- Email: [email protected]
Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin
EUF-249; No. of Pages 8 EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
available at www.sciencedirect.com journal homepage: www.europeanurology.com/eufocus
Testis Cancer
Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin Mikkel Bandak a,*, Niels Jørgensen b,c, Anders Juul b,c, Jakob Lauritsen a, Maria Gry Gundgaard Kier a,d, Mette Saksø Mortensen a, Gedske Daugaard a a
Department of Oncology, Copenhagen University Hospital, Rigshospitalet, Denmark;
b
Department of Growth and Reproduction, Copenhagen University
Hospital, Rigshospitalet, Denmark; c International Center for Research and Research Training in Endocrine Disruption of Male Reproduction and Child Health (EDMaRC), Rigshospitalet, University of Copenhagen, Denmark;
d
Unit of Survivorship, Danish Cancer Society Research Center, Danish Cancer Society,
Copenhagen, Denmark
Article info
Abstract
Article history: Accepted November 23, 2016
Background: Controversy exists whether Leydig cells recover after testicular cancer (TC) treatment or whether premature hormonal aging will occur. Objective: Evaluate serial changes in total testosterone (TT) and luteinizing hormone (LH) in patients treated with orchiectomy alone (Stage I) or combination chemotherapy with bleomycin, etoposide, and cisplatin (BEP). Design, settings, and participants: Changes in TT and LH were investigated during 5-yr follow-up (Stage I: n = 75, BEP: n = 81). A selected group of TC patients with mild Leydig cell dysfunction (LH 8 IU/l) were followed for a longer period (Stage I: n = 20, BEP: n = 23). An age-matched control group of 839 healthy men served as controls for TT and LH levels. Outcome measurements and statistical analysis: Changes in age-adjusted TT and LH were evaluated separately in each treatment group with univariate linear regression analysis. The proportion of patients initiating testosterone substitution during follow-up was calculated. Results and limitations: In the 75 Stage I patients there were no significant changes in LH and TT, while in the 81 BEP treated patients there was a significant decline in LH during follow-up (–24.2 percentage point/yr, 95% confidence interval: –38.5 to –9.9, p = 0.001). In total, 11% of Stage I patients and 15% of BEP-treated patients initiated testosterone substitution. In the 23 BEP-treated patients with mild Leydig cell dysfunction there was a significant decline in age-adjusted TT (–0.9 percentage point/yr, 95% confidence interval: –1.8 to –0.04, p = 0.04), while in the 20 Stage I patients there were no significant changes in age-adjusted LH and TT. Limitations include the retrospective study design. Conclusions: TT remained stable during 5-yr follow-up in TC patients treated with orchiectomy alone or BEP. BEP-treated patients with mild Leydig cell dysfunction during follow-up were at risk of long-term testicular failure and evaluation of Leydig cell function beyond follow-up should be considered in this group of patients. Patient summary: This study shows that the majority of testicular cancer survivors had stable testosterone levels after treatment for testicular cancer. However, 11–15% of patients needed testosterone substitution after treatment. # 2016 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Associate Editor: James Catto Keywords: Testicular cancer Testosterone Luteinizing hormone Leydig cell function
* Corresponding author. Department of Oncology, Copenhagen University Hospital, Rigshospitalet, 2100 Copenhagen, Denmark. Tel. +45 3545 6354; Fax: +45 3545 3909. E-mail address: [email protected] (M. Bandak). http://dx.doi.org/10.1016/j.euf.2016.11.018 2405-4569/# 2016 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 2
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
1.
Introduction
Testosterone substitution was the exclusion criteria. No patients in this cohort were included in the 5-yr follow-up cohort. A flow-chart of
It has been hypothesized that cisplatin-based chemotherapy, abdominal radiotherapy, and orchiectomy alone induce premature hormonal aging in testicular cancer (TC) survivors. Thus, it has been proposed to continue clinical followup beyond 5 yr [1]. However, it has also been suggested, that Leydig cells recover in the years following TC treatment [2,3]. In a recent meta-analysis, we found that cisplatinbased chemotherapy and abdominal radiotherapy increased the risk of low levels of total testosterone (TT) or use of testosterone substitution, when compared with orchiectomy alone [4]. However, most studies of Leydig cell function in TC survivors are based on a single analysis of serum levels of luteinizing hormone (LH) and TT after completion of treatment [5–9]. To clarify whether TC treatment induces premature hormonal aging and/or hormonal recovery takes place over time, serial measurements of TT and LH are needed. To answer these questions, we aimed to: (1) evaluate changes in age-adjusted TT and LH, (2) evaluate the proportion of TC patients who initiated testosterone substitution within a standard 5-yr follow-up program in an unselected cohort of TC patients treated with orchiectomy alone or BEP, (3) evaluate long-term changes in age-adjusted TT and LH in a selected group of TC patients who presented with mild Leydig cell dysfunction (elevated serum levels of LH in combination with low or normal TT levels) during the initial 5-yr follow-up period. As it has been suggested that Leydig cell function is better evaluated by combined assessment of LH and TT [10], we also estimated the proportion of patients with an abnormal LH/TT during follow-up.
included patients in this part of the study can be seen in Figure 1. As controls for the reproductive hormone values served results from 839 men included in two population based studies, assessing risk factors for cardiovascular disease [11], and time to pregnancy (in partner) [12]. Thus, controls were unselected men from the capital region of Denmark chosen without knowledge of fertility or body mass index. The study was approved by the Regional Ethical Committee (H-3-2013-175).
2.2.
Hormone analyses
During the 5-yr follow-up program, evaluation of serum levels of LH and TT was generally performed within the 1st yr after orchiectomy or BEP, after 1–3 yr, and by the end of the 5-yr follow-up. As not all patients had three evaluations of LH and TT, we based our calculations on two measurements: the primary measurement performed after treatment and the last measurement performed before the end of follow-up. Blood samples for analysis of reproductive hormones were generally drawn in the morning between 8 AM and 12 AM. Serum concentrations of LH were measured by time-resolved immunofluorometric assay (Delfia; Perkin Elmer, Turku, Finland) with detection limits of 0.05 IU/l. Intraassay and interassay coefficients of variation (CV) were below 5%. Testosterone was measured by radioimmunoassay (Siemens Coat-acount, Cruinn Diagnostics Limited, Dublin, Ireland) with a detection limit of <0.23 nmol/l and intra-assay and interassay CVs on 17% and 12.8%, respectively. Final testosterone analysis in the long-term cohort was performed by immunoassay (Testosterone Access 2; Beckman Coulter, Co. Clare, Ireland) with intra-assay and interassay CVs less than 5% and a detection limit <0.35 nmol/l. Evaluation of TT and LH were performed at the laboratory of the Department of Growth and Reproduction at Copenhagen University Hospital, Rigshospitalet.
2.3.
Statistical analyses
To adjust for the age-related changes in reproductive hormones, the
2.
Materials and methods
2.1.
Patients and controls
Patients were included from two distinct cohorts treated at Copenhagen University Hospital, Rigshospitalet:
percent of predicted values were calculated by dividing the observed value for each individual patient with the mean value of the agematched controls. Bivariate reference charts of TT in conjunction with the corresponding LH were constructed based on the controls and made by local linear regression smoothing as previously described [10]. Thus, the curve separates individuals with an abnormal LH/TT combination (to the right of the curve) from individuals with a normal combination of LH and TT (to the left of the curve; Fig. 3). Due to the method change in the
(1) Five-year follow-up cohort: all TC patients treated with orchiectomy
analysis of testosterone in the long-term cohort, bivariate evaluation of
alone or three cycles of standard dose BEP from March 2004 to May
LH and TT was not performed in this group. Furthermore, TT was
2011, alive, and without relapse at the time of their 5-yr follow-up visit (n = 390). Exclusion criteria were: testosterone substitution
analyzed with immunoassay in 158 controls, only. Thus, percent of predicted value of TT at the last visit in the long-term cohort was
before treatment (n = 9), less than two available analyses of LH and TT
calculated based on these 158 controls. Follow-up was calculated from
within 5-yr follow-up (n = 95), contralateral germ cell neoplasia in
orchiectomy (Stage I patients) or from initiation of BEP.
situ or bilateral tumors (n = 20), analysis of testosterone with another method than radioimmunoassay (n = 84), age <18 yr or age >65 yr (n = 13), or lost to follow-up (n = 13), leaving 156 evaluable patients; 75 in the orchiectomy alone (Stage I) cohort and 81 in the BEP cohort. (2) Long-term changes cohort: TC patients treated with orchiectomy alone
A univariate linear regression model for each treatment group was constructed with time from treatment as an independent variable and age-adjusted LH and TT as dependent variables. Each beta-value represents the age-adjusted annual change in each of the two hormones. A two-sided p value <0.05 was considered statistically significant.
or three to four cycles of standard dose BEP from 1984 to 2013 enrolled
Statistical tests were computed in IBM SPSS Statistics (Version 22.0; IBM
in the cross-sectional study ‘‘Symptoms and Clinical Signs of
Corp, Armonk, NY, USA).
