separated, pooled and stored a t -2OC until assay. Serum concentrations of luteinizing hormone and follicle-stimulating hormone were determined in dupli-
No.
Cisplatin/Etoposide/ Bleomycin (~
No.
CisplatinA'inblastine/ Bleomycidfosfamide/ Etoposide I r/, 1
Total No 1% I
9133.31 3tll.11 3111.11
17 (27.41 5 18.11
2 (2.71 6 (8.21 22 130.21 111.41 25 134.31 1 11.4)
6 122.2, 1 (3.71
23 I991 28 t 12 11 48(2071 1 I041 63 I27 21 14 I601
9 112.91
9 114.51
8 110.9)
3 111.11
29 (12 51
4 15.71
12 (19.31
8 t10.91
2 17.51
26 Ill 21
7 121.41 10 (18.61 23 (32.9) 15 (21.41 15 (21.4)
3 (4.81 10 (16.1I 12 (19.41 13 121.0) 24 (38.71
10 113.7) 19 126.0) 27 (37.01 11 115.1) 6 18.2)
5118.51 3 (11.1) 4 114.8) 6 (22.2) 9 (33.3)
25(1081 4 2 ( 1 8 1) 66(284) 45!194) 54 (23 3)
7 11O.Ol 15 121.41 3 118.61
5 18.1) 4 16.51 10 (16.11
15 (21.41 7 l10.01
846
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
months after orchiectomy and retroperitoneal l p p h a d e n e c bmy. After chemotherapy, mean luteinizing hormone concentration in all patients was within the normal range, but it was significantly higher in patients than i n healthy controls (5.78 t 0.28 versus 2.77 i- 0.11 I U k , p <0.001) as well a s in patients without cytotoxic treatment within the first year after surgery (3.4 2 0.48 IUfl., p <0.01). Serum luteinizing hormone levels were elevated in 31 of t h e 232 patients (13.4%).Of 73 patients studied within t h e first year after chemotherapy, serum luteinizing hormone levels were elevated in 24 patients (32.9%,fig. 1, A ) . Of 28 patients studied more than 8 years after chemotherapy (median 8.5 years, range 8.0 to 12.6),only 1 (3.6%) had elevated serum luteinizing hormone levels, which reflects a significant decrease in the percentage of patients with a n impairment of the Leydig cell function ( p <0.01), depending on the interval since treatment. Mean luteinizing hormone concentrations did not differ significantly in patients treated with cisplatidvinblastinehleomycin (5.37 i- 0.29 IUA.) or cisplatidvinblastine/ bleomycidifosfamide (5.73 2 0.27 IUA.) and patients treated with cisplatidetoposidehleomycin had serum luteinizing hormone levels of 5.81 -t 0.22 IUA. Germ cell tumor patients treated with cisplatin/vinblastinehleomycin/ifosfamide/ etoposide had higher serum luteinizing hormone concentrations (6.87 -t 0.38 IUA.) than those treated with the other regimens. Mean testosterone concentration in all patients was within RESULTS normal range and did not differ significantly from healthy Transi)ersaltrial. The mean age of the 232 patients treated controls or surveillance patients after surgical treatment for germ cell tumors was 31.5 t 0.51 years (range 18 to 64). (18.1 0.43 versus 18.5 -t 0.51 and 19.0 i- 1.15 nmo1.A.). Of No significant differences were observed i n the mean age these 232 patients 12 (5.2%),including 5 of 73 patients studof patients treated with cisplatidvinblastinehleomycin ied within the first year after chemotherapy (6.8%), had (33.6 2 0.57 years), cisplatin/vinblastinehleomycidifos- subnormal testosterone concentrations (range 2.15 to 9.2 famide ( 3 2 . 