- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Patients presenting with testicular cancer normally undergo radical orchidectomy as their initial treatment. However, a small group of patients present with advanced, life-threatening disease and require treatment with primary chemotherapy. The need for orchidectomy after the completion of platinum-containing chemotherapy has been investigated in several studies [1–5], with viable tumour and teratoma identified in up to 38% and 42% of post-chemotherapy orchidectomy specimens, respectively. The presence of the blood–testis barrier (BTB) is often cited as the reason for residual tumour due to impaired cytotoxic delivery to the testis [6,7].
This phenomenon was first observed in childhood acute lymphoblastic leukaemia, where isolated testicular relapses were seen in up to 13% of patients after complete remission [8,9]. Similarly, retrospective studies examining treatment failure in elderly men with testicular lymphoma show that the CNS (where the blood–brain barrier exist) and contralateral testis contribute to 40–45% of all relapses after complete remission . With testicular cancer there are patients who achieve a complete response to chemotherapy elsewhere but have persistent viable disease within the testis, supporting the theory of the BTB . Despite this, some cytotoxic drugs must influence the testicular epithelium, as up to 63% of post-chemotherapy orchidectomy specimens in some series contain no viable tumour or teratoma . In addition, the introduction of cisplatin chemotherapy for disseminated non-seminoma germ cell tumours (NSGCTs) led to a three-fold reduction in the rate of second contralateral testicular tumours, suggesting a role in the eradication of carcinoma in situ[12,13]. An alternative explanation given to explain residual disease within the testis in patients achieving a complete response elsewhere, particularly with NSGCT, is tumour heterogeneity. The presence of multiple cell lines within a single tumour as a result of spontaneous mutation, selective clonal metastasis or differentiation of multipotential tumour cells along different lines may explain the differential response to treatment between the primary tumour and metastases [1,14,15].
We sought to examine this issue further by determining the differential response to systemic chemotherapy in patients undergoing simultaneous orchidectomy and retroperitoneal lymph node dissection (RPLND) after treatment for metastatic testicular cancer.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
In all, 42 men undergoing post-chemotherapy simultaneous RPLND and orchidectomy were identified. Patient characteristics are shown in Table 1. The median (range) age at diagnosis was 33 (20–55) years. All patients presented with metastatic disease. Six patients (14.3%) had biopsy confirmed seminoma with normal α-fetoprotein and the remaining 36 (85.6%) had advanced NSGCT. The IGCCC prognostic groups are shown in Table 1. In all, 15 patients (35.7%) were classified as having good prognosis disease, eight (19.0%) intermediate and 18 (42.9%) poor prognosis. One patient presented with only RPLNs which were seminoma on biopsy; however, no baseline tumour markers were available to allow prognostic classification. In all, 16 patients (38.1%) had nodal disease only (including 12 with the RPLNs as the sole site of metastases), 14 patients (33.3%) had RPLNs with pulmonary metastases and 12 (28.6%) RPLNs with non-pulmonary visceral disease, including two patients with brain metastases. All but one patient had an abnormal testis on clinical or ultrasound examination.
Table 1. Patient characteristics
|Median (range) age, years||33 (20–55)|
|IGCCC Prognostic Group, n (%)|| |
| Seminoma:|| |
| Good||6 (14.3)|
| NSGCT:|| |
| Good||9 (21.4)|
| Intermediate||8 (19.0)|
| Poor||18 (42.9)|
| Unknown||1 (2.4)|
|Site of disease, n (%)|| |
| RPLN only||12 (28.6)|
| RPLN and other nodal||4 (9.5)|
| RPLN and lung||14 (33.3)|
| RPLN and non-lung visceral disease||10 (23.8)|
| RPLN, visceral disease and brain||2 (4.8)|
All patients received chemotherapy as shown in Table 2. In all, 36 (85.7%) patients underwent surgery after primary chemotherapy and six (14.3%) after salvage treatment. Most patients received BEP (bleomycin, etoposide and cisplatin) or POMB-ACE (cisplatin, vincristine, methotrexate, and bleomycin alternated with actinomycin D, cyclophosphamide and etoposide) as primary chemotherapy. Three patients received C-BOP-BEP (induction chemotherapy with carboplatin, bleomycin, vincristine and cisplatin followed by BEP) and one OX-BOP-BEP (induction chemotherapy with oxaliplatin, bleomycin, vincristine and cisplatin followed by BEP) in the context of a clinical trial , one (2.4%) patient was treated with etoposide and cisplatin only due to co-morbidities and one (2.4%) received TIP (paclitaxel, ifosfamide and cisplatin). Six patients received salvage chemotherapy: two after BEP [one POMB-ACE and one VIP (etoposide, ifosfamide and cisplatin)] and four after POMB-ACE (two received gemcitabine-TIP and two went on to receive high dose chemotherapy and autologous stem cell transplant).