Hypogonadism in Testicular Cancer Survivors’’ ClinicalTrials.gov number, NCT02240966 conducted from August 2014 to March
3.
Results
3.1.
Hormonal changes within 5-yr follow-up program
2016. Inclusion criteria were: LH 8 IU/l and TT <12 nmol/l (mild uncompensated Leydig cell dysfunction), or LH 8 IU/l and TT >12 nmol/l (mild compensated Leydig cell dysfunction) within the 5-yr follow-up program with at least one later blood sample, no evidence of relapse 1-yr after TC treatment, and age <18 yr or age >65 yr.
For the 75 Stage I patients, age-adjusted LH and TT did not change significantly within the follow-up period
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 3
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
GCC-survivors included in longterm cross seconal study (NCT02240966) (n = 158)
Excluded: Radiaon due to contralateral germ cell neoplasia in situ (n = 21) Abdominal radiotherapy (n = 23) More than one line of treatment (n = 18)
BEP (n = 47)
Stage I (n = 49)
Excluded: Normal Leydig cell funcon: during followup (n = 20) Unavailable TT and LH aer BEP: (n = 4)
Excluded: Normal Leydig cell funcon during followup (n = 29)
Stage I paents with mild Leydig cell dysfuncon (LH ≥ 8 IU/l) during follow-up (n = 20)
BEP treated paents with mild Leydig cell dysfuncon (LH ≥ 8 IU/l) during follow-up (n = 23)
Fig. 1 – Flow chart of patients with mild Leydig cell dysfunction (luteinizing hormone [LH] I 8 IU/l) included for long-term evaluation of serum levels of total testosterone and LH. BEP = bleomycin, etoposide, and cisplatin; GCC = germ-cell cancer; TT = total testosterone.
(p = 0.9 and p = 0.4, respectively; Fig. 2). Eight patients (11%) initiated testosterone substitution. For the 67 patients who did not initiate testosterone substitution during follow-up, LH remained approximately 150% of what was expected, while TT was 15% lower than expected from age-matched controls throughout the study period (Fig. 2, Table 1). In the combined evaluation of LH and TT, 37/67 (55%) and 38/67 (57%) were outside the 97.5 percentile at visit 1 and visit 2, respectively (Fig. 3). For the eight patients who initiated testosterone substitution the median time to initiation of treatment was 1.7 yr (range, 0.3–5.3 yr). The majority of these patients were outside the 97.5 percentile when using combined evaluation of LH and TT (Fig. 3). For the 81 BEP treated patients age-adjusted serum TT did not change significantly within the follow-up period (p = 0.3), while there was a significant decline in ageadjusted LH (DLH –24.2 percentage point, 95% confidence interval: –38.5 to –9.9, p = 0.001; Fig. 2). In total, 12 patients (15%) initiated testosterone substitution during follow-up. In the 69 patients who did not initiate testosterone substitution, age-adjusted TT was 83% and 86% of expected from age-matched controls throughout follow-up, while age-adjusted LH was 204% and 157% of expected at visit 1 and visit 2, respectively (Table 1). Bivariate evaluation of LH and TT showed that 46/69 (67%) and 34/69 (49%) were outside the 97.5 percentile at visit 1 and visit 2, respectively (Fig. 3).
For the 12 patients who initiated testosterone substitution, median time to testosterone substitution was 1.4 yr (range, 0.5–4 yr). All of these patients were outside the 97.5 percentile when using bivariate evaluation of LH and TT (Fig. 3). There was no significant difference between the proportion of patients who initiated testosterone substitution between Stage I patients and BEP-treated patients (p = 0.5, x2 test.) When evaluating Stage I patients and BEP-treated patients together, 2/55 (4%) of patients with combined LH/ TT within the 97.5 percentile at the first visit initiated testosterone substitution. In contrast, 18/101 (16%) of patients with combined LH/TT outside the 97.5 percentile at the first visit initiated testosterone substitution (Fig. 3). Thus, when considering testosterone substitution as outcome and combined LH/TT outside the 97.5 percentile at visit 1 as a positive test, this corresponds to a sensitivity of 90%, a specificity of 39%, a positive predictive value of 18%, and a negative predictive value of 96% (Supplementary Table 1). 3.2.
Long-term hormonal changes in patients with mild Leydig
cell dysfunction
In total, 20 Stage I patients and 23 BEP treated patients with mild Leydig cell dysfunction were included for evaluation of long-term changes in hormonal values (Fig. 1).
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 4
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
1000
β= –2.3 (–27.7–23.2); (p = 0.9)
800
Without T substuon
800
700
With T substuon
700
600 500 400
Without T substuon With T substuon
600 500 400
300
300
200
200
100
100 0
0 0
1
2
3 Follow-up (yr)
4
5
0
6
200
1
2
200
Stage 1 cohort 180
Without T substuon
β= 1.3 (–1.9–4.6); (p = 0.4)
3 Follow-up (yr)
4
5
6 Without T substuon
BEP cohort β= 1.3 (–1.1–3.7); (p = 0.3)
180
With T substuon
160
With T substuon
160
140
140 TT % of expected
TT % of expected
BEP cohort β= –24.2 (–38.5– –9.9); (p = 0.001)
900
LH % of expected
LH % of expected
1000
Stage 1 cohort
LH↑↑
900
120 100 80
120 100 80
60
60
40
40
20
20 0
0
0
1
2
3 Follow-up (yr)
4
5
6
0
1
2
3 Follow-up (yr)
4
5
6
Fig. 2 – Changes in total testosterone and luteinizing hormone (LH; % of predicted from age-matched controls) within 5-yr follow-up in 75 Stage I patients treated with orchiectomy alone and 81 patients treated with bleomycin, etoposide, and cisplatin (BEP). Each black dotted line represents the change of an individual patient who did not initiate testosterone (T) substitution, while each red line represents the change of a patient who initiated T substitution during follow-up (total [T]T and LH values before initiation of T substitution). The fat dotted-line represents the results of a linear regression analysis, with the beta-value representing the annual change with corresponding 95% confidence interval and p value.