9 i- 0.46 y e a r s ) a n d cisplatin/vinblastine/ nmolA., fig. 1, B ) . Of 28 patients studied longer than 8 years bleomycidifosfamide/etoposide (32.8 -t 0.47 years), respec- after chemotherapy (median 8.5 years, range 8.0 to 12.6), 2 tively, b u t p a t i e n t s t r e a t e d with cisplatin/etoposide/ (7.1%) had subnormal serum testosterone (5.34 and 7.95 bleomycin were significantly younger (27.9 -t 0.42 years, p nmo1.A. j. A recovery of endocrine functions, reflected by the <0.01) because of t h e later introduction of this chemotherapy percentage of patients with subnormal testosterone, could regimen. The interval between the completion of chemother- not be achieved. No significant differences in testosterone apy and hormone determinations did not differ significantly concentrations could be observed between patients treated between patients treated with cisplatidvinblastinehleomy- with cisplatidvinblastinehleomycin(17.5 2 0.43 nmo1.A.) or cin (45.2 t 2.28 months, range 1 to 134) and those treated cisplatin/vinblastine/bleomycidifosfamide( 18.1 t- 0.4 nmo1.i with cisplatidvinblastinehleomycidifosfamide(50.2 i- 2.39 1.). Patients treated with cisplatin/etoposidehleomycin months, 1 to 153),but this interval was significantly shorter (18.8 t 0.49 nmo1.fl.j and those treated with cisplatinl for patients treated with cisplatidetoposidehleomycin(18.8 2 vinblastinehleomycin/ifosfamide/etoposide h a d normal 1.1 months, 1 to 72, p <0.001) and cisplatidvinblastine/ mean testosterone (17.5 2 0.35 nmo1.A.). Mean folliclebleornycinhfosfamidetoposide (41.2 i- 2.24 months, 1 to 123, p stimulating hormone concentration in all patients was sig(0.05). Additionally, 11 patients with pathological stage I germ nificantly increased compared with healthy controls (18.8 -t cell tumors (age 30.2 -t 1.48 years, range 22 to 38) undergoing 0.93 versus 4.02 t 0.17 IUA., p <0.001) and to the 11 germ followup without chemotherapy were investigated 3.1 t 1.16 cell tumor patients without chemotherapy within the first year after orchiectomy and lymphadenectomy (8.5 2 1.15 * MAIAclone, Serono, Freiburg, Germany. IUA., p <0.01).Of these 11patients under surveillance 3 had t RSL, ICN Biomedicals, Inc., Costa Mesa, California
cate by immunoradiometric assay* and concentrations of testosterone by radioimmunoassay.: Normal ranges in men are 1.8 to 9.2 IUA. for luteinizing hormone, 1.6 to 9.0 IUA. for follicle-stimulating hormone, and 10.4 to 34.7 nmo1.A. for testosterone. The within-assay variation for luteinizing hormone (concentration 2.1 IUA. ), follicle-stimulating hormone (5.5 IUA.) and testosterone (12.5 nmo1.A.) was 3.5, 3.9 and 9.5%(20),respectively and the between-assay variation was 9.4, 5.3 and 9.8% (18). respectively. The testosterone-toluteinizing hormone ratio was used to describe the endocrine function of the Leydig cells more profoundly than luteinizing hormone or testosterone alone.” The normal range for the testosterone-to-luteinizing hormone ratio is 3.3 to 17.9 nmol./IU (105 men). Semen analyses. Semen analyses were performed simultaneously with serum follicle-stimulating hormone analyses in 46 patients with germ cell tumors using conventional methods to correlate sperm counts with serum follicle-stimulating hormone concentrations. I n each case, the abstinence interval was 4 days. Statistical analyses. Results were expressed as mean plus or minus standard error, as ranges and as the 5th and 95th percentile to establish reference values for the testosterone/ luteinizing hormone quotient. Statistical analyses were performed using the Mann-Whitney U test, the Wilcoxon test and a Spearman’s rank correlation (r). A p value <0.05 was considered significant.
A lo]..
T i m a(ur clumothuapy Imonth.)
Time after chamothenpy [monlhal
Time aner chemotherapy [months1
FIG. 1. Individual serum concentrations of luteinizing hormone (A), testosterone ( B , and follicle-stimulating hormone ( C Iin 232 patients with germ cell tumors u to 153 months after completion of chemotherapy. Lower and/or upper normal range is marked h y dottcd lines. L H . luteinizing hormone. F&f , follicle-stimulating hormone.
