University of Groningen Survivorship care after testicular cancer Boer, Hindrik

Survivorship care after testicular cancer

Boer, Hindrik

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2019

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Boer, H. (2019). Survivorship care after testicular cancer: New insights in late effects of treatment and approaches to shared-care follow-up. Rijksuniversiteit Groningen.

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

69

Single nucleotide polymorphism in the

5-α-reductase gene (SRD5A2) is associated

with increased prevalence of metabolic

syndrome in chemotherapy-treated

testicular cancer survivors

Chapter 4

4

Departments of 1_{Medical Oncology,} 2_{Laboratory Medicine,} 3_{Vascular Medicine and} 4_{Epidemiology,} University of Groningen, University Medical Center Groningen, Groningen, The Netherlands.

70

Abstract Purpose

Chemotherapy-treated testicular cancer survivors are at risk for development of the metabolic syndrome, especially in case of decreased androgen levels. Polymorphisms in the gene encoding steroid 5-α-reductase type II (SRD5A2) are involved in altered androgen metabolism. We investigated whether single-nucleotide polymorphisms (SNPs) rs523349 (V89L) and rs9282858 (A49T) in SRD5A2 are associated with cardiometabolic status in testicular cancer survivors. Methods

In 173 chemotherapy-treated testicular cancer survivors, hormone levels and cardiometabolic status were evaluated cross-sectionally (median 5 years [range 3-20] after chemotherapy) and correlated with SNPs in SRD5A2.

Results

The metabolic syndrome was more prevalent in survivors who were homozygous or heterozygous variant for SRD5A2 rs523349 compared to wild type (33% versus 19%, P = 0.032). In particular, patients with lower testosterone levels (<15 nmol/l) and a variant genotype showed a high prevalence of the metabolic syndrome (66.7%). Mean intima-media thickness of the carotid artery and urinary albumin excretion, both markers of vascular damage, were higher in the group of survivors homozygous or heterozygous variant for rs523349 (0.62 versus 0.57 mm, P = 0.026; 5.6 versus 3.1 mg/24h, P = 0.017, respectively). No association was found between cardiometabolic status and SNP rs9282858 in SRD5A2.

Conclusion

Metabolic syndrome develops more frequently in testicular cancer survivors homozygous or heterozygous variant for SNP rs523349 in SRD5A2. Altered androgen sensitivity appears to be involved in the development of adverse metabolic and vascular changes in testicular cancer survivors and is a target for intervention.

71

Introduction

Since the introduction of platinum-based chemotherapy, metastatic testicular cancer has become a curable disease and most testicular cancer patients have an excellent prognosis. However, long-term toxicity can undermine life after treatment of testicular cancer.1,2 _{Cardiovascular disease,} e.g. myocardial infarction, is more prevalent in testicular cancer survivors in comparison with age-matched controls.3,4 _{Several studies have shown that testicular cancer survivors are prone} to develop cardiovascular risk factors, often clustered in the metabolic syndrome, following orchidectomy and platinum-based chemotherapy.5-8

The metabolic syndrome comprises central obesity, dyslipidemia, hypertension and insulin resistance. Due to its pre-diabetic and pro-atherogenic characteristics the metabolic syndrome forms a seedbed for cardiovascular disease. Studies in the general population have shown that low levels of testosterone are associated with metabolic syndrome.9,10 _{The relationship} is probably bi-directional. On one hand, an increase in adipose tissue enhances aromatization of testosterone, thereby contributing to hypogonadism. On the other hand, testosterone and its more potent metabolite dihydrotestosterone have several regulatory functions in adipose tissue and lower levels of androgens may lead to adverse metabolic changes, such as central obesity and insulin resistance.11,12 _{Testosterone is also a vasoactive hormone, as well as a regulating factor} in glucose and lipid metabolism.13-15 _{Subclinical hypogonadism appears to be a prominent risk} factor for the metabolic syndrome in testicular cancer survivors.16 _{A substantial proportion of} testicular cancer survivors have a deteriorated gonadal function after chemotherapy that may persist for > 10 years.17,18 _{It is difficult to predict which cancer survivors would benefit from} testosterone supplementation. In a recent study by Finkelstein et al. in which 198 healthy men received different amounts of testosterone, the authors concluded that the amount of testosterone required for maintaining lean mass, fat mass, strength, and sexual function varied widely in men and argued that more personalized approaches for treating hypogonadism are needed.19

Genetic variations that result in functional changes in androgenic enzymes may explain why some patients are more at risk for hormonal or metabolic changes and may help identify patients that are more likely to benefit from interventions, like testosterone supplementation therapy. In a recent study by Aschim et al. it was shown that polymorphisms in genes involved in androgen metabolism may partially explain inter-individual differences in gonadal toxicity in testicular cancer survivors.20

Functional single-nucleotide polymorphisms (SNPs) in steroid 5-α-reductase type II (SRD5A2) result in variations in enzymatic activity of 5-α-reductase type II, affecting the conversion of testosterone into dihydrotestosterone.21 _{We investigated whether SNPs rs523349 (V89L) and} rs9282858 (A49T) in SRD5A2 are associated with cardiometabolic status in testicular cancer survivors.