Table 1 – Serum levels of total testosterone and luteinizing hormone (LH) during 5-yr follow-up in a cohort of testicular cancer patients treated with orchiectomy alone or bleomycin, etoposide, and cisplatin (BEP) Outcomes
Follow-up (yr) Visit 1 Visit 2 Age (yr) Visit 1 Visit 2 Histology Seminoma Nonseminoma Total testosterone (nmol/l) Visit 1 Visit 2 Total T (% of expected) Visit 1 Visit 2 LH (IU/l) Visit 1 Visit 2 LH (% of expected) Visit 1 Visit 2
Stage I without T substitution (n = 67)
0.4 (0.2–1.1) 2.2 (2.0–3.1) 35 (29–45) 36 (31–47) 40 27
Stage I with T substitutiona (n = 8)
0.2 (0.2–0.7) 1.7 (0–9–3.7) 37 (33–39) 39 (37–40) 7 1
BEP-cohort without T substitution (n = 69)
BEP-cohort with T substitutiona (n = 12)
1.0 (0.7–1.1) 4.1 (2.7–4.7)
0.8 (0.5–1.0) 1.4 (0.8–1.9)
33.5 (27.9–39.8) 35.4 (31.5–42.5) 23 45
37.8 (34.7–46.6) 39.2 (35.0–46.8) 6 6
14.4 (11.8–18.7) 14.5 (11.8–17.1)
11.4 (7.8–14.8) 9.4 (4.7–11.6)
14.6 (11.9–16.1) 14.1 (12.1–16.5)
11.4 (10.9–16.1) 8.6 (7.8–9.5)
87 (67–108) 84 (70–98)
65 (45–91) 55 (27–67)
83 (71–94) 86 (70–98)
68 (66–95) 49 (47–57)
4.9 (3.8–6.2) 5.1 (4.0–6.7) 143 (110–183) 153 (114–191)
11.5 (4.3–19.5) 12.0 (5.5 –21.4) 335 (138–577) 359 (121–615)
6.9 (5.11–10.0) 5.3 (4.0–7.4) 204 (145–293) 157 (118–216)
12.3 (8.9–20.5) 9.2 (8.0–13.7) 367 (266–586) 270 (236–391)
Continuous outcomes are presented as median with interquartile range. Visit 2 represents the last visit before initiation of testosterone substitution.
a
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 5
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
45
45
Stage I cohort without T substuon
35
35 Visit 1
30
Visit 1
30
Visit 2
TT (nmol/l)
TT (nmol/l)
BEP cohort without T substuon
40
40
25 20
Visit 2 25 20
15
15
10
10
5
5 0
0
0
5
10
15
20
25
30
0
35
5
10
15
45
45
Stage I cohort with T substuon
40
40
35
35
30
25
30
35
25
Visit 2
20
Visit 1
25
Visit 2
20
15
15
10
10
5
BEP cohort with T substuon
30
Visit 1 TT (nmol/l)
TT (nmol/l)
20
LH (IU/l)
LH (IU/l)
5
LH↑↑
0
0 0
5
10
15
20
25
30
35
0
5
10
LH (IU/l)
15
20
25
30
35
LH (IU/l)
Fig. 3 – Bivariate charts of luteinizing hormone (LH) and total testosterone (TT). The line represents the 97.5 percentile of healthy controls. Each & represents (LH,TT) for an individual patient at the first visit within the follow-up program, while each D represents (LH,TT) for an individual patient at the end of the follow-up program. BEP = bleomycin, etoposide, and cisplatin.
Table 2 – Long-term serum levels of total testosterone and luteinizing hormone (LH) in a selected group of testicular cancer patients with mild Leydig cell dysfunction (LH I 8 IU/l) during follow-up Outcomes Follow-up (yr) Visit 1 Visit 2 Age (yr) Visit 1 Visit 2 Histology Seminoma Nonseminoma Total testosterone (nmol/l) Visit 1 Visit 2 Total T (% of expected) Visit 1 Visit 2 LH (IU/l) Visit 1 Visit 2 LH (% of expected) Visit 1 Visit 2
Stage I (n = 20)
BEP (n = 23)
0.7 (0.2–2.0) 6.7 (3.0–10.4)
0.8 (0.4–3.3) 9.1 (3.6–20.6)
34.9 (30.6–39.6) 42.0 (33.7–49.1)
35.9 (30.7–38.7) 48.9 (39.7–52.3)
16 4
3 20
14.2 (11.2–18.0) 12.0 (9.3–14.5)
14.3 (12.0–17.9) 11.8 (9.4–14.7)
85 (63–96) 80 (65–102)
84 (69–110) 77 (64–97)
Median follow-up was 7 yr and 9 yr for patients treated with orchiectomy alone or BEP, respectively (Table 2). After adjustment for age, there was a significant decline in TT among BEP-treated patients DTT (–0.9 percentage point, 95% confidence interval: –1.8 to –0.04, p = 0.04). Apart from that, there were no significant long-term changes in Leydig cell function (Fig. 4). Among Stage I patients, 10/20 patients experienced an age-adjusted decline in TT and three patients had TT values <50% of expected at long-term follow-up. Among BEPtreated patients 15/23 experienced an age-adjusted decline in TT and four had TT values <50% of expected at long-term follow-up (Fig. 4) 4.
9.0 (8.2–10.6) 9.0 (7.0–12.9) 269 (243–312) 254 (206–384)
10.3 (9.0–15.4) 7.1 (5.5–11.1) 306 (266–454) 212 (155–316)
BEP = bleomycin, etoposide, and cisplatin. Continuous outcomes are presented as median with interquartile range.
Discussion
This study has several important findings: (1) TT remained stable in the majority of Stage I and BEP treated TC patients, while a significant decline in age-adjusted LH was observed in BEP-treated patients within 5-yr follow-up, (2) in total, 11% of Stage I TC patients and 15% of BEP-treated patients initiated testosterone substitution during 5-yr follow-up, (3) there was a significant long-term decline in age-adjusted TT in BEP-treated patients with mild Leydig
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 6
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
800
800
Stage I (mild Leydig celldysfuncon) β= –0.6 (–9.1– 7.9); (p = 0.9)
700 600
β= –1.4 (–6.2–3.5); (p = 0.6)
600 LH % of expected
LH % of expected
BEP (mild Leydig cell dysfuncon)
700
500 400 300
500 400 300
200
200
100
100
0
0 0
5
10
15
20
0
5
10
Follow-up (yr)
200
β= –0.9 (–2.8–1.0); (p = 0.4)
20
25
30
25
30
BEP (mild Leydig cell dysfuncon) β= –0.9 (–1.8– –0.04); (p = 0.04)
180 Total testosterone % of expected
Total testosterone % of expected
200
Stage I (mild Leydig celldysfuncon)
180
15 Follow-up (yr)
160 140 120 100
80 60 40
160 140 120 100 80 60 40 20
20
0
0 0
5
10
15
20
Follow-up (yr)
0
5
10
15
20
Follow-up (yr)
Fig. 4 – Long-term changes in total testosterone and luteinizing hormone (LH) (% of predicted from age-matched controls) in 20 Stage I patients treated with orchiectomy alone and 23 patients treated with bleomycin, etoposide, and cisplatin (BEP). All patients had mild Leydig cell dysfunction (LH I 8 IU/l) during follow-up. Each black dotted line represents the change of an individual patient. The fat dotted-line represents the results of a linear regression analysis, with the beta-value representing the annual change with corresponding 95% confidence interval and p value.
