Testicular carcinoma in situ in patients with extragonadal germ-cell tumours: the clinical role of pretreatment biopsy

S. D. Fosså1,+, N. Aass2, A. Heilo3, G. Daugaard4, N. E. Skakkebæk5, A. E. Stenwig6, J. M. Nesland6, L. H. J. Looijenga7 and J. W. Oosterhuis7

Departments of 1 Clinical Research, 2 Oncology, 3 Radiology and 6 Pathology, The Norwegian Radium Hospital, Oslo, Norway; Departments of 4 Oncology and 5 Growth and Reproduction, Danish National Hospital, Copenhagen, Denmark; 7 Department of Pathology, Laboratory for Experimental Patho-Oncology, Erasmus MC, University Medical Center Rotterdam, Josephine Nefkens Institute, Rotterdam, The Netherlands

Received 12 March 2002; accepted 26 May 2003


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

The aim of this study was to determine the prevalence of testicular carcinoma in situ (CIS) in patients with a malignant extragonadal germ-cell tumour (EGGCT) and the incidence of metachronous invasive testicular cancer (TC) in relation to the pretreatment demonstration of CIS.

Patients and methods:

Sixty-eight patients with EGGCT (53 retroperitoneal, 15 mediastinal) had pre-chemotherapy histological assessment of one (13) or both (55) testicle(s). A total of 123 testicles were examined for the presence of CIS.

Results:

Testicular CIS was found in 21 patients (31%) (18 retroperitoneal EGGCT, three mediastinal EGGCT). Two patients had bilateral CIS. Five patients, four of them with proven pretreatment CIS, developed a metachronous TC. The 10-year invasive-free TC survival rate for all 68 patients was 88%, but only 65% for those with proven pretreatment CIS. The overall 10-year survival rate for all patients was 82%. CIS was demonstrated in seven of 48 trans-scrotal core biopsies, in 10 of 56 trans-scrotal surgical biopsies and in five of 11 orchiectomy specimens.

Conclusions:

Approximately one-third of patients with EGGCT present with testicular CIS, predominantly those with a retroperitoneal tumour. These patients have a considerable risk of metachronous TC development in spite of chemotherapy. The pretreatment demonstration of testicular CIS in patients with EGGCT gives the possibility of individualised counselling and safe follow-up, and is therefore highly recommended. The data are in agreement with a multi-site development of malignant germ-cell tumours, but do not exclude the possibility that the retroperitoneal EGGCTs in particular represent metastases from a burned-out TC.

Key words: extragonadal germ-cell malignancy, invasive testicular cancer, testicular carcinoma in situ


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Between 2% and 5% of malignant germ-cell tumours in men arise at extragonadal sites [1, 2]. Cytogenetically most extragonadal germ-cell tumours (EGGCTs), i.e. the seminomas and non-seminomas, are similar to their testicular counterparts [3, 4]. In these tumours the malignant process is believed to start during early embryonic life, affecting primordial cells located in the midline of the embryo. Usually most of these cells regress or migrate towards the gonadal blastema to become the stem cells of spermatogenesis in the male. In some cases this regression might be disturbed, resulting in a misplacement of primordial cells at different sites at the body’s midline. If carcinogenic agents involved in malignant transformation of germ cells are present during embryonic or later life, carcinogenesis may take place in these residual primordial cells as in cells which subsequently develop into the testicular germinative epithelium. This hypothesis would explain the occurrence of germ-cell tumours at multiple sites. Micro-environmental conditions are believed to determine the final phenotype of the tumours at different sites with remarkably similar genotypic alterations.

True EGGCTs are thus diagnosed in the midline of the brain, in anterior mediastinum and in the retroperitoneal space. The latter localisation represents a particular clinical problem: the differential diagnosis as to metastases from a primary spontaneously regressed (‘burned-out’) invasive testicular cancer (TC) may be extremely difficult or impossible [57]. Clinicians are aware of this differential diagnosis, but discrimination between a true abdominal EGGCT and metastases from a completely regressed TC may be impossible in clinical practice. Most clinicians therefore define a germ-cell malignancy as an EGGCT if pretreatment examinations, including testicular biopsies, ultrasonography and/or orchiectomy have excluded the presence of an infiltrative TC. This definition has also been adopted in the present study.