72

Patients and methods

Study population

Patients belonged to a cohort of long-term survivors of metastatic non-seminomatous testicular cancer treated between 1977 and 2004 with platinum-based chemotherapy at University Medical Center Groningen.7 _{The study population consisted of 173 patients (Fig. 1). Inclusion criteria were} attained age <60 years and a follow-up duration of minimum 3 years and maximum 20 years. Because our aim was to investigate newly developed cardiovascular risk factors after therapy, exclusion criteria were history of cardiac disease (defined as myocardial infarction or coronary artery disease before or after chemotherapy) and radiotherapy to the mediastinum. Participants of a previous study on cardiovascular risk profiles in testicular cancer survivors were also excluded to prevent overlap among cohorts.22 _{The local medical ethics committee approved the study and} all participants gave written informed consent.

73 Original population

(n = 439)

Disseminated non-seminomatous testicular cancer Platinum-based chemotherapy < April 2004

Survivors approached for study participation (n = 212)

Study population (n = 173)

Cross-sectional evaluation of the metabolic syndrome and associated features

Excluded (n = 121):

≥ 60 years and/or follow-up period > 20 years (n = 46)

55)

No informed consent (n = 39) Not amenable for participation (n = 106):

74

Metabolic, vascular and hormonal assessments

For assessment of the metabolic syndrome, the AHA/NHLBI classification was used with the metabolic syndrome present in case of three or more of the following criteria: central obesity (waist circumference ≥102 cm), high triglycerides (≥1.7 mmol/l or medication), low HDL-cholesterol (<1.03 mmol/l or medication), high blood pressure (systolic ≥130 mmHg or diastolic ≥85 mmHg or medication), and high glucose (≥5.6 mmol/l or medication).23 _{Glucose, triglycerides, total} cholesterol, HDL cholesterol, testosterone, follicle-stimulating hormone and luteinizing hormone were measured in fasting blood samples collected in the morning to account for the circadian rhythm of testosterone. Von Willebrand factor (vWF), plasminogen activator inhibitor type-1 (PAI-1), tissue-type plasminogen activator (t-PA) (citrate plasma), and urinary albumin excretion (24-h urine sample) were measured as described previously.24 _{Blood pressure, weight, length, and} waist and hip circumference were measured during a visit at the outpatient clinic. Blood pressure was measured in supine position after a 10-minute resting period. To assess vascular structure, the intima-media thickness (IMT) of the common carotid artery was measured as described previously.24 _{Part of these data were previously reported by De Haas et al.}7

Urinary steroid profiling

Gas chromatography-mass spectrometry profiling of urinary steroids was performed to quantify urinary output of steroid metabolites. This assay is based on the method of Shackleton 25 and is performed routinely at the clinical chemistry department of our medical centre. Quantification was performed using an Agilent 6890N Gas Chromatograph (Agilent, Santa Clara, CA, USA), equipped with an Agilent 7683 Automatic Liquid Sampler, coupled to an Agilent 5973N Mass Selective Detector. The ratio between etiocholanolone (a 5β C19 steroid) and androsterone (a 5α C19 steroid) was determined as an index of 5-α-reductase activity.26

DNA collection and genotyping

Germline-DNA was isolated from ethylenediaminetetraacetic acid blood collected at the outpatient clinic. Serum-derived DNA was amplified and quality checked (REPLI- g Service; Qiagen, Hilden, Germany). Serum samples deemed ‘usable’ or ‘highly usable’ were used for genotyping. The A49T and L89V variants in the SRD5A2 gene were assessed with an allelic discrimination assay on an ABI Prism 7500 real-time PCR system (Applied Biosystems [Life Technologies], Carlsbad, CA, USA). Genotyping of the A49T variant (rs9282858) was performed with a custom Taqman SNP Genotyping Assay (Applied Biosystems) consisting of a forward primer 5’- GCACACGGAGAGCCTGAAG-3’ and a reverse primer 5’- CAGCTCCTGCAGGAACCA-3’ in combination with a 5’-VIC-CTGCCAGCCCGCG- MGB-NFQ probe to identify the A-variant and a 5’-FAM-CTGCCAACCCGCG-MGB- NFQ probe to identify the T-variant of the SRD5A2 gene. Genotyping of the L89V variant (rs523349) was performed with a validated TaqMan SNP Genotyping Assay from Applied Biosystems (ID C_2362601_10). To verify results, sequencing of samples representing each genotype was performed on an ABI Prism 3130 Genetic Analyzer (Applied Biosystems).