847
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
elevated follicle-stimulating hormone concentrations (27.3%, 11.0 to 16.2 IUI1.J. Serum follicle-stimulating hormone showed highest concentrations during the first year after therapy (fig. 1, C,, when 65 of 73 patients (89.0%)had elevated follicle-stimulating hormone (range 9.1 to 88.9 IUA.), then follicle-stimulating hormone gradually decreased. More than 8 years after chemotherapy (median 8.5 years, range 8.0 to 12.6),mean follicle-stimulating hormone was still elevated in 28 patients (14.3 2 2.48 IUA., median 12.5, range 3.3 to 41.5). A significantly lower percentage of patients with elevated serum follicle-stimulating hormone was evident ( p ~ 0 . 0 ),5 reflected by elevated follicle-stimulating hormone concentrations (range 10.2 to 41.5 IUA.) in 18 of these 28 patients. Still, a high percentage of patients showed elevated serum follicle-stimulating hormone as an indirect marker of Sertoli cell insufficiency (64.3%).An inverse correlation was observed between sperm count and follicle-stimulating hormone concentrations in 46 patients treated for germ cell tumors ( r = -0.77, p <0.001, fig. 2). Mean folliclestimulating hormone concentrations did not show significant differences in patients treated with cisplatin/vinblastine/ bleomycin ( 17.4 t 0.99 IU/l.) or cisplatin/vinblastine/ bleomycidifosfamide ( 18.7 ? 0.78 IUA. ). Cisplatidetoposidel bleomycin (18.6 -+ 0.86 IUA.) a s well as those treated with cisplatin/vinblastinemleomycin/ifosfamide/etoposide also showed elevated serum follicle-stimulating hormone concentrations (23.6 2 1.22 IUA.). Within the first year after completion of chemotherapy, the mean testosterone-to-luteinizing hormone ratio was within normal range but significantly lower in the patients (3.41 2 0.44 nmol./IU, p <0.001, 73) compared with the healthy controls (7.71 f 0.37 nmol./IU.) as well as the patients undergoing a followup protocol without chemotherapy (6.46 2 0.8 nmo1.A.). Of 73 patients 44 (60.3%)showed a subnormal testosterone-to-luteinizing hormone ratio (range 0.33 to 3.11 nmol./IU), whereas 1 patient in the surveillance group (9.1%) showed a subnormal testosterone-to-luteinizing hormone ratio (2.66 nmol./IU). More than 8 years after chemotherapy (median 8.5 years, range 8.0 to 12.6), the mean testosteroneto-luteinizing hormone ratio was within normal range but still significantly lower than in controls (5.01 2 0.56 nmol./ IU, p <0.01, 28). A significantly lower percentage of patients with a subnormal testosterone-to-luteinizinghormone ratio was observed (32.1%, 9 of 28 patients, range 0.67 to 2.74 nmol./IU, p <0.05) compared with patients within the first year after chemotherapy.
60
-
50
-0
-G
40
-
I
30-1.
2
2
To elucidate the influence of different chemotherapy regimens (cisplatidvinblastinehleomycin,cisplatin/vinblastine/ bleomycidifosfamide, cisplatidetoposidehleomycin, cisplatidvinblastinehleomycin/ifosfamide/etoposide)on germinative functions, we analyzed the percentage of patients with elevated follicle-stimulating hormone concentrations a t various times and intervals after chemotherapy. Within the first year a h r chemotherapy, a high percentage of patients with elevated serum follicle-stimulating hormone could be observed in all different chemotherapy regimens (cisplatidvinblastinehleomycin, 84.2% of 19 patients; cisplatidvinblastinehleomycidifosfamide, 90% of 20; cisplatidetoposidehleomycin,81.2% of 38; cisplatdvinblastinehleomycin/ifosfamide/etoposide,100% of 12). At 37 to 48 months after treatment the rate of elevated follicle-stimulating hormones was 60% in 6 of 10 patients treated with cisplatidetoposidehleomycin, 66.7% in 8 of 12 treated with cisplatidvinblastinehleomycin, 63.6% in 7 of 11 treated with cisplatidvinblastine/bleomycin/ifosfamide and 80% in 4 of 5 treated with cisplatidvinblastinehleomycidifosfamiddetoposide. At 48 to 60 months after treatment, 63.6% of the patients treated with cisplatidvinblastinehleomycin(7 of 11 patients) and 55.6% of the patients treated with cisplatid vinblastinehleomycidifosfamide (5 of 9) still had elevated follicle-stimulating hormone concentrations. Up to 72 months aRer chemotherapy, 504 of patients (5 of 10) treated with cisplatidvinblastinehleomycin still showed elevated serum follicle-stimulating hormone (fig. 3). Longitudinal trial. After orchiectomy and before chemotherapy, mean Iuteinizing hormone and testosterone concentrations were within the normal range, but mean luteinizing hormone was significantly higher than in healthy controls (3.51 ? 