EGGCT is today considered to be a curable malignancy, although patients most often present with advanced stages [2, 8]. The common aetiology of TC and EGGCT is probably one of the reasons for the high responsiveness of EGGCT to cisplatin, etoposide, ifosfamide and other cytostatic drugs successfully used in metastatic TC.

About 4% of patients with EGGCT develop a TC during a 60-month follow-up period [9], similar to patients with unilateral TC, in whom an infiltrative contralateral TC is observed in 4–5% of patients [10]. In both conditions testicular carcinoma in situ (CIS) is believed to be the precursor lesion of an infiltrative tumour [11, 12]. The cumulative risk of metachronous invasive TC is reported to be 14% in patients with retroperitoneal EGGCT, and 6% in those with a mediastinal EGGCT [9]. A figure of 4% is below the expected incidence, taking into account that testicular CIS has been demonstrated in 16 of 48 patients (33%) with EGGCT in one series [12]. This discrepancy between TC and EGGCT thus requires an explanation in expanded series with long follow-up. One may speculate whether cisplatin-based chemotherapy, given to almost all patients with EGGCT, abolishes testicular CIS or at least reduces its proliferative ability and invasive potential in spite of the claimed blood–testis barrier. This has been suggested in two previous series [13, 14]. Furthermore, the observed incidence of subsequent infiltrative TC is dependent on an institution’s strategy to apply immediate, delayed or no local treatment (orchiectomy or irradiation) to the CIS-involved testicle [13, 15].

On this background we have reviewed the experience of the Norwegian Radium Hospital (NRH), Oslo and that of the National Hospital, Copenhagen, as to the prevalence of testicular CIS in patients with EGGCT and the development of metachronous invasive TC. The aim was to support the future use of pretreatment testicular biopsies in patients with EGGCT or to abolish this strategy.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients
Patients with extracerebral EGGCT were eligible for the study if they had pretreatment histological assessment of at least one testicle, and had been treated from 1980 to 1997. As a rule the diagnosis of EGGCT was based on the histological and/or biochemical demonstration of a germ-cell malignancy in patients with midline tumours who were without clinical or ultrasonographic evidence of an infiltrative TC. In some patients a definite histological diagnosis of EGGCT was not possible and the diagnosis of EGGCT was finally based on the typical clinical presentation of a midline tumour and highly elevated serum tumour markers (human choriogonadotrophin and alpha-feto protein). The diagnosis of EGGCT was retained even though a pretreatment orchiectomy specimen of a CIS-affected testicle displayed a scar that was histologically suspicious of regressed TC [5, 6].

Based on the predominant clinical presentation, patients were grouped as those with a retroperitoneal EGGCT and those with a mediastinal EGGCT. The former category also contained patients with both retroperitoneal and mediastinal presentation. The latter group included patients with a supraclavicular or an anterior mediastinal mass without retroperitoneal manifestations. Histology discriminated between a pure seminoma and non-seminoma, the latter being further sub-grouped based on the Pugh classification [16].

As most patients presented with locally advanced and/or metastatic disease, cisplatin-based chemotherapy was the primary treatment followed by resection of residual tumour masses. Most patients received the so-called BEP combination, containing cisplatin 100 mg/m2, etoposide 500 mg/m2 and bleomycin (90 IU) per cycle given at 3-week intervals. A minority of patients also had ifosfamide.

Diagnosis of testicular CIS
All patients from the NRH with EGGCT underwent testicular ultrasonography before the start of any therapy. Only in the case of suspicious ultrasonographic or clinical findings did the patient undergo a surgical testicular biopsy, performed by an inguinal incision (‘open surgical biopsy’), followed by unilateral orchiectomy in case of infiltrative TC. From 1988 onwards performance of bilateral testicular biopsies was included in the routine pretreatment diagnostic armamentarium of all patients with EGGCT: transscrotal testicular biopsies (‘core biopsies’) were preferably obtained by an automatic high-speed 18-gauge core biopsy technique [17, 18]. Only exceptionally were open surgical biopsies and even orchiectomies performed as part of the diagnostic procedures, in particular at community hospitals before referral to the NRH. As a rule both testicles were to be biopsied, but open surgical biopsies were occasionally restricted to one—often clinically atrophic—testicle. At the Danish National Hospital, bilateral transscrotal surgical biopsy has been part of the pretreatment routine examination in patients with EGGCT since 1980 [12,19]. All biopsies were fixed in Bouin’s fixative.