75

Statistical analysis

Continuous variables were described with median and range, and categorical variables were described with counts and proportions. Univariate analyses were performed with t-test for normally distributed variables and Mann-Whitney U test for non-normally distributed variables. Hardy-Weinberg equilibrium was determined using Chi-square test. Univariate analysis of categorical variables was performed using Chi-square test. The effect of genotype on the risk of the metabolic syndrome, IMT and albumin was tested with multivariate regression analysis. All tests were two sided and were conducted at the 0.05 significance level. Statistical analyses were performed by using IBM SPSS statistics 20.0 (IBM Corp., Armonk, NY, USA).

Results

Study population characteristics

The basic characteristics of the study population are described in table 1. Median age at start of chemotherapy and during the follow-up assessment was 28 years and 37 years, respectively. Median follow-up of patients was 5 years (range 3-20). Metabolic syndrome was prevalent in 44 of 173 patients (25%), as reported previously by De Haas et al.7 _{Prevalence of individual components} of the metabolic syndrome and hormonal status are reported in table 1. A cut-off value of total testosterone <15 nmol/L was considered best discriminating for the metabolic syndrome in the study by De Haas et al.7

Genotyping results for SRD5A2 are presented in table 2. Genotype distribution of SNPs in SRD5A2 was comparable with the background population.27,28 _{Both SNPs were in Hardy-Weinberg} equilibrium. Because of the low frequency of homozygous variants in SNP rs523349 and SNP rs9282858 (10.4% and 0%, respectively), the heterozygous and homozygous variants were combined in further analyses. Thirteen patients in total were heterozygous for SNP rs9282858; ten of these patients were wild type for SNP rs523349 and three were heterozygous for SNP rs523349.

76

Table 1. Demographic, diagnostic, metabolic and hormonal characteristics of the study population,

previously reported by De Haas et al.7

BEP: bleomycin, etoposide, cisplatin; HDL: high-density lipoprotein; FSH: follicle-stimulating hormone; LH: luteinizing hormone. * Stage of disease according to the Royal Marsden Classification. **International Germ Cell Cancer Collaborative Group. *** Patients using testosterone supplementation excluded.

11

Tables chapter 4

Table 1. Demographic, diagnostic, metabolic and hormonal characteristics of the study population, previously reported by De Haas et al.7

median / n Range / %

Cohort size 173 100%

Age at chemotherapy (years) 28 16 – 52

Follow-up duration (years) 5 3 – 20

Age at end follow-up (years) 37 19 – 59 Initial stage of disease*

II 95 55%

III 12 7%

IV 66 38%

IGCCCG risk group** Good 103 60%

Intermediate 54 31%

Poor 16 9%

Chemotherapy regime BEP/EP 159 92%

Other 14 8%

Metabolic syndrome Overall prevalence 44 / 173 25% Central obesity 29 / 170 17% High triglycerides 50 / 173 29% Low HDL-cholesterol 76 / 173 44% Hypertension 100 / 171 58% High glucose 24 / 170 14% Gonadal status*** Total testosterone (nmol/l) 17.0 5 – 37

FSH (U/L) 14.6 2.38 – 87.5

LH (U/L) 6.1 1.59 – 33.1 Testosterone

supplementation

No supplementation 164 95% Yes, bilateral orchidectomy 7 4% Yes, low testosterone levels 2 1%

BEP: bleomycin, etoposide, cisplatin; HDL: high-density lipoprotein; FSH: follicle-stimulating hormone; LH: luteinizing hormone. *Stage of disease according to the Royal Marsden Classification. **International Germ Cell Cancer Collaborative Group. *** Patients using testosterone supplementation excluded.

77

Table 2. Genotyping results of SNP rs523349 and SNP rs9282858 in SRD5A2.

SNP: single-nucleotide polymorphism

* (Genotype rs9282858 not determinable for 1 patient)

12 Table 2. Genotyping results of SNP rs523349 and SNP rs9282858 in SRD5A2.

Genotype distribution of SNP rs523349 N % Background27

Wild type (VV) 91 52.6 61.2

Heterozygous (VL) 64 37.0 32.3

Homozygous (LL) 18 10.4 6.5

Genotype distribution of SNP rs9282858 N % Background28

Wild type (AA) 159 92.4 94.8

Heterozygous (AT) 13 7.6 4.7

Homozygous (TT) 0 0 0.5

Not determinable 1

Double mutants in SRD5A2 rs9282858

AA AT Total rs523349 VV 80 10 90 VL 61 3 64 LL 18 0 18 Total 159 13 172*

SNP: single-nucleotide polymorphism. *Genotype rs9282858 not determinable for 1 patient.

