SEARCH

SEARCH BY CITATION

Keywords:

  • Testicular neoplasms;
  • pathology;
  • surgical

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objective

To assess the value of central histopathological review of testicular tumours.

Materials and methods

The histopathology reports of the referring hospital and central review were compared for all referred testicular specimens (201 orchidectomies and seven testicular biopsies) from 11 hospitals in one region, between 1992 and 1997.

Results

There were two main findings: First, there was a major discrepancy of tumour type in 12 specimens (6%). The most frequent change (in five cases) was from seminoma to nonseminomatous germ cell tumour (NSGCT). There were also difficulties with uncommon tumours such as spermatocytic seminoma, lymphoma and adenocarcinoma. Treatment was changed as result of the altered diagnosis in eight of these 12 patients. Second, there were discrepancies of interpretation of vascular invasion in 12 of 59 NSGCTs (20%). Differences of tumour elements recognized in NSGCTs were frequent.

Conclusions

The errors discovered on central histopathological review of testicular tumours were significant, though small in number. This supports the idea of central review and is consistent with the specialization recommended in the Calman–Hine cancer arrangements.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The initial treatment for most testicular tumours is orchidectomy; subsequent treatment is determined by the pathology and staging [ 1]. The standard treatment for stage 1 classical seminoma (confined to the testis) is prophylactic radiotherapy to the ipsilateral para-aortic (and sometimes pelvic) lymph nodes. Radiotherapy is also used for low-volume metastatic disease in the abdomen. Patients with bulky metastases or relapse after radiotherapy are given chemotherapy.

Men with stage I nonseminomatous germ cell tumours (NSGCTs) with no vascular invasion (anticipated relapse rate of 15%) are managed with a policy of surveillance in the UK, although many oncologists currently offer limited chemotherapy if vascular invasion is present in the primary tumour (anticipated relapse rate of 40%). Retroperitoneal lymph node dissection is used in the USA and parts of Europe. In Britain, metastatic NSGCT is treated with cisplatin-based combination chemotherapy, although retroperitoneal lymph node dissection is used in the USA to treat low-volume para-aortic disease. Spermatocytic seminoma is treated by orchidectomy alone, epidermoid cysts may be treated by excision of the lesion and testicular lymphoma is treated by anthracycline-based combination chemotherapy. Thus an accurate pathological diagnosis is essential to ensure appropriate treatment.

The aim of this study was to determine whether central review of pathology is worthwhile, by comparing the diagnoses made in the referring hospital with that made on review. It was prompted by our impression that major discrepancies in diagnosis occurred in a significant minority of specimens. A secondary aim was to determine if there was any pattern to these discrepancies, and provide guidance to general pathologists to help them deal with potentially problematic cases.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The Department of Medical Oncology, Southampton University Hospitals provides a regional referral service for the management of testicular tumours. Most testicular specimens for suspected tumour related to this service are sent for histopathological review in Southampton before management decisions are made by an oncologist. Testicular specimens (208, comprising seven biopsies and 201 orchidectomies) reviewed between 1992 and 1997 were included in this study. Altogether, 220 testicular specimens were reviewed in this period. We excluded 12 patients; 10 who underwent biopsy or orchidectomy outside the region and two whose initial reports were well before the study period (in 1977 and 1981). Only occasional biopsies from Portsmouth were reviewed; otherwise, all NSGCTs from referring hospitals were reviewed. In the initial part of the study period only metastatic seminomas were referred and reviewed; later, all seminomas were assessed.

The pathology reports from the referring hospital were compared with the review diagnosis. The following diagnostic groups were used: NSGCT, classical seminoma, spermatocytic seminoma, adenocarcinoma, adenocarcinoma arising in NSGCT, epidermoid cyst, Leydig cell tumour, lymphoma, intratubular germ cell neoplasia, and ‘other’. For NSGCTs, the tumour elements described, i.e. embryonal carcinoma, yolk sac tumour, teratoma (mature and immature were not distinguished), choriocarcinoma and classical seminoma, were compared. Vascular invasion was commented on consistently in reports of NSGCTs from Southampton from the autumn of 1992. Therefore, only comments about vascular invasion in the period 1993–97 were compared.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In two biopsies there was no definite diagnosis in the initial report; of the remaining 206 specimens, there was a major discrepancy between the initial and review diagnosis in 12 (6%). In five patients an initial diagnosis of classical or anaplastic seminoma was changed to NSGCT, with embryonal carcinoma the only tumour type in four and the predominant type in one. As a result of this change in diagnosis, three of these patients were spared radiotherapy and instead were carefully observed, with no evidence of relapse to date. One patient was transferred to our care after chemotherapy at another hospital. A small-volume retroperitoneal mass persisted, and because of a diagnosis of seminoma, had been kept under surveillance. Once the diagnosis of NSGCT became apparent, resection of this mass was arranged. Mature teratoma was removed, confirming the change in diagnosis of the primary tumour. The change in diagnosis did not affect the management of the fifth patient who had metastatic disease and received chemotherapy.