For the purpose of the present study the histological material of the pretreatment testicular biopsies was reviewed by three of the authors (A.E.S., J.W.O., N.E.S.), who were blinded for the clinical observations of the patient follow-up. The pathologists recorded the presence or absence of testicular CIS in the haematoxylin and eosin-stained sections. The presence of a scar with or without CIS was evaluated only in biopsies taken from patients treated at the NRH. One pathologist (J.W.O.) reviewed all histological specimens during sessions with either A.E.S. or N.E.S.

Follow-up
After completion of treatment, Norwegian patients were followed up at the NRH’s out-patient department for 5 years at 3–6 month intervals, thereafter annually at their community hospitals. At the follow-up visits particular awareness was paid to testicular abnormalities. Annual testicular ultrasonography was performed. The suspicion of a new TC led to performance of an open biopsy and eventual orchiectomy. Post-chemotherapy follow-up biopsies were performed in patients with initially proven testicular CIS. The principal post-chemotherapy strategy in Danish patients was to remove the CIS-affected testicle after completion of chemotherapy, even though no invasive tumour was suspected [13]. A minority of Danish patients were, however, followed without such ‘prophylactic’ post-chemotherapy orchiectomy. These patients had annual testicular ultrasonography and biopsies every second year. None of the included patients with EGGCT had testicular irradiation of the CIS-affected testicle. All patients were followed up until death or to 30 June 2001.

Statistics
The PC-based SPSS programme (Version 10.1) was used for all analyses applying descriptive parametric or non-parametric tests, as appropriate. The overall survival and the survival without an invasive TC was displayed by the Kaplan–Meier procedure, in the latter case censoring patients at death or at the time of orchiectomy without demonstrating an invasive TC. Differences between survival curves were evaluated by the log-rank test. A P value of <0.05 was considered as statistically significant.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients
Out of a total of 91 patients with EGGCT a total of 68 men were identified with a pretreatment testicular biopsy. Fifty-three had a retroperitoneal EGGCT and 15 a mediastinal tumour (Table 1), the latter without any retroperitoneal manifestations. The histological diagnosis of EGGCT was a pure seminoma in 24 patients and non-seminoma in 20. The type of the malignant tumour remained unclassifiable in 24 patients with increased serum tumour markers. In these patients no biopsy was performed or the biopsy was described as ‘malignant tumour, not further classifiable’. All patients had cisplatin-based chemotherapy followed by surgery of residual masses.


View this table:
[in this window]
[in a new window]
 
Table 1. Demographics
 
Testicular histology
A total of 55 patients had histological evaluation of both testicles and 13 from only one side. CIS was found in 21 of the 68 patients (31%) [18 retroperitoneal EGGCT (34%); three mediastinal EGGCT (20%)] (Table 2). Unilateral CIS was demonstrated in 19 patients, and two patients had bilateral CIS (one within the retroperitoneal and one within the mediastinal group). In four of 11 evaluable patients CIS was associated with a scar (three retroperitoneal patients; one mediastinal, who also had contralateral CIS without a scar). Six biopsies from 30 evaluable patients without CIS, all of them with retroperitoneal EGGCTs, displayed a scar, most often associated with atrophic changes at the surrounding testicular tissue.


View this table:
[in this window]
[in a new window]
 
Table 2. Site of extragonadal germ-cell tumours and demonstration of testicular carcinoma in situ (CIS). Numbers in parenthesis are number of patients with subsequent invasive testicular cancer
 
A total of 123 testicles were histologically assessed before treatment (Table 3). No statistical significance as to the presence of CIS was observed comparing the types of biopsies with each other: CIS was demonstrated in seven of 48 testicles undergoing ultrasound-guided core biopsies and in 10 of 56 testicles examined in a surgical biopsy (Table 3). However, CIS positivity was significantly more often observed in the pre-chemotherapy orchiectomy specimens than in the biopsies (P = 0.04).