0.37 versus 2.77 -+ 0.11 IUA., p <0.001). One patient had supranormal serum luteinizing hormone ( 16.6 IUA.) and 4 patients had subnormal testosterone concentrations (4.4 to 7.9 nmo1.A. ). Mean serum follicle-stimulating hormone levels were significantly higher than in controls (7.79 -+ 0.13 versus 4.02 2 0.17 IUA., p <0.001), but mean follicle-stimulating hormone was still within the normal range. Before chemotherapy, 34 of these 51 patients (66.7%:) had folliclestimulating hormone concentrations within normal range, whereas 17 patients (33.3%) had elevated serum folliclestimulating hormone (range 9.4 to 26.6 IUA. ). After chemotherapy, mean luteinizing hormone and testosterone were within normal range (fig. 4, A and B However,
i
20
-
.0
1-12
OJ,
0
1
I
I
I
1
20
40
60
80
100
Sperm count [millJmL] FIG. 2. Inverse correlation of serum follicle-stimulatinghormone ( F S H )and sperm count in 46 patients treated for germ cell tumors.
13-24
25-36
37-48
49-60
61-72
Time after chemotherapy [months]
FIG. 3. Percentage of patients with elevated serum folliclestimulating hormone ( F S H ) concentrations after treatment with
different chemotherapy schedules for germ cell tumors. Small numbers on bars indicate number of patients with elevated folliclestimulating hormone concentrations compared with total number of patients.
848
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
... ...p 47
...
...
2'
12
6
0
24
36
0
60
48
12
6
24
36
48
60
Time aRer chemotherapy [months]
Time after chemotherapy [months] 301
*l
0
6
12
24
36
48
0
60
6
12
24
36
48
60
Time after chemotherapy [months]
Time after chemotherapy [months]
FIG.4. Serum concentrations of luteinizing hormone ( L H , A), testosterone ( B ) ,follicle-stimulating hormone (FSH, C) and testosteroneto-luteinizing hormone ratio tD) expressed as mean plus or minus standard error before and after chemotherapy for germ cell tumors performed in longitudinal trial. Numbers of patients are given on top of bars. *** p <0.001.** p 10.01.* p c0.05.
elevated serum luteinizing hormone was observed after 1 month of chemotherapy in 9 of 47 patients (19.1%),after 3 months in 12 of 38 (31.64), after 6 months in 5 of 39 (12.8%), after 12 months in 6 of 41 (14.6%1, after 24 months in 2 of 36 (5.55%)and after 36 months in 2 of 25 patients (7.69%).At 48
and 60 months after chemotherapy, supranormal serum luteinizing hormone was observed in 1 of 23 (4.3%)and in 2 of 19 patients (5.3%),respectively. Highest mean luteinizing hormone concentrations were observed 3 months after chemotherapy (7.17 5 0.1 I U k , 38), but up to 60 months after
TABLE2 Treatment interoal (months) and hormone concentratrons (expressed as mean plus or minus standard error, range and median) i n patients with germ cell tumors followed in a longitudinal study before and after chemotherapy ~~
Time After Chemotherapy Irnos.1 Before Range Median
~
~~
No.
47 38
38
41
24 Range Median
37
36 Range Median
27
48 Range Median 60 Rnnge Median -_ -
~~
FollicleStimulating Hormone tIU/l.,
7.792 0.13 1.0-26.6 7.8 20.9 z 0.25 1.7-59.7 19.4 23.9 2 0.3 1.O-54.1 20.8 24.4t 0.34 1.3-66.8 22.5 20.55 0.34 1.540.5 16.8 17.0 i 0.34 3.641.0 14.0 16.5t 0.42 3.8-51.7 13.9 15.2:0.41 3.S38.7 13.3 16.9I 0 71 2.1-579 14.9
6.0
12 Range Median
~~
3.45t 0.05 1.0-16.6 2.9 6.77 z 0.1 1.0-22.8
51
3 Range Median 6 Range Median
~~
(1UA.i
1 Range Median
~
Luteinizing Hormone
7.17f 0.11 1.2-17.9 5.35 5.94I0.1 1.4-21.6 4.7 5.68 f 0.09 i.sia.5 4.5 4.67t 0.49 1..5-17.1 4.1 5.29 r 0.1 1.0-13.3 4.9 4.83 0.08 2.8-9.2 4.2 5.26* 0.16 2.1-14.2 4.1 f
23 19 ~.