One patient had a contralateral testicular biopsy taken at another hospital; atrophy was diagnosed. However, on review, intratubular germ cell neoplasia was considered the correct diagnosis. Orchidectomy was performed as the testis was not functional; the resected specimen showed extensive intratubular germ cell neoplasia.

In one patient a diagnosis of lymphoma, rather than classical seminoma, resulted in adjuvant combination chemotherapy for stage I disease, as is standard practice. One man with an initial diagnosis of classical seminoma had radiotherapy cancelled after the diagnosis was revised to spermatocytic seminoma.

One patient with an initial diagnosis of NSGCT and a review diagnosis of adenocarcinoma arising in NSGCT received chemotherapy. The change in diagnosis meant that postchemotherapy laparotomy for metastatic disease was performed earlier than it would have been, and that the oncologist was alerted to the poor prognosis. One man with an initial diagnosis of NSGCT received radiotherapy to the groin at another hospital before the diagnosis was revised to adenocarcinoma; subsequent management was not changed. The two remaining cases (spermatocytic seminoma revised to Leydig cell tumour, and adenocarcinoma arising in NSGCT revised to NSGCT) resulted in no change in clinical management.

There were differences in the tumour elements described in 47 NSGCTs. Most of these differences were tumour elements that were described in the review but not in the initial report, particularly yolk sac tumour (24 tumours) and classical seminoma (17 tumours).

Vascular invasion was compared in 109 cases (1993–97) where both the initial and review diagnosis was NSGCT. The initial report commented on the definite presence or absence of vascular invasion in 59, reported possible vascular invasion in two, and made no comment in 48. Discrepancies about the interpretation of vascular invasion were present in 12 of 59 cases (20%). On review, the diagnosis of vascular invasion was changed to no vascular invasion (eight cases) more often than the reverse change (four cases). In seven of the initial reports, unconventional terminology or unclear description made interpretation difficult. For example, describing embryonal carcinoma as adenocarcinoma is potentially confusing.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In this study there were major discrepancies between the initial diagnosis and review diagnosis in 6% of testicular specimens, and discrepancies of interpretation of vascular invasion in 20% of NSGCTs. We assumed that the review diagnosis was correct; this was usually based on haematoxylin and eosin sections, supported in difficult cases by immunohistochemistry. The only similar previous study of which we are aware, by Segelov et al. from Australia [ 2], found a comparable rate of major discrepancies of 11% (10 of 87 testicular tumours). In both studies the discrepancies found were important, as they altered the treatment given. A diagnostic concordance rate of 94% is higher than that for tumours of the breast (92% for the presence of invasive carcinoma, with poor agreement on histological type [ 3]) and lung (72% for small cell carcinoma [ 4]). In seven biopsies the initial report used confusing terminology; it is essential that standard terminology is used in the pathology report.

The commonest change on review was from seminoma to NSGCT, in agreement with the previous study where six of the 10 discrepancies were of this type [ 2]. On review, these areas were interpreted as embryonal carcinoma. The characteristic cytological appearances, and immunohistochemistry with CAM5.2 and antibody to placental alkaline phosphatase, are useful in distinguishing classical seminoma and embryonal carcinoma [ 5]. Confusion between classical seminoma and spermatocytic seminoma, classical seminoma and lymphoma, and spermatocytic seminoma and Leydig cell tumour, are understandable as the latter diagnosis in each of these pairs is uncommon. Again, the cytological appearance of the cells is useful in making these diagnoses and can be supported by immunohistochemistry for placental alkaline phosphatase and lymphoid markers. The characteristic cytological appearance of embryonal carcinoma is useful in distinguishing it from adenocarcinoma. Immunohistochemistry for placental alkaline phosphatase or detection of cytoplasmic glycogen with periodic acid-Schiff stain can help in recognizing intratubular germ cell neoplasia [ 5].