View this table:
[in this window]
[in a new window]
 
Table 3. Type of biopsy and testicular carcinoma in situ (CIS)
 
In five of the 21 patients the diagnosis of testicular CIS had been established in an orchiectomy specimen before systemic treatment. No CIS was found in the contralateral testicle in four of these patients, whereas the fifth unilaterally orchiectomised patient was diagnosed with bilateral CIS. Seventeen men thus retained at least one testicle with pretreatment-demonstrated testicular CIS and were followed up for the development of invasive TC.

Outcome
At the date of last observation 14 patients were dead, 12 of them due to EGGCT, and 55 patients were alive, all but one without evidence of EGGCT (median observation time 119 months). The 10-year actuarial overall survival rate was 82% [95% confidence interval (CI) 73% to 91%, Figure 1]. Five of the 68 patients subsequently had the diagnosis of an invasive TC (four retroperitoneal EGGCT, one mediastinal EGGCT), two of them with regional lymph node metastases at the time of diagnosis of their invasive TC (Table 4). Four of the five patients had a pretreatment CIS-positive testicular biopsy (Table 4) and all metachronous TCs developed in the CIS-affected testicle. In the fifth patient with a retroperitoneal EGGCT a mature testicular teratoma was diagnosed 38 months after the CIS-negative pretreatment core biopsy, showing normal testicular tissue. In four patients the CIS-involved testicle was removed immediately after chemotherapy according to the principal strategy at the National Hospital, Copenhagen. Histology revealed ‘Sertoli cells only’ (three patients) or high-grade atrophy (one patient), but no scar suspicious of a burned-out TC. No CIS was demonstrated in repeated post-chemotherapy biopsies in eight of the other patients with an initial CIS.



View larger version (20K):
[in this window]
[in a new window]
 
Figure 1. Overall survival of 68 patients with extragonadal germ-cell tumour (EGGCT) and 10-year survival rate (95% confidence interval).

 

View this table:
[in this window]
[in a new window]
 
Table 4. Clinical course in 17 patients retaining one or more carcinoma in situ-involved testicles during chemotherapy
 
Figure 2 displays the Kaplan–Meier plot for the invasive tumour-free survival in all 68 patients and separately for patients with histological proof of testicular CIS (21 patients) and those in whom the pretreatment biopsy did not demonstrate testicular CIS (47 patients) (Figure 2B). The 10-year invasive TC-free survival rate for all patients was 88% (95% CI 78% to 98%). Patients with proven CIS had a significantly higher risk of developing a metachronous invasive TC than those with a CIS-negative pretreatment biopsy, the 10-year survival rates being, respectively, 65% (95% CI 36% to 94%) and 97% (95% CI 92% to 100%) (P = 0.007).



View larger version (19K):
[in this window]
[in a new window]
 
Figure 2. Invasive testicular cancer (TC)-free survival and 10-year survival rates (95% confidence intervals). (A) All patients; (B) patients with (1; n = 21) or without (2; n = 47) pretreatment testicular carcinoma in situ. EGGCT, extragonadal germ-cell tumour.

 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Testicular CIS was demonstrated in 18 of 53 men (34%) with a retroperitoneal EGGCT and in three of 15 men (20%) with a mediastinal EGGCT. Five men subsequently had a diagnosis of an invasive TC. The 10-year survival rate of freedom from invasive TC was 88% for all 68 patients, but only 65% in patients with locally untreated testicular CIS. The 10-year overall survival rate in all 68 patients was 82%. Biopsy material obtained by the automatic high-speed transscrotal core biopsy method revealed testicular CIS as often as biopsies achieved by a transscrotal surgical biopsy.