~
~~
~~~
Testosterone
(nmol./l.)
Testosterone-to-Luteinizing Hormone Ratio (nmol./lU~
18.61 0.17 4.44-53.9 17.8 21.5z 0.24 0.56-69.4 20.9 19.6t 0.19 10.7-52.4 18.7 18.5 t 0.13 9.9-30.3 17.8 18.3z 0.17 8.043.1 16.8 17.8t 0.14 9.96-32.8 16.9 16.2 i 0.2 7.9-29.5 15.7 17.6I0.22 1o.R-32.9 17.3 17.5z 0.31 10.1-29.3 16.4
6.38 I0.72 1.12-16.1
~~~
3.972 0.4 0.23-12.7 3.59 4.1 t 0.73 0.73-10.1 3.3 4.12 I0.52 0.85-7.1 3.67 4.632 0.59 0.92-13.5 3.67 4.76f 0.43 0.58-12.5 4.12 3.27? 0.27 0.89-5.45 3.43 4.33I 0.52 1.76-11.9 3.59 4.73I0.59 1.91-8.17 4.84 .---
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
treatment, luteinizing hormone was still significantly higher than before chemotherapy (5.26 -+ 0.16 versus 3.45 -+ 0.05 IUA., p <0.001). Mean serum testosterone was almost within normal range and vaned between 16.5 and 21.5 nmo1.A. (table 2). A subnormal serum testosterone was observed in 2.5 to 4.2%of the patients under observation. Serum follicle-stimulating hormone increased with a significant 6-month peak (24.4 5 0.34 IUA., p <0.001, 38, fig. 4, C). Of 38 patients 37 (97.4%)had elevated folliclestimulating hormone concentrations (range 10.7 to 66.8 IU/ 1. ). Then follicle-stimulating hormone gradually decreased, but 60 months after chemotherapy, mean follicle-stimulating hormone was still significantly elevated (16.9 ? 0.71 IUA., p <0.01, 19) compared with the pretreatment values. Of 19 patients 13 (68.4%) still showed elevated serum folliclestimulating hormone concentrations (range 9.6 to 43.4 IUA.). Before chemotherapy, no significant difference could be observed in the mean testosterone-to-luteinizing hormone ratio of germ cell tumor patients and healthy controls (6.38 ? 0.72 versus 7.71 t 0.37 nmol./IU) and 12 patients (23.1%) showed a subnormal testosterone-to-luteinizing hormone ratio. One month after chemotherapy, the testosterone-toluteinizing hormone ratio significantly declined (3.97 ? 0.40 nmol./IU, p <0.05, fig. 4, D),and 3 months after treatment, a maximum of 52.6%of patients (20 of 38 patients) showed a subnormal ratio. Sixty months after therapy, this ratio still was significantly lower (4.73 -+ 0.59 nmol./IU, p <0.05) than the pretreatment ratio and the testosterone-to-luteinizing hormone ratio in healthy controls, but within the lower normal range. Of 19 patients 6 (31.6%)investigated 60 months after chemotherapy still showed a subnormal testosteroneto-luteinizing hormone ratio (1.1 to 2.6 nmol./IU).