cell dysfunction (LH 8 IU/l) during the standard 5-yr follow-up period. It has been hypothesized that Leydig cell function recovers over time after the initial injury caused by TC treatment [2,3]. Indeed, the significant decline in LH in BEPtreated patients within 5-yr follow-up suggests a degree of Leydig cell recovery in this group of patients. However, we found that median LH was1.5 times higher than expected from age-matched controls at the end of follow-up and 15% of patients experienced a decline in TT requiring testosterone substitution. Furthermore, the long-term results indicate no signs of Leydig cell recovery as LH remained elevated, while TT remained lower than expected from agematched controls. Our finding of increased LH are in line with previous publications [1,5,7,9,13] and shows that an increased LHdrive is continuously needed to compensate for the removal of the tumor-bearing testicle. Interestingly, the change in TT and the proportion of patients who initiated testosterone substitution were similar between BEP-treated patients and Stage I patients within follow-up. This indicates that orchiectomy seems to be the most significant factor in the development of hypogonadism, while BEP appears to be
less important. This is in contrast to a recent meta-analysis showing that BEP increases the risk of testosterone deficiency when compared with orchiectomy alone [4]. We defined mild Leydig cell dysfunction as having serum levels of LH 8 IU/l, which corresponds to approximately two standard deviations above healthy controls. The significant decline in age-adjusted TT in BEP-treated patients with mild Leydig cell dysfunction indicates that this subgroup of patients is at risk of developing long-term primary testicular failure. On the contrary, patients with a combined LH/TT within the 97.5 percentile at the first visit within 5-yr follow-up had a negligible risk (4%) of testosterone substitution during follow-up. The present study suggests that the majority of TC patients develop primary testicular failure within few years after treatment. Our practice has been to evaluate TT levels and LH within 1 yr after orchiectomy or BEP, again at 1–3 yr, and by the end of the 5-yr follow up. Testosterone substitution has been initiated if TT fell to subnormal levels with concomitantly elevated LH combined with clinical symptoms of testosterone deficiency. Thus, it should be emphasized that testosterone substitution should not be initiated based on low levels of testosterone and
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
elevated LH alone. However, as the sensitivity and specificity of clinical symptoms are low, it is important to measure LH and TT. Clinical guidelines suggest that a diagnosis of androgen deficiency should be based on repeated measures of low levels of testosterone in combination with clinical symptoms [14]. In a long-term study of TC survivors, Nord et al [5] found that 13 of 373 (3%) patients treated with standard dose cisplatin-based chemotherapy and seven of 251 (3%) patients followed on surveillance alone received testosterone substitution after 12-yr follow-up. The proportion of TC patients on testosterone substitution is considerably smaller than in our study. As there are no uniform international guidelines concerning testosterone substitution, different clinical practice or follow-up programs might explain the conflicting results. It has uniformly been reported that low levels of TT is associated with the metabolic syndrome in testicular cancer survivors [8,15–17]. However, little is known about the clinical consequences of slightly subnormal TT levels and elevated LH. A recent population-based study including 5350 Danish men found that increased LH and LH/TT ratio were associated with increased all-cause mortality, while there was no association between level of TT alone and allcause mortality [18]. It is not known whether these findings can be extrapolated to TC survivors, where the majority have elevated LH levels. A limitation of the present study is the retrospective study design and the small number of patients with mild Leydig cell dysfunction included. Furthermore, a relatively large proportion of patients were excluded from the cohort within 5-yr follow-up as TT and LH were evaluated in another laboratory which did not use radioimmunoassay. Finally, patients treated with abdominal RT or more treatment lines were not included and extrapolation to these groups cannot be made with certainty.
7
Author contributions: Mikkel Bandak had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Bandak, Juul, Daugaard, Jørgensen, Lauritsen, Kier. Acquisition of data: Bandak, Mortensen, Kier. Analysis and interpretation of data: Bandak, Juul, Daugaard, Jørgensen, Lauritsen, Kier, Mortensen. Drafting of the manuscript: Bandak. Critical revision of the manuscript for important intellectual content: Juul, Daugaard, Jørgensen, Lauritsen, Kier, Mortensen. Statistical analysis: Bandak, Lauritsen. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: None. Other: None. Financial disclosures: Mikkel Bandak certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: None. Acknowledgments: We thank the following source of funding: grants from Copenhagen University Hospital, Rigshospitalet, The Danish Cancer Society, and the Preben and Anna Simonsen Foundation.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.euf.2016.11. 018.
References [1] Sprauten M, Brydøy M, Haugnes HS, et al. Longitudinal serum
5.
Conclusions
testosterone, luteinizing hormone, and follicle-stimulating hormone levels in a population-based sample of long-term testicular
In summary, TT remained stable in the majority of patients during 5-yr follow-up, while there was a significant decline in LH in patients treated with BEP, suggesting a degree of Leydig cell recovery. We recommend to evaluate levels of TT and LH s at regular intervals during the 1st yr after TC treatment in order to identify the 11–15% of patients who according to our findings develop testicular failure requiring testosterone substitution. We suggest that patients are informed about symptoms of testosterone deficiency at the end of follow-up. Our results indicate that evaluation of Leydig cell function should be considered beyond clinical follow-up in BEPtreated patients with mild Leydig cell dysfunction (LH > 8 IU/l) during follow-up. Patients with combined LH/TT within the 97.5 percentile of healthy controls within follow-up had only small risk (4%) of developing primary testicular failure, and further evaluation of Leydig cell function might be omitted in this group of patients. However, this finding should be validated in a prospective study of a larger group of TC patients.
cancer survivors. J Clin Oncol 2014;32:571–8. [2] Aass N, Fossa˚ SD, Theodorsen L, Norman N. Prediction of long-term gonadal toxicity after standard treatment for testicular cancer. Eur J Cancer 1991;27:1087–91. [3] Eberhard J, Sta˚hl O, Cwikiel M, et al. Risk factors for post-treatment hypogonadism in testicular cancer patients. Eur J Endocrinol 2008;158:561–70. [4] Bandak M, Jørgensen N, Juul A, et al. Testosterone deficiency in testicular cancer survivors - a systematic review and meta-analysis. Andrology 2016. http://dx.doi.org/10.1111/andr.12177. [5] Nord C, Bjøro T, Ellingsen D, et al. Gonadal hormones in long-term survivors 10 years after treatment for unilateral testicular cancer. Eur Urol 2003;44:322–8. [6] Huddart RA, Norman A, Moynihan C, et al. Fertility, gonadal and sexual function in survivors of testicular cancer. Br J Cancer 2005;93:200–7. [7] Wiechno P, Demkow T, Kubiak K, Sadowska M, Kamin´ska J. The quality of life and hormonal disturbances in testicular cancer survivors in cisplatin era. Eur Urol 2007;52:1448–54. [8] Nuver J, Smit AJ, Wolffenbuttel BHR, et al. The metabolic syndrome and disturbances in hormone levels in long-term survivors of disseminated testicular cancer. J Clin Oncol 2005;23:3718–25.
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 8
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
[9] Gerl A, Mu¨hlbayer D, Hansmann G, Mraz W, Hiddemann W. The
[14] Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy
impact of chemotherapy on Leydig cell function in long term
in men with androgen deficiency syndromes: an Endocrine Society
survivors of germ cell tumors. Cancer 2001;91:1297–303.
clinical practice guideline. J Clin Endocrinol Metab 2010;95:
[10] Aksglaede L, Andersson AM, Jørgensen N, et al. Primary testicular
2536–59.
failure in Klinefelter’s syndrome: The use of bivariate luteinizing
[15] de Haas EC, Altena R, Boezen HM, et al. Early development of the
hormone-testosterone reference charts. Clin Endocrinol (Oxf)
metabolic syndrome after chemotherapy for testicular cancer. Ann
2007;66:276–81.
Oncol 2013;24:749–55.
[11] Rasmussen SL, Torp-Pedersen C, Borch-Johnsen K, Ibsen H. Normal
[16] Haugnes HS, Aass N, Fossa˚ SD, et al. Components of the metabolic
values for ambulatory blood pressure and differences between
syndrome in long-term survivors of testicular cancer. Ann Oncol
casual blood pressure and ambulatory blood pressure: results from a Danish population survey. J Hypertens 1998;16:1415–24.
2007;18:241–8. [17] Willemse PM, Burggraaf J, Hamdy NAT, et al. Prevalence of the
[12] Jensen TK, Andersson AM, Hjollund NH, et al. Inhibin B as a serum
metabolic syndrome and cardiovascular disease risk in chemother-
marker of spermatogenesis: correlation to differences in sperm
apy-treated testicular germ cell tumour survivors. Br J Cancer
concentration and follicle-stimulating hormone levels. A study of 349 Danish men. J Clin Endocrinol Metab 1997;82:4059–63.
2013;109:60–7. [18] Agergaard Holmboe S, Vradi E, Kold Jensen T, et al. The association
[13] Palmieri G, Lotrecchiano G, Ricci G, et al. Gonadal function after
of reproductive hormone levels and all-cause, cancer and cardio-
multimodality treatment in men with testicular germ cell cancer.
vascular disease mortality in men. J Clin Endocrinol Metab 2015;
Eur J Endocrinol 1996;134:431–6.
100:4472–80.