Table 2. Chemotherapy regimens and indications for surgery
|Chemotherapy regime, n (%)|| |
| Primary|| |
| BEP||19 (45.2)|
| POMB/ACE||11 (26.2)|
| C-BOP-BEP||3 (7.1)|
| Other||3 (7.1)|
| Salvage|| |
| BEP + other||2 (4.8)|
| POMB/ACE + other||2 (4.8)|
| POMB/ACE + high dose||2 (4.8)|
|Median (range; IQR) days between chemotherapy and surgery||51 (14–742; 31–88)|
|Indication for RPLND, n (%)|| |
| Residual mass after chemotherapy||33 (78.6)|
| Growing teratoma syndrome||7 (16.7)|
| Progressive disease on chemotherapy||1 (2.4)|
| Unfit for further chemotherapy||1 (2.4)|
Surgery was performed a median (range; interquartile range) of 51 (14–742; 31–88) days after the completion of chemotherapy. One patient had a significant delay in their surgery (742 days). The reason for this was multiple medical co-morbidities including thrombo-embolic disease. The indications for post-chemotherapy RPLND are shown in Table 2. RPLND was performed in most patients due to residual retroperitoneal mass after chemotherapy (33 patients, 78.6%). Growing teratoma was suspected in seven patients (16.7%). One patient proceeded to surgery as they were unfit for further chemotherapy after two cycles of BEP and one patient was operated on due to progressive disease despite POMB/ACE and gemcitabine-TIP chemotherapy.
Tumour markers were normal before surgery in 34 patients (81.0%) and persistently elevated in six (14.3%). In the patients with elevated markers; three had falling markers before surgery which normalised postoperatively, one was suspected of having growing teratoma syndrome, one unfit for further chemotherapy and another had progressive disease despite two chemotherapy regimens. Tumour markers before surgery were unavailable in two patients (4.8%), both had elevated hCG before chemotherapy.
Pathology findings are summarised in Table 3. Necrosis, teratoma and cancer were identified in 25 (59.5%), 14 (33.3%) and three (7.1%) of RPLN specimens and 15 (35.7%), 15 (35.7%) and 12 (28.6%) of orchidectomy specimens, respectively.
Table 3. Pathological findings in testis and RPLN specimens
|RPLN, n (%)||Testis, n (%)||Total|
|Total||12 (28.6)||15 (35.7)||15 (35.7)||42 (100)|
In all, 12 orchidectomy specimens contained viable cancer, eight (66.7%) of the corresponding RPLN contained necrotic or scar tissue only, two contained teratoma and two cancer. Of the 25 patients with necrotic RPLNs, 12 (48.0%) had active disease within the orchidectomy specimen (eight invasive cancer and four mature teratoma). In 15 of 17 (88.2%) patients with active disease within the RPLN specimen also had active disease within the orchidectomy specimen. Only two patients had viable disease within the RPLN specimen but necrosis within the orchidectomy specimen. The overall histological discordance rate was 38.1% (16 patients). Findings in the orchidectomy specimens were more aggressive than those in the RPLN specimens (i.e. cancer worse than teratoma, which is worse than necrosis) in 33.3% and the same in 61.9%. In the patients with corresponding pathology 13 (50.0%) contained necrosis/scar tissue, 11 (42.3%) teratoma, and two (7.7%) viable cancer.