849
Zated. Our data are supported by Stephenson et al,Z0 who -eported on oligospermia or azoospermia in 19 of 30 patients 63.370) treated with 2 t o 4 courses of chemotherapy according to the cisplatidetoposidehleomycin regimen and by Sietema et al,” who reported on a protracted luteinizing hormone and follicle-stimulating hormone elevation more than 5 years after 4 cycles of chemotherapy. In a small collective of 25 patients, after cisplatidvinblastinehleomycin treatment for metastatic testicular cancer, Hansen et a1 demonstrated that azoospermia or oligospermia was associated with elevated serum follicle-stimulating hormone within the first 2 years after treatment, followed by a gradual recovery in the further followup.18All patients showed at least partial regeneration of spermatogenesis, but 5 years after therapy, sperm density had returned to pretreatment sperm counts in only 46%. Hansen et a1 reported on higher serum folliclestimulating hormone and lower sperm density in patients treated with cisplatidvinblastinehleomycin compared with patients treated with orchiectomy alone, but this “control collective” was older than the chemotherapy-treated patients.ls To our knowledge, no conclusive data were available concerning the comparison of gonadotoxic effects- especially the influence on spermatogenesis reflected by elevated folliclestimulating hormone concentrations-of different cisplatin based chemotherapy regimens for germ cell tumors. In the first 12 months after chemotherapy, there was no difference in the percentage of patients with elevated serum folliclestimulating hormone levels in these 4 therapy groups. During the followup, the percentage of patients with elevated follicle-stimulating hormone levels decreased in the cisplatid vinblastinehleomycin, cisplatidvinblastinehleomycidifosfamide and cisplatidetoposidehleomycin groups, which reflects a gradual recovery of spermatogenesis in some of the DISCUSSION patients. Patients treated with cisplatidvinblastinehleomyIn our transversal study, we had the opportunity to screen cidifosfamide/etoposide had elevated follicle-stimulating the gonadal function of 232 patients with germ cell tumors up hormone concentrations without a clear decrease most freto 12.6 years after completion of cisplatin based chemother- quently because of the higher dose of cisplatin in this subapy. We observed a n elevation of serum luteinizing hormone group. These data are in agreement with those of Hansen et in 24 and subnormal testosterone in 5 of 73 patients studied a1.18 who found dose-dependent gonadal damage from use of within the first year after chemotherapy, which reflects a n cytostatic drugs. Now, 48 months after completion of chemoimpairment of the function of the Leydig cells. The majority therapy, no conclusion can yet be drawn as to which regimen of patients had luteinizing hormone and testosterone concen- has the least gonadotoxic side effects. The data of the cross-sectional trial were confirmed by our trations within normal range. Longer than 8 years after chemotherapy, 1 of 28 patients still had elevated luteinizing longitudinal study. After orchiectomy and before chemotherhormone and 2 patients had subnormal testosterone concen- apy, 1 of 51 patients showed elevated luteinizing hormone trations. Our results agree with those of Hansenls and and 2 patients showed subnormal testosterone concentraFossa19 et a1 who reported on luteinizing hormone elevation tions, which reflects a n impaired function of the Leydig cells. and subnormal testosterone concentrations as signs of dis- In accordance with several investigators, mean luteinizing turbed Leydig cell function in some patients treated for germ hormone and testosterone concentrations were within norcell tumors with cisplatin based chemotherapy. Levels of mal range after chemotherapy.18.19 However, mean luteinizSerum follicle-stimulating hormone-a sensitive parameter ing hormone was still significantly higher years after comof germinative function12-1s (fig. 2 t w e r e significantly ele- pletion of chemotherapy compared with pretreatment values, vated after chemotherapy. In the first year after treatment, which reflects a compensated insufficiency of the Leydig 65 of 73 patients (89%)showed elevated follicle-stimulating cells. In addition, we used the testosterone-to-luteinizing horhormone concentrations that reflected spermatogenesis impairment and follicle-stimulating hormone then gradually mone ratio, which is considered to be a more sensitive marker decreased. More than 8 years after chemotherapy, however, of the Leydig cell function than testosterone or luteinizing 18 of 28 patients (64.3%)still had elevated serum follicle- hormone alone.17 Before chemotherapy and after orchiecstimulating hormone. Thus, a significant