Table 3: Comparison of metabolic, vascular and hormonal characteristics in genotype groups for SNP rs523349 (V89L) in SRD5A2.

78

Metabolic status and SNPs in SRD5A2

Metabolic syndrome was more prevalent in testicular cancer survivors homozygous or heterozygous variant for SNP rs523349 compared with wild type (33% versus 19%, P = 0.032) (Table 3). Age, follow-up duration, cisplatin dose, body mass index (BMI) and testosterone levels were comparable in both groups. Triglycerides were higher in the group of survivors homozygous or heterozygous variant for rs523349, 1.44 versus 1.29 mmol/l (P = 0.049). Multiple logistic regression analysis revealed that the odds ratio for metabolic syndrome as predicted by a variant genotype (VL/LL) for SNP rs523349 is 2.56 (95% confidence interval [CI] 1.02 - 6.40, P = 0.044), adjusted for age at follow-up, follow-up duration, BMI and testosterone level. (Appendix table A.1). The prevalence of the metabolic syndrome was particularly increased in the group of patients homozygous or heterozygous variant for SNP rs523349 combined with testosterone levels <15 nmol/l (66.7%) (Fig. 2). For SRD5A2 SNP rs9282858 we found no differences in metabolic status between the 159 wild-type patients and the 13 patients with a variant genotype.

79

Table 3. Comparison of metabolic, vascular and hormonal characteristics in genotype groups for SNP

rs523349 (V89L) in SRD5A2.

E/A ratio: ratio between etiocholanolone and androsterone; IMT: Intima Media Thickness; vWF: von Willebrand Factor; PAI: plasminogen activator inhibitor-1; t-PA: tissue plasminogen activator; SD: standard deviation; SNP: single-nucleotide polymorphism. The P-values below 0.05 (significance level) are represented in bold.

*Chi-square test, t-test or Mann-Whitney U test used when appropriate. **One hundred sixty-six patients received cisplatin-based chemotherapy. ***Patients using testosterone supplementation excluded.

13

Table 3: Comparison of metabolic, vascular and hormonal characteristics in genotype groups for SNP rs523349

(V89L) in SRD5A2.

Wildtype (VV) (n=91) Variant (VL/LL) (n=82)

Median Range Median Range p*

Age years 36.8 19.2 - 58.6 36.5 22.8 - 55.7 .39

Follow-up duration years 4.9 2.8 - 20.2 5.5 2.9 - 19.6 .23 Cisplatin dose** mg/m2 _{400 200 - 800 400 275 - 800 .13}

Metabolic and hormonal status

Metabolic syndrome 19% 33% .032

Body mass index kg/m2 _{24.6 18.5 - 38.8 24.7 17.4 - 38.7 .95}

Triglycerides mmol/l 1.29 0.35 - 5.13 1.44 0.46 - 10.94 .049

Testosterone*** nmol/l 17.5 5 - 32 17.0 9 - 37 .91

Vascular status

IMT (mean value, SD) mm 0.57 0.12 0.62 0.12 .026

Albumin excretion mg/24h 3.1 0.1 - 615 5.6 0.1 - 158 .017

vWF % 100 28 - 296 94 47 - 205 .16

PAI-1/-tPA ratio 3.4 0.6 - 16.2 3.7 0.5 - 31.6 .42

Urinary steroid profile

Etiocholanolone (E) µmol/24h 7.7 1.6 - 25.0 7.9 2.2 - 32.7 .61 Androsterone (A) µmol/24h 11.6 2.3 - 33.7 12.4 2.6 - 28.5 .79

E/A ratio 0.67 0.25 - 3.42 0.66 0.30 - 2.26 .48

E/A ratio: ratio between etiocholanolone and androsterone; IMT: Intima Media Thickness; vWF: von Willebrand Factor; PAI: plasminogen activator inhibitor-1; t-PA: tissue plasminogen activator; SD: standard deviation; SNP: single-nucleotide polymorphism.

The p-values below 0.05 (significance level) are represented in bold. *Chi-square test, t-test or Mann-Whitney U test used when appropriate. **One hundred sixty-six patients received cisplatin-based chemotherapy. ***Patients using testosterone supplementation excluded.

80

Vascular status and SNPs in SRD5A2

IMT of the carotid artery was higher in patients homozygous or heterozygous variant for SRD5A2 SNP rs523349 (0.62 versus 0.57 mm, P = 0.026) (Table 3). In a multiple linear regression model in which IMT was predicted by genotype and adjusted for age at follow-up and follow-up duration, this association remained significant (beta 0.15, P = 0.037, Appendix table A.2). Albumin excretion was higher in patients with a variant genotype for SNP rs523349 (5.6 versus 3.1 mg/24h, v = 0.017). In a multiple linear regression model in which albumin excretion was predicted by genotype and adjusted for age at follow-up and follow-up duration, this association was also significant (beta 0.16, P = 0.040, Appendix table A.3). Levels of vWF and PAI / t-PA ratio were comparable in both genotype groups (Table 3). For rs9282858, we did not find differences in status between the 159 wild-type patients and the 13 patients who were heterozygous for the variant.