Patient outcomes are summarised in Table 4. Patients were followed for a median (range) of 49 (0.5–143) months after surgery. In all, 24 patients (57.1%) remained alive and free from disease a median (range) of 57 (0.5–143) months after surgery. In all, 14 patients relapsed after surgery: nine (21.4%) are currently alive and disease free after further salvage chemotherapy a median (range) of 36 (8–132) months after surgery. Five (12.0%) patients died from progressive disease at a median (range) of 18 (8–25) months after surgery. The outcome of four overseas patients is unknown.
Table 4. Patient outcome by pathological findings
| ||Overall||Alive and relapse free||Relapsed or died from disease||Unknown|| P * |
|Overall, n (%)||4249 (1–143)||24 (57.1)||14 (33.3)||4 (9.5)||NA|
|Median (range) survival, months||57 (1–143)||42 (8–132)|
|Testis, n (%)|| || || || ||0.639|
| Viable cancer||12 (28.6)||7||4||1|| |
| Teratoma||15 (35.7)||10||4||1|| |
| Necrosis||15 (35.7)||7||6||2|| |
|RPLND, n (%)|| || || || ||0.556|
| Viable cancer||3 (7.1)||2||1||0|
| Teratoma||14 (33.3)||9||2||2|
| Necrosis||25 (59.5)||13||6||2|
For the patients with known outcome and residual disease within the testis, seven of 11 with viable cancer and 10 of 14 with teratoma remained disease free. Four patients with viable cancer relapsed, two were salvaged with further chemotherapy and two died from progressive disease despite further treatment. Four patients with residual teratoma relapsed, three within the lung and one in the CNS. All were treated with a combination of further chemotherapy and surgery, two remain disease free and two died from progressive disease.
Of the three patients with residual cancer within the RPLN specimen, two remain disease free and one was salvaged with further chemotherapy. Of the 12 patients with known outcome and residual RPLN teratoma, nine are free from disease, two relapsed and were salvaged with further chemotherapy and one died from progressive disease.
Using Pearson's chi squared independence test between pathological subtype and survival outcome, there was no significant association between pathological findings in either orchidectomy or RPLN specimens and outcome (P = 0.639 and P = 0.556 respectively).
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The most striking finding of the present study was that almost half of the patients (48%) with necrotic RPLN tissue had viable disease (either cancer or teratoma) within the testis. The vast majority of patients with testicular cancer undergo orchidectomy at initial presentation and it is accepted that completion orchidectomy is required after chemotherapy in patients presenting with disseminated disease. The effect of systemic chemotherapy on the primary tumour has been examined in several published series with rates of residual cancer of 8–37% and teratoma of 25–50% [3–5,11,17]. The present findings of viable tumour (28%), teratoma (36%) and necrosis (36%) in post-chemotherapy orchidectomy specimens are similar to those found in the largest published series by Leibovitch et al. . They examined 160 NSGCT post-chemotherapy orchidectomy specimens and found viable cancer, teratoma and necrosis in 25%, 31% and 44%, respectively. Whilst it is clear that residual disease persists within the testis after chemotherapy, it is not possible to determine whether this represents a true differential response to treatment, either due to the BTB or tumour heterogeneity, or simply represents incomplete response to systemic therapy.
Compared with the many studies examining post-chemotherapy orchidectomy specimens there are very few published reports of comparative histopathology in patients undergoing simultaneous orchidectomy and resection of extra-gonadal masses. A series by Simmonds et al.  identified 11 patients undergoing simultaneous resection of extra-gonadal masses and found similar pathology in nine of 11 cases. Conversely, in the study by Leibovitch et al. , which examined 160 paired post-chemotherapy orchidectomy and RPLN samples (although it is not stated whether the procedures were performed simultaneously) there was concordance in less than half of cases. In the present study, 16 cases (38.1%) had discordant pathology; most (14, 87.5%) had more aggressive pathology (cancer > teratoma > necrosis) within the testis: 12 with active disease (eight cancer, four teratoma) with associated necrotic RPLNs and two with cancer in the testis and teratoma within the RPLNs.