disturbance of sper- tomy, the mean testosterone-to-luteinizinghormone ratio did matogenesis reflected by elevated follicle-stimulating hor- not differ significantly among patients, healthy controls and mone concentrations was still evident in the majority of the germ cell tumor patients without chemotherapy, excluding patients investigated. Our data contrast with the observa- an impairment of the function of the Leydig cells caused tions of Fossa et a1,19 who reported on median follicle- by orchiectomy. Twelve patients (23.1%) had subnormal stimulating hormone levels within the upper normal range in testosterone-to-luteinizing hormone ratios and 3 months af47 germ cell tumor patients with a median followup of 9 ter treatment, a maximum of 52.6%of the patients showed a Years after chemotherapy generally treated according t o the subnormal ratio. For up to 60 months in the postclsphtidvinblastinehleomycin schedule, which may imply chemotherapy period, the testosterone-to-luteinizing horthat many of these patients had supranormal follicle- mone ratio declined and was significantly lower in patients stimulating hormone levels. Unfortunately, the authors didI than in controls. However, 60 months after completion of not show individual hormone data of the patients investi- chemotherapy 31.6% still had a subnormal testosterone-to-
850
CISPLATIN BASED CHEMOTHERAPY AND GONADAL FUNCTION
luteinizing hormone ratio, which reflects a g r a d u a l recovery of Leydig cell function. This finding suggests that y e a r s after chemotherapy, higher endogenous luteinizing hormone secretion i s still necessary t o s t i m u l a t e testosterone synthesis and/or secretion t o t h e s a m e degree as in healthy subjects. These findings confirm that a f t e r chemotherapy, a n impairm e n t of testosterone synthesis/secretion-although generally still p r e s e n t in a compensated form-is based on a primary testicular defect in p a t i e n t s treated for g e r m cell tumors. This i m p a i r m e n t c a n be caused directly by chemotherapy a n d by reduced testicular blood flow a f t e r treatment w i t h cytotoxic agents.22 The clinical relevance of compensated Leydig cell disturbance in p a t i e n t s treated for g e r m cell tumors m u s t be illuminated by sexual a n a m n e s i s that details reduction of libido a n d erectile dysfunction. Before chemotherapy, 17 of 51 p a t i e n t s with germ cell t u m o r s (33.3%) had elevated follicle-stimulating hormone concentrations. which reflects a n insufficiency of Sertoli cell function. After completion of chemotherapy, s e r u m folliclestimulating hormone levels increased, with a significant peak a t 6 months, t h e n gradually decreased. Even 60 m o n t h s after chemotherapy, follicle-stimulating hormone w a s significantly higher t h a n before t r e a t m e n t and an elevation of s e r u m follicle-stimulating hormone was still obvious in m o r e t h a n 68%of the patients under observation. CONCLUSIONS
A compensated insufficiency of Leydig cell function reflected b y p e a k levels of s e r u m luteinizing hormone and a low testosterone-to-luteinizinghormone ratio occurs i n u p to 528 of patients with g e r m cell t u m o r s within the first m o n t h s after chemotherapy, but t h i s generally compensated impairment of the Leydig cells w a s still evident i n a b o u t 31%of the patients 60 m o n t h s a f t e r completion of treatment. The clinical relevance of t h i s compensated endocrine disturbance still needs to be investigated. The elevation of s e r u m folliclestimulating hormone, which reflects an i m p a i r m e n t of germinative functions, i s more severe a n d protracted 6 m o n t h s after chemotherapy. At 60 m o n t h s after chemotherapy, more t h a n 68% of the p a t i e n t s still showed elevated folliclestimulating hormone concentrations. Because some of these patients will be infertile after curative t r e a t m e n t , the option of s p e r m cryopreservation should be considered for patients with g e r m cell t u m o r s before chemotherapy. To date i n our study, i t could not be decided which chemotherapy regimen investigated had the l e a s t gonadotoxic side effects.
Mrs. M. Lafosse, I. Horn a n d U. Bartholomay provided technical assistance. REFERENCES
1. Fossa, S. D., Aass, S. D. and Kaalhus, 0.: Testicular cancer in young Norwegians. J . Surg. Oncol., 3 9 43, 1988. 2. Mackay, E. N. and Sellers, A. H.: A statistical review of malignant testicular tumours based on the experience of the Ontario Cancer Foundation Clinics, 1938-1961. Canad. Med. Ass. J., %k889, 1966. 3. Mostofi, F. K. and Sellers, A. H.: Histology typing of testis tumours. Geneva, Switzerland: World Health Organization, 1977. 4. Wheeler, J. E.: The history of teratomas. In: The Teratomas.