In 1987 Daugaard et al. [11] reported the concomitant existence of EGGCT and testicular CIS in eight of 12 biopsied patients with retroperitoneal EGGCT. Three of the eight patients subsequently had an invasive TC. No CIS was found in three patients with mediastinal EGGCT. In a subsequent expanded series the Danish group reported testicular CIS in 42% of patients with retroperitoneal EGGCT [12], but no CIS in eight patients with mediastinal EGGCT. Thereafter other investigators have published case reports and small series showing an increased prevalence of testicular CIS in patients with EGGCT, and the development of invasive TC in patients cured of EGGCT [2022]. Our expanded analysis confirms the Danish observations of CIS in about one-third of the patients with EGGCT and emphasises the clinically important risk of post-chemotherapy metachronous invasive TC in these patients, in particular if testicular CIS is demonstrated initially.

As also observed for unilateral TC [23], today’s chemotherapy of the germ-cell malignancy seems to delay the risk of metachronous TC in patients with EGGCT, but does not completely abolish this risk [9, 13, 14]. Persistent conservation of a CIS-affected testicle is only possible in some cases. On the other hand, our observations of delayed invasive metachronous TC, post-chemotherapy findings of orchiectomy specimens of initially CIS-affected testicles [24] and recent immunohistochemical studies in testicular tissue [25] underscore that the effect of chemotherapy on CIS is unpredictable.

Our clinical diagnosis of EGGCT can be debated on the background of a histology suspicious of a ‘burned-out’ TC leading to a scar in some patients with or without residual CIS. The histopathological features of such scars are areas of fibrous tissue with pasting borders relative to surrounding testicular parenchyma. Calcifications are frequent, leading to microlithiasis visible by ultrasonography [26]. It can be claimed that a ‘true’ EGGCT requires the absence of the above typical histological features. As a typical testicular scar was seen in six of 30 pre-chemotherapy histological specimens without CIS, all of them from patients with retroperitoneal tumours, and in four of 11 specimens with CIS, one from a mediastinal EGGCT, the above hypothesis cannot be opposed. Some cases of malignant germ-cell tumour clinically presenting as EGGCT may indeed represent metastases from a burned-out primary TC. This might be relevant for retroperitoneal EGGCTs in particular, if the tumour lesions are predominantly located laterally to the midline (such analysis was not possible in our series). Mediastinal EGGCTs without retroperitoneal tumour manifestations should probably be viewed as primary tumours. In the Danish series testicular CIS was never observed in patients with mediastinal EGGCT [12], whereas testicular CIS has subsequently been described in several patients with a mediastinal EGGCT [21, 22]. We report three such cases, two of them without a scar typical for a burned-out TC, the third presenting with bilateral CIS.

Besides the possibility that a scar from a burned-out TC may remain undetected in small biopsies, an alternative explanation for the association between testicular CIS and EGGCT may be that testicular CIS in some patients does not develop into an invasive lesion, although some tumour cells may give rise to metastases. Multiplicity of the carcinogenic process with primary tumour development in the testicle(s), in the retroperitoneum and the mediastinum would represent a third explanation of our findings. This latter view might be supported by ploidy investigations [27]: mediastinal EGGCTs are often near-diploid or near-tetraploid, whereas testicular germ-cell malignancies are usually peri-triploid. However, EGGCTs associated with testicular CIS have never been investigated in this context. It is expected that molecular biological investigations, planned for the current histological samples, will increase our understanding of the tumour biology of EGGCT and the significance of testicular CIS in these cases.