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
available at www.sciencedirect.com journal homepage: www.europeanurology.com/eufocus
Testis Cancer
Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin Mikkel Bandak a,*, Niels Jørgensen b,c, Anders Juul b,c, Jakob Lauritsen a, Maria Gry Gundgaard Kier a,d, Mette Saksø Mortensen a, Gedske Daugaard a a
Department of Oncology, Copenhagen University Hospital, Rigshospitalet, Denmark;
b
Department of Growth and Reproduction, Copenhagen University
Hospital, Rigshospitalet, Denmark; c International Center for Research and Research Training in Endocrine Disruption of Male Reproduction and Child Health (EDMaRC), Rigshospitalet, University of Copenhagen, Denmark;
d
Unit of Survivorship, Danish Cancer Society Research Center, Danish Cancer Society,
Copenhagen, Denmark
Article info
Abstract
Article history: Accepted November 23, 2016
Background: Controversy exists whether Leydig cells recover after testicular cancer (TC) treatment or whether premature hormonal aging will occur. Objective: Evaluate serial changes in total testosterone (TT) and luteinizing hormone (LH) in patients treated with orchiectomy alone (Stage I) or combination chemotherapy with bleomycin, etoposide, and cisplatin (BEP). Design, settings, and participants: Changes in TT and LH were investigated during 5-yr follow-up (Stage I: n = 75, BEP: n = 81). A selected group of TC patients with mild Leydig cell dysfunction (LH 8 IU/l) were followed for a longer period (Stage I: n = 20, BEP: n = 23). An age-matched control group of 839 healthy men served as controls for TT and LH levels. Outcome measurements and statistical analysis: Changes in age-adjusted TT and LH were evaluated separately in each treatment group with univariate linear regression analysis. The proportion of patients initiating testosterone substitution during follow-up was calculated. Results and limitations: In the 75 Stage I patients there were no significant changes in LH and TT, while in the 81 BEP treated patients there was a significant decline in LH during follow-up (–24.2 percentage point/yr, 95% confidence interval: –38.5 to –9.9, p = 0.001). In total, 11% of Stage I patients and 15% of BEP-treated patients initiated testosterone substitution. In the 23 BEP-treated patients with mild Leydig cell dysfunction there was a significant decline in age-adjusted TT (–0.9 percentage point/yr, 95% confidence interval: –1.8 to –0.04, p = 0.04), while in the 20 Stage I patients there were no significant changes in age-adjusted LH and TT. Limitations include the retrospective study design. Conclusions: TT remained stable during 5-yr follow-up in TC patients treated with orchiectomy alone or BEP. BEP-treated patients with mild Leydig cell dysfunction during follow-up were at risk of long-term testicular failure and evaluation of Leydig cell function beyond follow-up should be considered in this group of patients. Patient summary: This study shows that the majority of testicular cancer survivors had stable testosterone levels after treatment for testicular cancer. However, 11–15% of patients needed testosterone substitution after treatment. # 2016 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Associate Editor: James Catto Keywords: Testicular cancer Testosterone Luteinizing hormone Leydig cell function
* Corresponding author. Department of Oncology, Copenhagen University Hospital, Rigshospitalet, 2100 Copenhagen, Denmark. Tel. +45 3545 6354; Fax: +45 3545 3909. E-mail address: [email protected] (M. Bandak). http://dx.doi.org/10.1016/j.euf.2016.11.018 2405-4569/# 2016 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 2
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
1.
Introduction
Testosterone substitution was the exclusion criteria. No patients in this cohort were included in the 5-yr follow-up cohort. A flow-chart of
It has been hypothesized that cisplatin-based chemotherapy, abdominal radiotherapy, and orchiectomy alone induce premature hormonal aging in testicular cancer (TC) survivors. Thus, it has been proposed to continue clinical followup beyond 5 yr [1]. However, it has also been suggested, that Leydig cells recover in the years following TC treatment [2,3]. In a recent meta-analysis, we found that cisplatinbased chemotherapy and abdominal radiotherapy increased the risk of low levels of total testosterone (TT) or use of testosterone substitution, when compared with orchiectomy alone [4]. However, most studies of Leydig cell function in TC survivors are based on a single analysis of serum levels of luteinizing hormone (LH) and TT after completion of treatment [5–9]. To clarify whether TC treatment induces premature hormonal aging and/or hormonal recovery takes place over time, serial measurements of TT and LH are needed. To answer these questions, we aimed to: (1) evaluate changes in age-adjusted TT and LH, (2) evaluate the proportion of TC patients who initiated testosterone substitution within a standard 5-yr follow-up program in an unselected cohort of TC patients treated with orchiectomy alone or BEP, (3) evaluate long-term changes in age-adjusted TT and LH in a selected group of TC patients who presented with mild Leydig cell dysfunction (elevated serum levels of LH in combination with low or normal TT levels) during the initial 5-yr follow-up period. As it has been suggested that Leydig cell function is better evaluated by combined assessment of LH and TT [10], we also estimated the proportion of patients with an abnormal LH/TT during follow-up.
included patients in this part of the study can be seen in Figure 1. As controls for the reproductive hormone values served results from 839 men included in two population based studies, assessing risk factors for cardiovascular disease [11], and time to pregnancy (in partner) [12]. Thus, controls were unselected men from the capital region of Denmark chosen without knowledge of fertility or body mass index. The study was approved by the Regional Ethical Committee (H-3-2013-175).
2.2.
Hormone analyses
During the 5-yr follow-up program, evaluation of serum levels of LH and TT was generally performed within the 1st yr after orchiectomy or BEP, after 1–3 yr, and by the end of the 5-yr follow-up. As not all patients had three evaluations of LH and TT, we based our calculations on two measurements: the primary measurement performed after treatment and the last measurement performed before the end of follow-up. Blood samples for analysis of reproductive hormones were generally drawn in the morning between 8 AM and 12 AM. Serum concentrations of LH were measured by time-resolved immunofluorometric assay (Delfia; Perkin Elmer, Turku, Finland) with detection limits of 0.05 IU/l. Intraassay and interassay coefficients of variation (CV) were below 5%. Testosterone was measured by radioimmunoassay (Siemens Coat-acount, Cruinn Diagnostics Limited, Dublin, Ireland) with a detection limit of <0.23 nmol/l and intra-assay and interassay CVs on 17% and 12.8%, respectively. Final testosterone analysis in the long-term cohort was performed by immunoassay (Testosterone Access 2; Beckman Coulter, Co. Clare, Ireland) with intra-assay and interassay CVs less than 5% and a detection limit <0.35 nmol/l. Evaluation of TT and LH were performed at the laboratory of the Department of Growth and Reproduction at Copenhagen University Hospital, Rigshospitalet.
2.3.
Statistical analyses
To adjust for the age-related changes in reproductive hormones, the
2.
Materials and methods
2.1.
Patients and controls
Patients were included from two distinct cohorts treated at Copenhagen University Hospital, Rigshospitalet:
percent of predicted values were calculated by dividing the observed value for each individual patient with the mean value of the agematched controls. Bivariate reference charts of TT in conjunction with the corresponding LH were constructed based on the controls and made by local linear regression smoothing as previously described [10]. Thus, the curve separates individuals with an abnormal LH/TT combination (to the right of the curve) from individuals with a normal combination of LH and TT (to the left of the curve; Fig. 3). Due to the method change in the
(1) Five-year follow-up cohort: all TC patients treated with orchiectomy
analysis of testosterone in the long-term cohort, bivariate evaluation of
alone or three cycles of standard dose BEP from March 2004 to May
LH and TT was not performed in this group. Furthermore, TT was
2011, alive, and without relapse at the time of their 5-yr follow-up visit (n = 390). Exclusion criteria were: testosterone substitution
analyzed with immunoassay in 158 controls, only. Thus, percent of predicted value of TT at the last visit in the long-term cohort was
before treatment (n = 9), less than two available analyses of LH and TT
calculated based on these 158 controls. Follow-up was calculated from
within 5-yr follow-up (n = 95), contralateral germ cell neoplasia in
orchiectomy (Stage I patients) or from initiation of BEP.
situ or bilateral tumors (n = 20), analysis of testosterone with another method than radioimmunoassay (n = 84), age <18 yr or age >65 yr (n = 13), or lost to follow-up (n = 13), leaving 156 evaluable patients; 75 in the orchiectomy alone (Stage I) cohort and 81 in the BEP cohort. (2) Long-term changes cohort: TC patients treated with orchiectomy alone
A univariate linear regression model for each treatment group was constructed with time from treatment as an independent variable and age-adjusted LH and TT as dependent variables. Each beta-value represents the age-adjusted annual change in each of the two hormones. A two-sided p value <0.05 was considered statistically significant.
or three to four cycles of standard dose BEP from 1984 to 2013 enrolled
Statistical tests were computed in IBM SPSS Statistics (Version 22.0; IBM
in the cross-sectional study ‘‘Symptoms and Clinical Signs of
Corp, Armonk, NY, USA).