Functionality of the 5-α-reductase enzyme

The ratio between etiocholanolone and androsterone (E/A ratio) as an index for the enzymatic activity of 5-α-reductase was comparable in patients homozygous or heterozygous for SNP rs523349 versus wild type: 0.66 versus 0.67 (Table 3). Levels of etiocholanolone, androsterone, and the E/A ratio in subgroups based on testosterone level are shown in Table 4.

17% 33% 20% 67% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Testosterone ≥ 15 nmol/L Testosterone < 15 nmol/L

Pr ev al en ce o f M et ab ol ic S yn dr om e (% ) Wildtype Variant

Fig. 2. Prevalence of metabolic syndrome in groups based on single-nucleotide

polymorphism rs523349 and testosterone levels higher or lower than 15 nmol/l. Body mass index pre-chemotherapy was comparable in wild type and variant groups.

81

Table 4. Comparison of the urinary steroid profiles in genotype groups for SNP rs523349 (V89L) in SRD5A2,

subdivided in groups based on testosterone levels (patients receiving testosterone supplementation excluded).

E/A ratio: ratio between etiocholanolone and androsterone; SNP: single-nucelotide polymorphism. *Mann-Whitney U test.

Discussion

We found an association between SNP rs523349 in the gene SRD5A2 and the prevalence of the metabolic syndrome in testicular cancer survivors. Our findings suggest a role of impaired androgen metabolism in the etiology of the metabolic syndrome in testicular cancer patients after orchidectomy and cisplatin-based chemotherapy. The metabolic syndrome was present in 25% of the total testicular cancer survivor population studied, as reported earlier.7 Patients with testosterone levels <15 nmol/l (22% of the cohort) had an odds ratio of 4.1 (95% CI 1.8 – 9.3) for developing or having the metabolic syndrome after chemotherapy. The current study suggests that genetic variations related to the androgen metabolism are involved in the development of an adverse cardiometabolic profile after treatment for testicular cancer. The group of patients with both a low testosterone (<15 nmol/l) and the variant genotype for SNP rs523349 showed a high prevalence (66.7%) of the metabolic syndrome.

14

Table 4. Comparison of the urinary steroid profiles in genotype groups for SNP rs523349 (V89L) in SRD5A2, subdivided in groups based on testosterone levels (patients receiving testosterone supplementation excluded).

Wildtype (VV) (n=67) Variant (VL/LL) (n=52)

Testosterone ≥15 nmol/l Median Range Median Range p* Etiocholanolone (µmol/24h) 7.6 1.6 - 25.0 8.7 2.2 - 29.7 .25 Androsterone (µmol/24h) 11.6 2.3 - 33.7 13.1 3.6 - 27.6 .50 E/A ratio 0.68 0.30 - 3.42 0.69 0.30 - 2.26 .40

Wildtype (VV) (n=14) Variant (VL/LL) (n=20) Testosterone <15 nmol/l Median Range Median Range p* Etiocholanolone (µmol/24h) 7.5 3.1 - 14.3 6.4 2.8 - 32.7 .36 Androsterone (µmol/24h) 11.3 6.1 - 15.7 10.3 2.6 - 28.5 .62 E/A ratio 0.67 0.39 - 1.28 0.61 0.32 - 1.34 .92

E/A ratio: ratio between etiocholanolone and androsterone; SNP: single-nucelotide polymorphism. *Mann-Whitney U test.

82

The gene SRD5A2 encodes the 5-α-reductase enzyme, which converts testosterone into dihydrotestosterone. Although dihydrotestosterone levels are considerably lower than testosterone, dihydrotestosterone has a five-fold stronger binding capacity to the androgen receptor and, thereby, a considerably more potent androgenic effect. In vitro the variant genotype for SNP rs523349 is associated with decreased activity of the 5-α-reductase enzyme, thus leading to lower levels of dihydrotestosterone.21 _{Compared to testosterone, the effect} of dihydrotestosterone on metabolic risk factors is less well understood. In a population-based study of Yeap et al., several associations with metabolic risk factors were equal between dihydrotestosterone and testosterone, but additional distinct associations with diabetes and fasting glucose indicated an additive role of dihydrotestosterone.29