The question remains whether all patients fulfilling our criteria for EGGCT should undergo bilateral testicular biopsy. Of our 17 patients with testicular CIS receiving chemotherapy without prior orchiectomy, four men were subsequently diagnosed with an infiltrative TC. The prognosis of patients with metachronous TC was excellent: no patients have died. The overall TC-free 10-year survival rate for all 68 patients was as high as 88%, in agreement with the international series [9], although our series probably represents a favourably biased cohort: patients whose clinical condition allowed a few days’ treatment delay for performance of the testicular biopsy. Lastly, as also shown in the present series, a CIS-negative pretreatment biopsy does not completely exclude the subsequent development of an invasive TC. This may indicate that the routine performance of biopsies is less justifiable. However, the demonstration of testicular CIS has consequences as to local treatment (radiotherapy or orchiectomy) and follow-up of the individual patient, compared with the histologically proven absence of testicular CIS. Our data clearly indicate that the demonstration of CIS reflects an increased 10-year risk for subsequent tumour development even after chemotherapy for EGGCT. Performing pre-chemotherapy testicular biopsies in patients with EGGCT offers the possibility of individualised counselling and patient management. Patients with EGGCT and testicular CIS and possibly those with retroperitoneal tumour, even without CIS, should be informed about their risk of metachronous TC and necessary close follow-up routines. Fertility issues and sperm-banking possibilities can be discussed with the individual patient, emphasising that post-treatment fathership is possible, even in patients with testicular CIS and in those at risk of a subsequent TC [28]. Based on these considerations our group favours the view of routinely performing pretreatment testicular biopsies in patients with EGGCT. These biopsies can be done as transscrotal core biopsies or transscrotal surgical biopsies with similar accuracy as to the demonstration of CIS, as also shown for patients with unilateral TC [29]. Orchiectomy without preceding testicular biopsy should not be used as a diagnostic tool, even though ultrasonographic findings may indicate some pathology (tiny hyperdense areas or microlithiasis).

In conclusion, in ~30% of the patients with EGGCT, testicular histology displays CIS, mainly in those patients presenting with a retroperitoneal disease. Patients with this condition have a 10-year invasive TC-free survival rate of 65%. Chemotherapy for EGGCT conserves a CIS-affected testis in some patients for many years, but very careful follow-up is needed for these patients. As a rule we recommend performance of testicular biopsies in patients with EGGCT.


    Acknowledgements
 
The authors thank the Norwegian Cancer Society and the Health and Rehabilitation Foundation for financial support of this study.


    Footnotes
 
+ Correspondence to: Dr S. D. Fosså, The Norwegian Radium Hospital, Montebello, 0310 Oslo, Norway. Tel: +47-22-93-40-00; Fax: +47-22-93-45-53; E-mail: s.d.fossa{at}klinmed.uio.no Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Collins DH, Pugh RCB. Classification and frequency of testicular cancer. Br J Urol 1964; 36: 1–11.

2. Dueland S, Stenwig AE, Heilo A et al. Treatment and outcome of patients with extragonadal germ-cell tumours—the Norwegian Radium Hospital’s experience 1979–1994. Br J Cancer 1998; 77: 329–335.[ISI][Medline]

3. Chaganti RSK, Rodriguez E, Mathew S. Origin of adult male mediastinal germ-cell tumours. Lancet 1994; 343: 1130–1132.[ISI][Medline]

4. de Bruin TWA, Slater RM, Defferrari R et al. Isochromosome (12p) positive pineal germ-cell tumor. Cancer Res 1994; 54: 1542–1544.[Abstract]

5. Shawker TH, Javadpour N, O’Leary T et al. Ultrasonographic detection of ‘burned-out’ primary testicular germ-cell tumours in clinically normal testis. J Ultrasound Med 1983; 2: 477–479.[ISI][Medline]

6. Comiter CV, Renshaw AA, Benson CB, Loughlin KR. Burned-out primary testicular cancer: sonographic and pathological characteristics. J Urol 1996; 156: 85–88.[ISI][Medline]

7. Böhle A, Studer UE, Sonntag RW et al. Primary or secondary extragonadal germ-cell tumours? J Urol 1986; 135: 933–943.

8. Bokemeyer C, Nichols CR, Droz JP et al. Extragonadal germ-cell tumours of the mediastinum and retroperitoneum: results from an international analysis. J Clin Oncol 2002; 20: 1864–1873.[Abstract/Free Full Text]

9. Hartmann JT, Fosså SD, Nichols CR et al. Incidence of metachronous testicular cancer in patients with extragonadal germ-cell tumours. J Natl Cancer Inst 2001; 93: 1733–1738.[Abstract/Free Full Text]

10. von der Maase H, Rørth M, Walbom-Jørgensen S et al. Carcinoma in situ of contralateral testis in patients with testicular germ-cell cancer. Study of 27 cases in 500 patients. Br Med J 1986; 293: 1398–1401.[ISI][Medline]

11. Daugaard G, von der Maase H, Olsen J et al. Carcinoma-in-situ of testis in patients with assumed extragonadal germ-cell tumours. Lancet 1987; 2: 528–530.