Hypogonadism in Testicular Cancer Survivors’’ ClinicalTrials.gov number, NCT02240966 conducted from August 2014 to March
3.
Results
3.1.
Hormonal changes within 5-yr follow-up program
2016. Inclusion criteria were: LH 8 IU/l and TT <12 nmol/l (mild uncompensated Leydig cell dysfunction), or LH 8 IU/l and TT >12 nmol/l (mild compensated Leydig cell dysfunction) within the 5-yr follow-up program with at least one later blood sample, no evidence of relapse 1-yr after TC treatment, and age <18 yr or age >65 yr.
For the 75 Stage I patients, age-adjusted LH and TT did not change significantly within the follow-up period
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 3
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
GCC-survivors included in longterm cross seconal study (NCT02240966) (n = 158)
Excluded: Radiaon due to contralateral germ cell neoplasia in situ (n = 21) Abdominal radiotherapy (n = 23) More than one line of treatment (n = 18)
BEP (n = 47)
Stage I (n = 49)
Excluded: Normal Leydig cell funcon: during followup (n = 20) Unavailable TT and LH aer BEP: (n = 4)
Excluded: Normal Leydig cell funcon during followup (n = 29)
Stage I paents with mild Leydig cell dysfuncon (LH ≥ 8 IU/l) during follow-up (n = 20)
BEP treated paents with mild Leydig cell dysfuncon (LH ≥ 8 IU/l) during follow-up (n = 23)
Fig. 1 – Flow chart of patients with mild Leydig cell dysfunction (luteinizing hormone [LH] I 8 IU/l) included for long-term evaluation of serum levels of total testosterone and LH. BEP = bleomycin, etoposide, and cisplatin; GCC = germ-cell cancer; TT = total testosterone.
(p = 0.9 and p = 0.4, respectively; Fig. 2). Eight patients (11%) initiated testosterone substitution. For the 67 patients who did not initiate testosterone substitution during follow-up, LH remained approximately 150% of what was expected, while TT was 15% lower than expected from age-matched controls throughout the study period (Fig. 2, Table 1). In the combined evaluation of LH and TT, 37/67 (55%) and 38/67 (57%) were outside the 97.5 percentile at visit 1 and visit 2, respectively (Fig. 3). For the eight patients who initiated testosterone substitution the median time to initiation of treatment was 1.7 yr (range, 0.3–5.3 yr). The majority of these patients were outside the 97.5 percentile when using combined evaluation of LH and TT (Fig. 3). For the 81 BEP treated patients age-adjusted serum TT did not change significantly within the follow-up period (p = 0.3), while there was a significant decline in ageadjusted LH (DLH –24.2 percentage point, 95% confidence interval: –38.5 to –9.9, p = 0.001; Fig. 2). In total, 12 patients (15%) initiated testosterone substitution during follow-up. In the 69 patients who did not initiate testosterone substitution, age-adjusted TT was 83% and 86% of expected from age-matched controls throughout follow-up, while age-adjusted LH was 204% and 157% of expected at visit 1 and visit 2, respectively (Table 1). Bivariate evaluation of LH and TT showed that 46/69 (67%) and 34/69 (49%) were outside the 97.5 percentile at visit 1 and visit 2, respectively (Fig. 3).
For the 12 patients who initiated testosterone substitution, median time to testosterone substitution was 1.4 yr (range, 0.5–4 yr). All of these patients were outside the 97.5 percentile when using bivariate evaluation of LH and TT (Fig. 3). There was no significant difference between the proportion of patients who initiated testosterone substitution between Stage I patients and BEP-treated patients (p = 0.5, x2 test.) When evaluating Stage I patients and BEP-treated patients together, 2/55 (4%) of patients with combined LH/ TT within the 97.5 percentile at the first visit initiated testosterone substitution. In contrast, 18/101 (16%) of patients with combined LH/TT outside the 97.5 percentile at the first visit initiated testosterone substitution (Fig. 3). Thus, when considering testosterone substitution as outcome and combined LH/TT outside the 97.5 percentile at visit 1 as a positive test, this corresponds to a sensitivity of 90%, a specificity of 39%, a positive predictive value of 18%, and a negative predictive value of 96% (Supplementary Table 1). 3.2.
Long-term hormonal changes in patients with mild Leydig
cell dysfunction
In total, 20 Stage I patients and 23 BEP treated patients with mild Leydig cell dysfunction were included for evaluation of long-term changes in hormonal values (Fig. 1).
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 4
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
1000
β= –2.3 (–27.7–23.2); (p = 0.9)
800
Without T substuon
800
700
With T substuon
700
600 500 400
Without T substuon With T substuon
600 500 400
300
300
200
200
100
100 0
0 0
1
2
3 Follow-up (yr)
4
5
0
6
200
1
2
200
Stage 1 cohort 180
Without T substuon
β= 1.3 (–1.9–4.6); (p = 0.4)
3 Follow-up (yr)
4
5
6 Without T substuon
BEP cohort β= 1.3 (–1.1–3.7); (p = 0.3)
180
With T substuon
160
With T substuon
160
140
140 TT % of expected
TT % of expected
BEP cohort β= –24.2 (–38.5– –9.9); (p = 0.001)
900
LH % of expected
LH % of expected
1000
Stage 1 cohort
LH↑↑
900
120 100 80
120 100 80
60
60
40
40
20
20 0
0
0
1
2
3 Follow-up (yr)
4
5
6
0
1
2
3 Follow-up (yr)
4
5
6
Fig. 2 – Changes in total testosterone and luteinizing hormone (LH; % of predicted from age-matched controls) within 5-yr follow-up in 75 Stage I patients treated with orchiectomy alone and 81 patients treated with bleomycin, etoposide, and cisplatin (BEP). Each black dotted line represents the change of an individual patient who did not initiate testosterone (T) substitution, while each red line represents the change of a patient who initiated T substitution during follow-up (total [T]T and LH values before initiation of T substitution). The fat dotted-line represents the results of a linear regression analysis, with the beta-value representing the annual change with corresponding 95% confidence interval and p value.