Published series show that in patients with necrotic RPLNs the incidence of viable disease appears to be higher in the testis than at other metastatic sites [3,18,19]. Almost half of the patients in the present study with necrotic RPLNs had viable disease within the testis; these findings are consistent with other published series . In comparison, amongst 59 patients with NSGCT undergoing RPLND and hepatic resection, liver histology was worse in only 3% of patients . Similarly in a large study of patients undergoing pulmonary resection for NSGCT, 89% of patients with necrotic RPLNs also had necrotic lung pathology . The higher rates of viable disease within the testis suggest that it is the BTB impairing chemotherapy delivery to the primary tumour rather than a differential response to chemotherapy due to tumour heterogeneity.
The normal function of the BTB is to shield developing germ cells from agents that are harmful to the meiotic process and to prevent the immune system recognising expressed proteins as foreign antigens. The BTB is composed of three parts; the physiochemical barrier, an efflux-pump system and an immunological barrier. It is the effectiveness of this barrier, especially the physiochemical and efflux pump, which impedes chemotherapy delivery. Sertoli cells are connected together with tight junctions, which form a physical barrier preventing entry of large molecules; this is further strengthened by tight junctions between the capillary endothelial cells and the surrounding myoid cell layer [6,20,21].
Despite this tight physical barrier, smaller or lipophilic molecules may enter the testicular tissue by diffusion across cellular membranes. However, although many cytotoxic drugs are small and lipophilic, concentrations in mice testis are lower than found in other tissues . Efflux pumps limit testis tissue concentrations by extruding substance back into the blood. ATP-binding cassette transporters, which mediate resistance to anticancer drugs, are expressed on the BTB. These include p-glycoprotein, which is expressed on the luminal side of capillary endothelial cells in the testis and brain compared with the basolateral side in the liver, kidney and adrenal gland [23,24]. Animal models have confirmed that knock-out mice lacking p-glycoprotein show increased testis concentrations of cytotoxic agents, e.g. vinblastine , and higher concentration of radioactive cyclosporine compared with wild-type mice . The complementary action of the efflux pumps and physiochemical barrier restricts chemotherapy delivery to the testis resulting in persistent viable disease.
No such barrier exists within the systemic circulation, which may explain the higher incidence of necrosis within RPLN compared with orchidectomy specimens in the present series (60% vs 36%). RPLND was performed in most patients (95%) because of a residual mass or suspicion of growing teratoma syndrome. Residual post-chemotherapy RPLN masses in patients with NSGCT are routinely resected [26–28]. Imaging methods are currently unable to differentiate between necrosis and residual teratoma. Whilst teratoma is considered benign the natural history is unpredictable, it is resistant to chemotherapy, and has the potential for malignant degeneration or the development of growing teratoma syndrome [29–31]. Teratoma was found in 36% and 33% of orchidectomy and RPLN specimens respectively, and in both in 26% in the present series, confirming the need for the combined procedure. RPLND can also make a major contribution to curing those with drug-resistant viable cancer. Of the three patients in the present series with viable cancer in the RPLNs, two did not require further chemotherapy and remain disease free at 11.9 and 6.6 years after surgery.
Completion RPLND is not routinely performed if imaging confirms a complete response. A recent study has confirmed that observation of RPLNs in patients with NSCGT is acceptable in those with complete remission on imaging . In patients with seminoma most centres limit surgery to patients with residual masses of >3 cm or in those with positron emission tomography positive disease opting for observation in the reminder of cases [32–34]. The present series included six patients with pure seminoma who all underwent RPLND for residual masses after chemotherapy. Two of these patients were operated on in the late 1990s before the introduction of surveillance policies at our institutions, one patient had an intra-abdominal testis requiring laparotomy for removal and the remaining three patients had large (>3 cm) residual masses after chemotherapy. Only one of these six RPLN specimens contained viable seminoma, whereas seminoma was present in four orchidectomy specimens.
In conclusion, there is a high risk of finding active tumour or teratoma in the testis of patients with metastatic testicular GCTs treated with primary chemotherapy. Thus, patients treated with primary chemotherapy who have had a suspicious lesion identified in a testis at any time, should undergo orchidectomy of that testis to complete treatment. This remains the case even when the RPLN specimen shows no viable cancer or indeed when there is no evidence of disease on imaging and no RPLND is performed. In selected cases, particularly if only one testis is present and a discrete abnormality is visible on ultrasound, partial orchidectomy can be considered .