12. Daugaard G, Rorth M, von der Maase H, Skakkebaek NE. Management of extragonadal germ-cell tumours and the significance of bilateral testicular biopsies. Ann Oncol 1992; 3: 283–289.[Abstract]

13. Christensen TB, Daugaard G, Geertsen PF, von der Maase H. Effect of chemotherapy on carcinoma in situ of the testis. Ann Oncol 1998; 9: 657–660.[Abstract]

14. Dieckmann KP, Loy V. Intratesticular effects of cisplatin-based chemotherapy. Eur Urol 1995; 28: 25–30.[ISI][Medline]

15. Meidahl Pettersen P, Giwercman A, Daugaard G et al. Effect of graded testicular doses of radiotherapy in patients treated for carcinoma-in-situ in the testis. J Clin Oncol 2002; 20: 1537–1543.[Abstract/Free Full Text]

16. Collins DH, Pugh RCB. The pathology of testicular tumours. Br J Urol 1964; 36: 1–112.

17. Parker SH, Hopper KD, Yakes WF et al. Image-directed percutaneous biopsies with a biopsy gun. Radiology 1989; 171: 663–669.[Abstract]

18. Heikkilä R, Heilo A, Stenwig AE, Fosså SD. Testicular ultrasonography and 18G biopsy for clinically undetected cancer or carcinoma in situ in patients with germ-cell tumours. Br J Urol 1993; 71: 214–216.[ISI][Medline]

19. Rorth M Rajpert-De Meyts E, Andersson L et al. Carcinoma in situ in the testis. Scand J Urol Nephrol Suppl 2000; 205: 166–186.[Medline]

20. von Ostau C, Krege S, Hartmann M, Rübben H. Metachronous contralateral germ-cell tumour 7 years after management of testicular intraepithelial neoplasia by chemotherapy and multiple control biopsies. Scand J Urol Nephrol 2001; 35: 430–431.[CrossRef][ISI][Medline]

21. Hailemariam S, Engeler DS, Bannwart F, Amin MB. Primary mediastinal germ-cell tumour with intratubular germ-cell neoplasia of the testis—further support for germ-cell origin of these tumours. Cancer 1997; 79: 1031–1036.[CrossRef][ISI][Medline]

22. Lee KC. Primary mediastinal germ-cell tumour with intratubular germ-cell neoplasia of the testis—further support for germ-cell origin of these tumours. Cancer 1997; 80: 1007–1008.[CrossRef][ISI][Medline]

23. Christensen TB, Daugaard G, Geertsen PF et al. Effect of chemotherapy on carcinoma in situ of the testis. Ann Oncol 1998; 9: 657–660.[Abstract]

24. Simmonds PD, Mead GM, Lee AH et al. Orchiectomy after chemotherapy in patients with metastatic testicular cancer. Is it indicated? Cancer 1995; 75: 1018–1024.[ISI][Medline]

25. Bart J, Groen HJM, van der Graaf WTA et al. An oncological view on the blood–testis barrier. Lancet Oncol 2002; 3: 357–363.[CrossRef][ISI][Medline]

26. Berger A, Brabrand K. Testicular microlithiasis—a possibly premalignant condition. Acta Radiol 1998; 39: 583–586.[ISI][Medline]

27. Oosterhuis JW, Rammeloo RHU, Cornelisse CJ et al. Ploidy of malignant mediastinal germ-cell tumors. Hum Pathol 1990; 21: 728–732.

28. Jacobsen KD, Fosså SD, Bjøro T et al. Gonadal function and fertility in patients with bilateral testicular germ-cell malignancy. Eur Urol 2002; 42: 229–238.[CrossRef][ISI][Medline]

29. Harland SJ, Cook PA, Fosså SD et al. Intratubular germ-cell neoplasia of the contralateral testis in testicular cancer: defining a high-risk group. J Urol 1998; 160: 1353–1357.[ISI][Medline]