Table 1 – Serum levels of total testosterone and luteinizing hormone (LH) during 5-yr follow-up in a cohort of testicular cancer patients treated with orchiectomy alone or bleomycin, etoposide, and cisplatin (BEP) Outcomes
Follow-up (yr) Visit 1 Visit 2 Age (yr) Visit 1 Visit 2 Histology Seminoma Nonseminoma Total testosterone (nmol/l) Visit 1 Visit 2 Total T (% of expected) Visit 1 Visit 2 LH (IU/l) Visit 1 Visit 2 LH (% of expected) Visit 1 Visit 2
Stage I without T substitution (n = 67)
0.4 (0.2–1.1) 2.2 (2.0–3.1) 35 (29–45) 36 (31–47) 40 27
Stage I with T substitutiona (n = 8)
0.2 (0.2–0.7) 1.7 (0–9–3.7) 37 (33–39) 39 (37–40) 7 1
BEP-cohort without T substitution (n = 69)
BEP-cohort with T substitutiona (n = 12)
1.0 (0.7–1.1) 4.1 (2.7–4.7)
0.8 (0.5–1.0) 1.4 (0.8–1.9)
33.5 (27.9–39.8) 35.4 (31.5–42.5) 23 45
37.8 (34.7–46.6) 39.2 (35.0–46.8) 6 6
14.4 (11.8–18.7) 14.5 (11.8–17.1)
11.4 (7.8–14.8) 9.4 (4.7–11.6)
14.6 (11.9–16.1) 14.1 (12.1–16.5)
11.4 (10.9–16.1) 8.6 (7.8–9.5)
87 (67–108) 84 (70–98)
65 (45–91) 55 (27–67)
83 (71–94) 86 (70–98)
68 (66–95) 49 (47–57)
4.9 (3.8–6.2) 5.1 (4.0–6.7) 143 (110–183) 153 (114–191)
11.5 (4.3–19.5) 12.0 (5.5 –21.4) 335 (138–577) 359 (121–615)
6.9 (5.11–10.0) 5.3 (4.0–7.4) 204 (145–293) 157 (118–216)
12.3 (8.9–20.5) 9.2 (8.0–13.7) 367 (266–586) 270 (236–391)
Continuous outcomes are presented as median with interquartile range. Visit 2 represents the last visit before initiation of testosterone substitution.
a
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 5
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
45
45
Stage I cohort without T substuon
35
35 Visit 1
30
Visit 1
30
Visit 2
TT (nmol/l)
TT (nmol/l)
BEP cohort without T substuon
40
40
25 20
Visit 2 25 20
15
15
10
10
5
5 0
0
0
5
10
15
20
25
30
0
35
5
10
15
45
45
Stage I cohort with T substuon
40
40
35
35
30
25
30
35
25
Visit 2
20
Visit 1
25
Visit 2
20
15
15
10
10
5
BEP cohort with T substuon
30
Visit 1 TT (nmol/l)
TT (nmol/l)
20
LH (IU/l)
LH (IU/l)
5
LH↑↑
0
0 0
5
10
15
20
25
30
35
0
5
10
LH (IU/l)
15
20
25
30
35
LH (IU/l)
Fig. 3 – Bivariate charts of luteinizing hormone (LH) and total testosterone (TT). The line represents the 97.5 percentile of healthy controls. Each & represents (LH,TT) for an individual patient at the first visit within the follow-up program, while each D represents (LH,TT) for an individual patient at the end of the follow-up program. BEP = bleomycin, etoposide, and cisplatin.
Table 2 – Long-term serum levels of total testosterone and luteinizing hormone (LH) in a selected group of testicular cancer patients with mild Leydig cell dysfunction (LH I 8 IU/l) during follow-up Outcomes Follow-up (yr) Visit 1 Visit 2 Age (yr) Visit 1 Visit 2 Histology Seminoma Nonseminoma Total testosterone (nmol/l) Visit 1 Visit 2 Total T (% of expected) Visit 1 Visit 2 LH (IU/l) Visit 1 Visit 2 LH (% of expected) Visit 1 Visit 2
Stage I (n = 20)
BEP (n = 23)
0.7 (0.2–2.0) 6.7 (3.0–10.4)
0.8 (0.4–3.3) 9.1 (3.6–20.6)
34.9 (30.6–39.6) 42.0 (33.7–49.1)
35.9 (30.7–38.7) 48.9 (39.7–52.3)
16 4
3 20
14.2 (11.2–18.0) 12.0 (9.3–14.5)
14.3 (12.0–17.9) 11.8 (9.4–14.7)
85 (63–96) 80 (65–102)
84 (69–110) 77 (64–97)
Median follow-up was 7 yr and 9 yr for patients treated with orchiectomy alone or BEP, respectively (Table 2). After adjustment for age, there was a significant decline in TT among BEP-treated patients DTT (–0.9 percentage point, 95% confidence interval: –1.8 to –0.04, p = 0.04). Apart from that, there were no significant long-term changes in Leydig cell function (Fig. 4). Among Stage I patients, 10/20 patients experienced an age-adjusted decline in TT and three patients had TT values <50% of expected at long-term follow-up. Among BEPtreated patients 15/23 experienced an age-adjusted decline in TT and four had TT values <50% of expected at long-term follow-up (Fig. 4) 4.
9.0 (8.2–10.6) 9.0 (7.0–12.9) 269 (243–312) 254 (206–384)
10.3 (9.0–15.4) 7.1 (5.5–11.1) 306 (266–454) 212 (155–316)
BEP = bleomycin, etoposide, and cisplatin. Continuous outcomes are presented as median with interquartile range.
Discussion
This study has several important findings: (1) TT remained stable in the majority of Stage I and BEP treated TC patients, while a significant decline in age-adjusted LH was observed in BEP-treated patients within 5-yr follow-up, (2) in total, 11% of Stage I TC patients and 15% of BEP-treated patients initiated testosterone substitution during 5-yr follow-up, (3) there was a significant long-term decline in age-adjusted TT in BEP-treated patients with mild Leydig
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 6
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
800
800
Stage I (mild Leydig celldysfuncon) β= –0.6 (–9.1– 7.9); (p = 0.9)
700 600
β= –1.4 (–6.2–3.5); (p = 0.6)
600 LH % of expected
LH % of expected
BEP (mild Leydig cell dysfuncon)
700
500 400 300
500 400 300
200
200
100
100
0
0 0
5
10
15
20
0
5
10
Follow-up (yr)
200
β= –0.9 (–2.8–1.0); (p = 0.4)
20
25
30
25
30
BEP (mild Leydig cell dysfuncon) β= –0.9 (–1.8– –0.04); (p = 0.04)
180 Total testosterone % of expected
Total testosterone % of expected
200
Stage I (mild Leydig celldysfuncon)
180
15 Follow-up (yr)
160 140 120 100
80 60 40
160 140 120 100 80 60 40 20
20
0
0 0
5
10
15
20
Follow-up (yr)
0
5
10
15
20
Follow-up (yr)
Fig. 4 – Long-term changes in total testosterone and luteinizing hormone (LH) (% of predicted from age-matched controls) in 20 Stage I patients treated with orchiectomy alone and 23 patients treated with bleomycin, etoposide, and cisplatin (BEP). All patients had mild Leydig cell dysfunction (LH I 8 IU/l) during follow-up. Each black dotted line represents the change of an individual patient. The fat dotted-line represents the results of a linear regression analysis, with the beta-value representing the annual change with corresponding 95% confidence interval and p value.