Recently, Aschim et al., reported on several polymorphisms that modify the effect of different treatments on endocrine status.20 _{They found that patients heterozygous for SNP rs9282858} in SRD5A2 had lower testosterone levels after radiotherapy compared with patients who were treated with surgery only. Because of the low numbers of patients heterozygous for this SNP rs9282858, which results in increased functionality of 5-α-reductase in in vitro studies, we could not properly analyze this in our study. SNPs in SRD5A2, in particular rs523349, have been studied extensively to determine polymorphisms that are associated with prostate cancer risk and biochemical recurrence after prostate cancer. Although individual studies have yielded conflicting results, meta-analyses point to a lack of association between rs523349 and prostate cancer risk.30

In the urinary steroid profile analysis we observed no differences in E/A ratio between wild type and variant genotype for rs523349. One would expect lower levels of androsterone and a relatively high E/A ratio in case of lower functionality of 5-α- reductase. E/A ratio in the normal population shows a wide variation; for men aged 17 - 50 years, the reference value is 0.4 - 1.8.31 Differences in functionality of 5-α- reductase are probably too subtle and not detectable by comparing the ratio of these metabolites. By using direct determination of dihydrotestosterone levels, we would have been able to get a more accurate estimate of 5-α-reductase activity. However, accurate determination of plasma dihydrotestosterone remains a technical challenge, although recently progress has been made.32,33

We also observed differences in vascular parameters: IMT and albumin excretion were higher in patients who were homozygous or heterozygous variant for SNP rs523349. This could indicate that patients with a variant genotype are prone to subclinical vascular damage, although the vascular changes could also be secondary to adverse metabolic changes. The vascular effects of testosterone and dihydrotestosterone are not yet fully understood. Testosterone was found to be a negative predictor of arterial stiffness and the carotid IMT, but the specific effect of dihydrotestosterone is largely unknown.15

Strong points of the current study are the relatively large study-population size, extent and variety of measured parameters, and measurements of androgenic profile and urinary steroid profile. The cross-sectional design and the large range in follow-up duration hamper the possibility to analyze causal factors in the development of metabolic syndrome over time.

These results are hypothesis generating and need to be validated in an independent larger cohort of testicular cancer patients. Exploration of this hypothesis in a cohort of testicular cancer patients, treated with chemotherapy or orchiectomy only, and age-matched controls

83

would be preferable to determine the effect of different treatment modalities.

In conclusion, we found an association between SNP rs523349 in SRD5A2 and the metabolic syndrome, which is an important late effect of platinum-based chemotherapy. Patients homozygous or heterozygous variant for SNP rs523349 have an odds ratio of 2.56 for the metabolic syndrome after treatment for metastatic testicular cancer compared to wild-type patients. In patients with lower testosterone levels and a variant genotype for SNP rs523349, the prevalence of the metabolic syndrome is even higher. Validation of these findings is necessary, and when confirmed, these data may help to identify patients who are particularly prone to develop cardiovascular disease as a late effect of chemotherapy for testicular cancer. These patients may benefit from early intervention strategies, i.e. a physical exercise programme or testosterone supplementation. Further exploration of the role of androgens and genetic susceptibility in testicular cancer survivors is needed to facilitate these early interventions.

Funding

84

References

Travis LB, Beard C, Allan JM, et al. Testicular cancer survivorship: research strategies and recommendations. J Natl Cancer Inst. 2010;102:1114-1130.

Beyer J, Albers P, Altena R, et al. Maintaining success, reducing treatment burden, focusing on survivorship: highlights from the third European consensus conference on diagnosis and treatment of germ-cell cancer. Ann Oncol. 2013;24:878-888.

van den Belt-Dusebout AW, Nuver J, de Wit R, et al. Long-term risk of cardiovascular disease in 5-year survivors of testicular cancer. J Clin Oncol. 2006;24:467-475.

van den Belt-Dusebout AW, de Wit R, Gietema JA, et al. Treatment-specific risks of second malignancies and cardiovascular disease in 5-year survivors of testicular cancer. J Clin Oncol. 2007;25:4370-4378. Nuver J, Smit AJ, Wolffenbuttel BH, et al. The metabolic syndrome and disturbances in hormone levels in long-term survivors of disseminated testicular cancer. J Clin Oncol. 2005;23:3718-3725.

Haugnes HS, Aass N, Fosså SD, et al. Components of the metabolic syndrome in long-term survivors of testicular cancer. Ann Oncol. 2007;18:241-248.

de Haas EC, Altena R, Boezen HM, et al. Early development of the metabolic syndrome after chemotherapy for testicular cancer. Ann Oncol. 2013;24:749-755.

