Somatic Transformation in Metastatic Testicular Germ Cell Tumours – A Different Disease Entity

Discussion

The incidence of somatic transformation in GCT has been described in other centres. Ahmed et al. (8) have reported 17 cases with an established diagnosis of malignant teratoma transformation out of 580 GCT patients treated with chemotherapy. Malignant mesodermal differentiation was the most common, seen in 15/17 patients including rhabdomyosarcoma, leiomyosarcoma and others. Guo et al. (18) have reported 30 confirmed cases of GCT with sarcomatous component from 1985 to 2007, all containing a teratomatous component and a sarcomatous component in the primary (n=19) or metastatic tumours (n=11) or both. The commonest sarcomatous component was rhabdomyosarcoma, found in 79% of cases.

The pathogenesis of somatic transformation remains unclear. It may derive from pluripotent germ cells and malignant transformation of teratomatous elements (19). There are potentially other mechanisms that contribute to the development of a sarcomatoid change specifically within NSGCT (non seminomatous germ cell tumour), and one hypothesis implicates exposure to DNA damaging agents. Other centres have found that the sarcoma element can appear after chemotherapy in metastasis rather than being present in the primary excision (20) and this is certainly evident in our cohort presented above. A recent, elegant, genomic study on chemotherapy resistant and refractory GCT demonstrated that loss of pluripotency markers such as NANOG is strongly associated with chemo resistance and poor outcome (21). Therefore, GCT tumours that have undergone terminal differentiation towards a specific somatic phenotype such as rhabdomyosarcoma may, in keeping with this, be expected to be more chemoresistant.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Survival by histology at post-chemotherapy resection. Comparison between necrosis, mature teratoma, non-sarcomatoid and sarcomatoid variants.

At present, it is unclear to the treating physician whether this histological diagnosis should change the therapeutic algorithm in these patients. We report on the presentation and clinical course of 30 patients with GCT that underwent somatic transformation, and find evidence for a different prognostic significance depending on the timepoint during therapy at which the somatic transformation event is seen. Despite the small number of patients, our data suggest that patients with somatic transformation at presentation appear to have a better outcome than patients who present with somatic transformation at post chemotherapy surgery for residual cancer, or at a post-chemotherapy relapse. In a retrospective study, Malagon et al. (16) have reported the clinicopathological features of 46 patients with GCT that showed additional sarcomatous components involving either the primary site or their metastases. They showed a statistically significant difference in overall survival between patients with sarcomatous containing GCT and age and stage matched control group without sarcoma in the primary site (p<0.001). Guo et al. have also found a higher risk of mortality in patients with sarcoma in any metastasis (18).

Currently, there is no consensus on the treatment of GCTs with a sarcomatous component due to the rarity of this tumour. Cisplatin-based combination chemotherapy remains the backbone of treatment for disseminated NSGCTs achieving excellent anticipated cure rates in good risk patients. However, treatment of platinum refractory disease is challenging and a significant number of patients with intermediate and poor risk NSGCT will still require combination chemotherapy for relapsed disease.

In many centres, ifosfamide and platinum-containing regimens such as TIP, VIP or VeIP (4) are utilised as first line therapy of relapse, with high dose chemotherapy and stem cell rescue held in reserve for third line salvage therapy. However, none of the large prospective randomised studies utilising salvage chemotherapy have reported on whether any of the patients had variant histology. The total numbers are likely therefore to be small.

In the case series presented above, a number of patients who progressed following first-line platinum-based chemotherapy were salvaged utilising ifosfamide-containing regimens such as TIP, G-TIP, cisplatin doxorubicin and ifosfamide. Germ cell tumours are generally highly sensitive to cisplatin-based chemotherapy whilst pure sarcomas are generally resistant to cisplatin. Doxorubicin and ifosfamide have been the backbone of treatment of metastatic sarcoma over the last decade, and gemcitabine-docetaxel combination has recently emerged as a useful alternative in several phase II studies. Hensley et al. have found a 53% response rate including 50% of patients previously treated with doxorubicin (22). Some pre-clinical data suggests that there seems to be a degree of synergistic activity between docetaxel and gemcitabine (23). Donadio et al. (24) administered chemotherapy, based on the specific single cell type observed in the transformed histology, to 12 GCT patients, of which 5 presented with undifferentiated rhabdomyosarcoma, and a partial response was achieved in 7 cases. The choice of salvage chemotherapy should therefore take into consideration the data above indicating that the sarcomatous element is likely to affect outcome in relapsed disease and we would suggest utilising regimens that have documented crossover efficacy in these tumours.

The studies highlighted above (13, 17, 23) indicate that the presence of viable cancer with sarcomatoid change may confer a poorer outcome in the post chemotherapy surgery setting compared to the presence of viable non-transformed cancers. We were able to compare a cohort of patients (n=23) who underwent resection of residual disease post chemotherapy for NSGCT and were found to have viable cancer (but not transformed) with the cohort of patients described above with sarcomatoid change and post chemotherapy resection (n=8). Although there is the caveat of the small number of cases, the data presented in Figure 3 indicate that the presence of residual active cancer post chemotherapy confers a poor prognosis-whether sarcomatoid or not, implying that the failure of first line systemic treatment is in itself a key determinant of survival.

In these cases, it is clear that surgery with the goal of clearance of all sites of residual disease is clearly also of vital importance. Recent literature has highlighted the importance of surgical resection of residual disease in other chemotherapy “resistant” phenotypes of GCT such as late relapse (25) and this must be echoed in patients with variant histology. In our cohort patients, whenever feasible surgery was performed, including retroperitoneal lymph node dissection (RPLND), thoracotomy and craniotomy to achieve resection of residual disease. Radiotherapy was also utilised where surgery was not deemed possible.