cell dysfunction (LH 8 IU/l) during the standard 5-yr follow-up period. It has been hypothesized that Leydig cell function recovers over time after the initial injury caused by TC treatment [2,3]. Indeed, the significant decline in LH in BEPtreated patients within 5-yr follow-up suggests a degree of Leydig cell recovery in this group of patients. However, we found that median LH was1.5 times higher than expected from age-matched controls at the end of follow-up and 15% of patients experienced a decline in TT requiring testosterone substitution. Furthermore, the long-term results indicate no signs of Leydig cell recovery as LH remained elevated, while TT remained lower than expected from agematched controls. Our finding of increased LH are in line with previous publications [1,5,7,9,13] and shows that an increased LHdrive is continuously needed to compensate for the removal of the tumor-bearing testicle. Interestingly, the change in TT and the proportion of patients who initiated testosterone substitution were similar between BEP-treated patients and Stage I patients within follow-up. This indicates that orchiectomy seems to be the most significant factor in the development of hypogonadism, while BEP appears to be
less important. This is in contrast to a recent meta-analysis showing that BEP increases the risk of testosterone deficiency when compared with orchiectomy alone [4]. We defined mild Leydig cell dysfunction as having serum levels of LH 8 IU/l, which corresponds to approximately two standard deviations above healthy controls. The significant decline in age-adjusted TT in BEP-treated patients with mild Leydig cell dysfunction indicates that this subgroup of patients is at risk of developing long-term primary testicular failure. On the contrary, patients with a combined LH/TT within the 97.5 percentile at the first visit within 5-yr follow-up had a negligible risk (4%) of testosterone substitution during follow-up. The present study suggests that the majority of TC patients develop primary testicular failure within few years after treatment. Our practice has been to evaluate TT levels and LH within 1 yr after orchiectomy or BEP, again at 1–3 yr, and by the end of the 5-yr follow up. Testosterone substitution has been initiated if TT fell to subnormal levels with concomitantly elevated LH combined with clinical symptoms of testosterone deficiency. Thus, it should be emphasized that testosterone substitution should not be initiated based on low levels of testosterone and
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
elevated LH alone. However, as the sensitivity and specificity of clinical symptoms are low, it is important to measure LH and TT. Clinical guidelines suggest that a diagnosis of androgen deficiency should be based on repeated measures of low levels of testosterone in combination with clinical symptoms [14]. In a long-term study of TC survivors, Nord et al [5] found that 13 of 373 (3%) patients treated with standard dose cisplatin-based chemotherapy and seven of 251 (3%) patients followed on surveillance alone received testosterone substitution after 12-yr follow-up. The proportion of TC patients on testosterone substitution is considerably smaller than in our study. As there are no uniform international guidelines concerning testosterone substitution, different clinical practice or follow-up programs might explain the conflicting results. It has uniformly been reported that low levels of TT is associated with the metabolic syndrome in testicular cancer survivors [8,15–17]. However, little is known about the clinical consequences of slightly subnormal TT levels and elevated LH. A recent population-based study including 5350 Danish men found that increased LH and LH/TT ratio were associated with increased all-cause mortality, while there was no association between level of TT alone and allcause mortality [18]. It is not known whether these findings can be extrapolated to TC survivors, where the majority have elevated LH levels. A limitation of the present study is the retrospective study design and the small number of patients with mild Leydig cell dysfunction included. Furthermore, a relatively large proportion of patients were excluded from the cohort within 5-yr follow-up as TT and LH were evaluated in another laboratory which did not use radioimmunoassay. Finally, patients treated with abdominal RT or more treatment lines were not included and extrapolation to these groups cannot be made with certainty.
7
Author contributions: Mikkel Bandak had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Bandak, Juul, Daugaard, Jørgensen, Lauritsen, Kier. Acquisition of data: Bandak, Mortensen, Kier. Analysis and interpretation of data: Bandak, Juul, Daugaard, Jørgensen, Lauritsen, Kier, Mortensen. Drafting of the manuscript: Bandak. Critical revision of the manuscript for important intellectual content: Juul, Daugaard, Jørgensen, Lauritsen, Kier, Mortensen. Statistical analysis: Bandak, Lauritsen. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: None. Other: None. Financial disclosures: Mikkel Bandak certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: None. Acknowledgments: We thank the following source of funding: grants from Copenhagen University Hospital, Rigshospitalet, The Danish Cancer Society, and the Preben and Anna Simonsen Foundation.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.euf.2016.11. 018.
References [1] Sprauten M, Brydøy M, Haugnes HS, et al. Longitudinal serum
5.
Conclusions
testosterone, luteinizing hormone, and follicle-stimulating hormone levels in a population-based sample of long-term testicular
In summary, TT remained stable in the majority of patients during 5-yr follow-up, while there was a significant decline in LH in patients treated with BEP, suggesting a degree of Leydig cell recovery. We recommend to evaluate levels of TT and LH s at regular intervals during the 1st yr after TC treatment in order to identify the 11–15% of patients who according to our findings develop testicular failure requiring testosterone substitution. We suggest that patients are informed about symptoms of testosterone deficiency at the end of follow-up. Our results indicate that evaluation of Leydig cell function should be considered beyond clinical follow-up in BEPtreated patients with mild Leydig cell dysfunction (LH > 8 IU/l) during follow-up. Patients with combined LH/TT within the 97.5 percentile of healthy controls within follow-up had only small risk (4%) of developing primary testicular failure, and further evaluation of Leydig cell function might be omitted in this group of patients. However, this finding should be validated in a prospective study of a larger group of TC patients.
cancer survivors. J Clin Oncol 2014;32:571–8. [2] Aass N, Fossa˚ SD, Theodorsen L, Norman N. Prediction of long-term gonadal toxicity after standard treatment for testicular cancer. Eur J Cancer 1991;27:1087–91. [3] Eberhard J, Sta˚hl O, Cwikiel M, et al. Risk factors for post-treatment hypogonadism in testicular cancer patients. Eur J Endocrinol 2008;158:561–70. [4] Bandak M, Jørgensen N, Juul A, et al. Testosterone deficiency in testicular cancer survivors - a systematic review and meta-analysis. Andrology 2016. http://dx.doi.org/10.1111/andr.12177. [5] Nord C, Bjøro T, Ellingsen D, et al. Gonadal hormones in long-term survivors 10 years after treatment for unilateral testicular cancer. Eur Urol 2003;44:322–8. [6] Huddart RA, Norman A, Moynihan C, et al. Fertility, gonadal and sexual function in survivors of testicular cancer. Br J Cancer 2005;93:200–7. [7] Wiechno P, Demkow T, Kubiak K, Sadowska M, Kamin´ska J. The quality of life and hormonal disturbances in testicular cancer survivors in cisplatin era. Eur Urol 2007;52:1448–54. [8] Nuver J, Smit AJ, Wolffenbuttel BHR, et al. The metabolic syndrome and disturbances in hormone levels in long-term survivors of disseminated testicular cancer. J Clin Oncol 2005;23:3718–25.
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018
EUF-249; No. of Pages 8 8
EUROPEAN UROLOGY FOCUS XXX (2016) XXX–XXX
[9] Gerl A, Mu¨hlbayer D, Hansmann G, Mraz W, Hiddemann W. The
[14] Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy
impact of chemotherapy on Leydig cell function in long term
in men with androgen deficiency syndromes: an Endocrine Society
survivors of germ cell tumors. Cancer 2001;91:1297–303.
clinical practice guideline. J Clin Endocrinol Metab 2010;95:
[10] Aksglaede L, Andersson AM, Jørgensen N, et al. Primary testicular
2536–59.
failure in Klinefelter’s syndrome: The use of bivariate luteinizing
[15] de Haas EC, Altena R, Boezen HM, et al. Early development of the
hormone-testosterone reference charts. Clin Endocrinol (Oxf)
metabolic syndrome after chemotherapy for testicular cancer. Ann
2007;66:276–81.
Oncol 2013;24:749–55.
[11] Rasmussen SL, Torp-Pedersen C, Borch-Johnsen K, Ibsen H. Normal
[16] Haugnes HS, Aass N, Fossa˚ SD, et al. Components of the metabolic
values for ambulatory blood pressure and differences between
syndrome in long-term survivors of testicular cancer. Ann Oncol
casual blood pressure and ambulatory blood pressure: results from a Danish population survey. J Hypertens 1998;16:1415–24.
2007;18:241–8. [17] Willemse PM, Burggraaf J, Hamdy NAT, et al. Prevalence of the
[12] Jensen TK, Andersson AM, Hjollund NH, et al. Inhibin B as a serum
metabolic syndrome and cardiovascular disease risk in chemother-
marker of spermatogenesis: correlation to differences in sperm
apy-treated testicular germ cell tumour survivors. Br J Cancer
concentration and follicle-stimulating hormone levels. A study of 349 Danish men. J Clin Endocrinol Metab 1997;82:4059–63.
2013;109:60–7. [18] Agergaard Holmboe S, Vradi E, Kold Jensen T, et al. The association
[13] Palmieri G, Lotrecchiano G, Ricci G, et al. Gonadal function after
of reproductive hormone levels and all-cause, cancer and cardio-
multimodality treatment in men with testicular germ cell cancer.
vascular disease mortality in men. J Clin Endocrinol Metab 2015;
Eur J Endocrinol 1996;134:431–6.
100:4472–80.
Please cite this article in press as: Bandak M, et al. Longitudinal Changes in Serum Levels of Testosterone and Luteinizing Hormone in Testicular Cancer Patients after Orchiectomy Alone or Bleomycin, Etoposide, and Cisplatin. Eur Urol Focus (2016), http:// dx.doi.org/10.1016/j.euf.2016.11.018