Willemse PM, Burggraaf J, Hamdy NA, et al. Prevalence of the metabolic syndrome and cardiovascular disease risk in chemotherapy-treated testicular germ cell tumour survivors. Br J Cancer. 2013;109:60-67. Brand JS, van der Tweel I, Grobbee DE, Emmelot-Vonk MH, van der Schouw YT. Testosterone, sex hormone-binding globulin and the metabolic syndrome: a systematic review and meta-analysis of observational studies. Int J Epidemiol. 2011;40:189-207.

Haring R, Völzke H, Steveling A, et al. Low serum testosterone levels are associated with increased risk of mortality in a population-based cohort of men aged 20-79. Eur Heart J. 2010;31:1494-1501.

Ruige JB, Ouwens DM, Kaufman JM. Beneficial and adverse effects of testosterone on the cardiovascular system in men. J Clin Endocrinol Metab. 2013;98:4300-4310.

Brand JS, Rovers MM, Yeap BB, et al. Testosterone, sex hormone-binding globulin and the metabolic syndrome in men: an individual participant data meta-analysis of observational studies. PLoS One. 2014;9:e100409.

Blouin K, Veilleux A, Luu-The V, Tchernof A. Androgen metabolism in adipose tissue: recent advances. Mol Cell Endocrinol. 2009;301:97-103.

Kelly DM, Jones TH. Testosterone: a metabolic hormone in health and disease. J Endocrinol. 2013;217:R25-45.

Kelly DM, Jones TH. Testosterone: a vascular hormone in health and disease. J Endocrinol. 2013;217:R47-71.

de Haas EC, Oosting SF, Lefrandt JD, Wolffenbuttel BH, Sleijfer DT, Gietema JA. The metabolic syndrome in cancer survivors. Lancet Oncol. 2010;11:193-203.

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;44:322-328.

Sprauten M, Brydøy M, Haugnes HS, et al. Longitudinal serum testosterone, luteinizing hormone, and follicle-stimulating hormone levels in a population-based sample of long-term testicular cancer survivors. J Clin Oncol. 2014;32:571-578.

Finkelstein JS, Lee H, Burnett-Bowie SA, et al. Gonadal steroids and body composition, strength, and sexual function in men. N Engl J Med. 2013;369:1011-1022.

Aschim EL, Oldenburg J, Kristiansen W, et al. Genetic variations associated with the effect of testicular cancer treatment on gonadal hormones. Hum Reprod. 2014;29:2844-2851.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

85 Makridakis NM, di Salle E, Reichardt JK. Biochemical and pharmacogenetic dissection of human steroid 5 alpha-reductase type II. Pharmacogenetics. 2000;10:407-413.

Meinardi MT, Gietema JA, van der Graaf WT, et al. Cardiovascular morbidity in long-term survivors of metastatic testicular cancer. J Clin Oncol. 2000;18:1725-1732.

Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112:2735-2752.

Nuver J, Smit AJ, Sleijfer DT, et al. Microalbuminuria, decreased fibrinolysis, and inflammation as early signs of atherosclerosis in long-term survivors of disseminated testicular cancer. Eur J Cancer. 2004;40:701-706. Shackleton CH. Profiling steroid hormones and urinary steroids. Journal of Chromatography 1986; 379: 91-156

Mareck U, Geyer H, Opfermann G, Thevis M, Schänzer W. Factors influencing the steroid profile in doping control analysis. J Mass Spectrom. 2008;43:877-891.

NCBI. dbSNP: rs523349. http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?rs=523349. NCBI. dbSNP: rs9282858. http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?rs=9282858.

Yeap BB, Knuiman MW, Divitini ML, et al. Differential associations of testosterone, dihydrotestosterone and oestradiol with physical, metabolic and health-related factors in community-dwelling men aged 17-97 years from the Busselton Health Survey. Clin Endocrinol (Oxf ). 2014;81:100-108.

Li Q, Zhu Y, He J, et al. Steroid 5-alpha-reductase type 2 (SRD5A2) V89L and A49T polymorphisms and sporadic prostate cancer risk: a meta-analysis. Mol Biol Rep. 2013;40:3597-3608.

Weykamp CW, Penders TJ, Schmidt NA, Borburgh AJ, van de Calseyde JF, Wolthers BJ. Steroid profile for urine: reference values. Clin Chem. 1989;35:2281-2284.

Wilton JH, Titus MA, Efstathiou E, et al. Androgenic biomarker prof|ling in human matrices and cell culture samples using high throughput, electrospray tandem mass spectrometry. Prostate. 2014;74:722-731. Ke Y, Bertin J, Gonthier R, et al. A sensitive, simple and robust LC-MS/MS method for the simultaneous quantification of seven androgen- and estrogen-related steroids in postmenopausal serum. J Steroid Biochem Mol Biol. 2014;144:523-534.