Failure to recognize yolk sac tumour has also been a common finding in other studies [ 2, 6], and is probably a reflection of the histological overlap with embryonal carcinoma and immature teratoma. We base our diagnosis of yolk sac tumour on sections stained with haematoxylin and eosin, and do not find immunohistochemistry for α-fetoprotein useful. Failure to recognize classical seminoma as an element of NSGCTs was more surprising, and may reflect a failure to search for other elements once NSGCT has been diagnosed. A useful clue to microscopic foci of classical seminoma is the associated lymphocytic infiltrate.

There were differences in the interpretation of the presence or absence of vascular invasion in 20% of NSGCTs. Similar discrepancies have been described in testicular tumours [ 6], and in other tumours such as carcinoma of the breast [ 7]. A common problem with germ cell tumours is the smearing of cells across the section, especially with classical seminoma, which can hamper interpretation of vascular invasion. There is an element of subjectivity in the diagnosis of vascular invasion, which even with strict criteria is probably difficult to eliminate. One advantage of central review is that the interpretation should be more consistent. Our interpretation of vascular invasion in stage 1 NSGCT is predictive of relapse in men treated with a policy of surveillance [ 8].

There are two possible responses to the discrepancies shown in this study. The first would be to educate the pathologists in the referring hospitals about the areas of difficulty. The problem with this is that testicular tumours are uncommon (the referring pathologists in this study reported an average of two testicular tumours each per year), and major discrepancies are infrequent (6%). Thus each pathologist makes an error on average only once every 10 years.

The alternative approach is to appreciate the value of central review. In this study, one pathologist (J.M.T.) reviewed all but 14 of the 208 specimens, and reported about 100 specimens from patients operated on in Southampton during the study period, i.e. in total, about 50 specimens per year. By examining this number of testicular tumours, familiarity with the appearances of the different tumour types can be maintained, and one pathologist can achieve a more consistent interpretation of more subjective features such as vascular invasion. Such specialization is consistent with the Calman–Hine cancer arrangements [ 9]. We suggest that central review has an important role in ensuring the accurate diagnosis of testicular tumours and thus ensuring appropriate treatment.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank the pathologists from the following hospitals for their collaboration in this study: North Hampshire Hospital, Basingstoke; Royal Bournemouth Hospital; Princess Elizabeth Hospital, Guernsey; Royal Hospital, Haslar; St Mary’s Hospital, Isle of Wight; General Hospital St Helier, Jersey; Poole General Hospital; Queen Alexandra Hospital, Portsmouth; Salisbury District Hospital; West Dorset Hospital, Dorchester; and Royal Hampshire County Hospital, Winchester.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    O’Callaghan A & Mead GM Progress in the management of solid tumours. Testicular carcinoma. Postgrad Med J 1997; 73: 481 6
  • 2
    Segelov E, Cox KM, McNeil RE, Lancaster L, Rogers J The impact of histological review on clinical management of testicular cancer. Br J Urol 1993; 71: 736 8
  • 3
    Ellman R & Anderson TJUK National Coordinating Group for Breast Screening Pathology; Sloan JP, et al. Consistency of histopathological reporting of breast lesions detected by screening: findings of the UK National External Quality Assessment (EQA) Scheme. Eur J Cancer 1994; 30A: 1414 9
  • 4
    Keehn R, Auerbach O, Nambu S et al. Reproducibility of major diagnoses in a binational study of lung cancer of uranium miners and atomic bomb survivors. Am J Clin Pathol 1994; 101: 478 82
  • 5
    Damjanov ITumors of the testis and epididymis. In Murphy WM, ed, Urological Pathology Chapt 6 Philadelphia: Saunders, 1989; 314 79
  • 6
    Sesterhenn IA, Weiss RB, Mostofi K et al. Prognosis and other clinical correlates of pathologic review in stage I and II testicular carcinoma: a report from the Testicular Cancer Intergroup Study. J Clin Oncol 1992; 10: 69 78
  • 7
    Pinder SE, Ellis IO, Galea M, O’Rouke S, Blamey RW, Elston CW Pathological prognostic factors in breast cancer. III. Vascular invasion: relationship with recurrence and survival in a large study with long-term follow-up. Histopathology 1994; 24: 41 7
  • 8
    Maher TM & Lee AHS Vascular density does not predict future metastatic disease in clinical stage 1 non-seminomatous germ cell tumours of the testis. Histopathology 1998; 32: 217 24
  • 9
    Department of Health Policy Framework for Commissioning Cancer Services; Guidance for Purchasers and Providers of Cancer Services. London: HMSO, 1995