Edited by I. Damjanov, B. B. Knowles and D. Solter. Clifton, New Jersey: Humana Press, pp. 1-22. 5. Nichols, C. R. and Roth, J. B.: Management of metastatic nonseminoma, development of effective chemotherapy. In: Testicular Cancer: Investigation and Management. Edited by A. Horwich. New York: Chapman and Hall Medical, chapt. 14, pp. 185-203. 6. Einhorn, L. H. and Williams, S. D.: Chemotherapy of disseminated testicular cancer. A random prospective study. Cancer, 4 6 1339, 1980. 7. Ellis, M. and Sikora, K.: The current management of testicular cancer. Brit. J. Urol., 5 9 2, 1987. 8. Hartlapp, J. H.. Weissbach, L. and Bussar-Maatz, R.: Adjuvant chemotherapy in nonseminomatous testicular tumour stage 11. Int. J. Androl., 10: 277, 1987. 9. Klingmiiller, D., Brenneman, W., Brensing, K. A., Deitenbeck, R., Stoffel-Wagner, B., Hartlapp, J. H., Jager, N. and Mumperow, E.: Hodenfunktionsstorungen bei Patienten mit Hodentumoren. Urologe LBI, 32: 9, 1992. 10. Brenneman, W., Stoffel-Wagner, B., Bidlingmaier, F. and Klingmiiller, D.: Immunoreactive plasma inhibin levels in men after polyvalent chemotherapy of germinal cell cancer. Acta Endocr., 126 224, 1992. 11. Grigor, K. M. and Donohue, J. P.: Reproductive aspects of testicular germ cell cancer: general discussion. Eur. Urol., 23: 177, 1993. 12. Palmieri, G., Lotrecchiano, G . , Ricci, G., Spiezia, R., Lombardi, G., Bianco, A. R. and Torino, G.: Gonadal function after multimodality treatment in men with testicular germ cell cancer. Eur. J. Endocr., 134: 431, 1996. 13. Kreuser, E. D., Harsch, U., Hetzel, W. D. and Schreml, W.: Chronic gonadal toxicity in patients with testicular cancer after chemotherapy. Eur. J. Cancer Clin. Oncol., 22: 289,1986. 14. de Kretser, D. M., Burger, H. G. and Hudson, B.: The relationship between germinal cells and serum FSH levels in males with infertility. J . Clin. Endocr. Metab., 3 8 787, 1974. 15. de Kretser, D. M., Burger, H. G. and Bremner, W. J.: Control of FSH and LH secretion. Monogr. Endocr., 3 5 12, 1983. 16. Franchimot, O., Millet, D., Vendrely, E., Letawe, J., Legros, J. J. and Netter, A,: Relationship between spermatogenesis and serum gonadotropin levels in azoospermia and oligospermia. J. Clin. Endocr. Metab., 3 4 1003, 1972. 17. Giagulli, V. A. and Vermeulen, A,: Leydig cell function in infertile men with idiopathic oligospermic infertility. J. Clin. Endocr. Metab., 66:62, 1988. 18. Hansen, P. V., Trykker, H., Helkjoer, P. E. and Andersen, J.: Testicular function in patients with testicular cancer treated with orchiectomy alone or orchiectomy plus cisplatin-based chemotherapy. J. Natl. Cancer Inst., 81: 1246. 1989. 19. Fossa, S. D., Lehne, G., Heimdal, K. and Theodorsen, L.: Chemical and biochemical long-term toxicity after postoperative cisplatin-based chemotherapy in patients with low-stage testicular cancer. Oncology, 53: 300, 1995. 20. Stephenson, W. T., Poirier, S. M., Rubin, L. and Einhorn, L. H.: Evaluation of reproductive capacity in germ cell tumor patients following treatment with cisplatin, etoposide and bleomycin. J. Clin. Oncol., 13: 2278, 1995. 21. Gietema, J. A., Sleijfer, D. T., Willemse, P. H., Schraffordt Koops, H., van Ittersum, E., Verschuren, W. M., Kromhout, D., Sluiter, W. J., Mulder, N. H. and de Vries, E. G.: Long-term follow-up of cardiovascular risk factors in patients given chemotherapy for disseminated nonseminomatous testicular cancer. Ann. Intern. Med., 116 709, 1992. 22. Huane. H. F.. Poaach. - L. M.. Nathan, E. and Gielio. W.: Acute and chronic affects of cisplatinium upon testicular function in the rat. J. Androl., 11: 436, 1990. I
Y