21 22 23 24 25 26 27 28 29 30 31 32 33

86

15

Supplementary table 1-1. Multiple logistic regression model for metabolic syndrome at follow-up with SNP

rs523349 (V89L) and known confounding factors as predictor (n = 164, patients using testosterone supplementation excluded).

_{O.R. 95% C.I. p-value}

SNP rs523349 2.56 1.02 - 6.40 .044

Age at follow-up 1.04 0.98 - 1.10 .234

Follow-up duration 1.02 0.93 - 1.12 .691

Body Mass Index* 1.54 1.30 - 1.82 .000

Testosterone** 0.99 0.91 - 1.08 .851

*Body mass index (BMI) was added in this model because of the association between testosterone and BMI.

**Testosterone were added in this model because of the association between testosterone and metabolic syndrome.

Supplementary table 1-2. Multiple logistic regression model for metabolic syndrome at follow-up with SNP

rs523349 (V89L) and known confounding factors as predictor.

_{O.R. 95% C.I. p-value}

SNP rs523349 2.13 1.03 - 4.37 .040

Age at follow-up 1.05 1.00 - 1.10 .027

Follow-up duration 1.04 0.96 - 1.13 .338

Appendix tables

Supplementary table A.1-1. Multiple logistic regression model for metabolic syndrome at follow-up with

SNP rs523349 (V89L) and known confounding factors as predictor (n = 164, patients using testosterone supplementation excluded)

*Body mass index (BMI) was added in this model because of the association between testosterone and BMI. **Testosterone were added in this model because of the association between testosterone and metabolic syndrome.

87

Supplementary table A.1-2. Multiple logistic regression model for metabolic syndrome at follow-up

with SNP rs523349 (V89L) and known confounding factors as predictor

Supplementary table 1-3. Multiple linear regression model for intima media thickness at follow-up with

SNP rs523349 (V89L) and known confounding factors as predictor.

Supplementary table 1-4. Multiple linear regression model for albumin excretion in 24-h urine at follow-up

(log-transformed) with SNP rs523349 (V89L) and known confounding factors as predictor

SNP in the SRD5A2 gene and the metabolic syndrome in testicular cancer survivors

16

Supplementary table 1 -3 . Multiple linear regression model for intima media thickness at follow-up with SNP

rs523349 (V89L) and known confounding factors as predictor.

Supplementary table 1 -4 . Multiple linear regression model for albumin excretion in 24-h urine at follow-up

(log-transformed) with SNP rs523349 (V89L) and known confounding factors as predictor.

B Std. Error β p-value Constant _{-.514 .665 .441} SNP rs523349 _{.539 .261 .157 .040} Age at follow-up _{.019 .017 .093 .275} Follow-up duration _{.040 .033 .103 .225} B Std. Error β p-value Constant _{.316 .044 .000} SNP rs523349 _{.037 .017 .150 .037} Age at follow-up _{.006 .001 .440 .000} Follow-up duration _{.000 .002 -.012 .875}

16

Supplementary table 1 -3 . Multiple linear regression model for intima media thickness at follow-up with SNP

rs523349 (V89L) and known confounding factors as predictor.

Supplementary table 1 -4 . Multiple linear regression model for albumin excretion in 24-h urine at follow-up

(log-transformed) with SNP rs523349 (V89L) and known confounding factors as predictor.

B Std. Error β p-value Constant _{-.514 .665 .441} SNP rs523349 _{.539 .261 .157 .040} Age at follow-up _{.019 .017 .093 .275} Follow-up duration _{.040 .033 .103 .225} B Std. Error β p-value Constant _{.316 .044 .000} SNP rs523349 _{.037 .017 .150 .037} Age at follow-up _{.006 .001 .440 .000} Follow-up duration _{.000 .002 -.012 .875}

15

Supplementary table 1-1. Multiple logistic regression model for metabolic syndrome at follow-up with SNP

rs523349 (V89L) and known confounding factors as predictor (n = 164, patients using testosterone supplementation excluded).

_{O.R. 95% C.I. p-value}

SNP rs523349 2.56 1.02 - 6.40 .044

Age at follow-up 1.04 0.98 - 1.10 .234

Follow-up duration 1.02 0.93 - 1.12 .691

Body Mass Index* 1.54 1.30 - 1.82 .000

Testosterone** 0.99 0.91 - 1.08 .851

*Body mass index (BMI) was added in this model because of the association between testosterone and BMI.

**Testosterone were added in this model because of the association between testosterone and metabolic syndrome.

Supplementary table 1-2. Multiple logistic regression model for metabolic syndrome at follow-up with SNP

rs523349 (V89L) and known confounding factors as predictor.

_{O.R. 95% C.I. p-value}

SNP rs523349 2.13 1.03 - 4.37 .040

Age at follow-up 1.05 1.00 - 1.10 .027

Follow-up duration 1.04 0.96 - 1.13 .338