Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
In 2015, an estimated number of 8430 new cases of germ cell tumour (GCT) will be diagnosed in the USA.1 Although GCTs show a high sensitivity to cisplatin-based chemotherapy, 10%–15% of patients fail first-line chemotherapy and 3%–5% of all patients with GCT will eventually die of their disease.2 Despite a response rate above 95% to cisplatin-based chemotherapy, the search for new treatment strategies remains worthwhile in accordance to reduce treatment toxicity and offer therapeutic options in non-responding patients.3–5 Various tumours have been described to express L1 cell adhesion molecule (L1-CAM) including lung carcinoma, gliomas, melanoma, renal, ovarian, endometrial and colon carcinoma.6 L1-CAM is associated with tumour cell dissemination via the regulation of prometastatic MMP-2 and MMP-9 in solid and non-solid tumours7 as well as in brain metastases.8 L1-CAM is involved in epithelial-to-mesenchymal transition.9 In various malignancies, there is evidence showing that the expression of L1-CAM is associated with a subset of highly aggressive tumours with adverse clinical outcome and might serve as a therapeutic target.10 We aimed to investigate the expression of L1-CAM in the different GCT subtypes.
Materials and methods
The construction of the tissue micro array (TMA) was described before.9 L1-CAM immunohistochemistry (IHC) was performed using the monoclonal antibody anti-L1-CAM (clone 14.10, directed to the ectodomain, 1:200). The antibody was generated as described previously11 and was tested on a multi-tissue TMA for the appropriate dilution. Peripheral neuronal tissue served as internal positive control for L1-CAM staining. Sertoli cells and Leydig cells were negative for L1-CAM. Two experienced surgical pathologists (PKB, VT) evaluated the L1-CAM stained TMA. Samples were dichotomised into positive and negative. The threshold for positivity was defined at 5% of cells with a moderate or strong staining. The expression patterns were separately analysed for each tumour component, GCNIS and normal tissue. The study was approved by the local ethics committee (reference number KEK StV. 25-2008).
The series included a total of 207 seminomas, 4 spermatocytic tumours, 19 pure embryonal carcinomas, 1 pure mature teratoma and 94 mixed GCT. Mixed GCT included the following components: seminoma, embryonal carcinoma, yolk sac tumour, choriocarcinoma and teratoma (49 with and 45 without a seminomatous component).
The individual tumour components consisted of 253 seminomas, 89 embryonal carcinomas, 52 yolk sac tumours, 53 teratomas, 10 choriocarcinomas. In addition, we included non-tumorous testicular tissue from 20 patients with tumour and GCNIS from 20 patients.
L1-CAM IHC staining showed an exclusively membranous staining pattern in GCTs with moderate-to-strong intensity. In seminoma, IHC staining of L1-CAM showed a predominantly heterogeneous membranous pattern and 95 cases (38%) were L1-CAM positive (see table 1 and figure 1).
The yolk sac tumours with microcystic, glandular, solid and spindle cell growth patterns showed a heterogeneous expression of L1-CAM in 26 cases (50%). The glandular components were positive in a L1-CAM staining, whereas areas with stromal or spindle cell differentiation were negative. The results of L1-CAM staining in teratoma components were very heterogeneous depending on the tissue types found in the teratoma. Ten cases (19%) showed L1-CAM positive structures, for example, glands with a membranous pattern. Stromal components as well as chondrocytes were L1-CAM negative. Five (50%) choriocarcinoma cases expressed L1-CAM. The expression was more restricted to the syncytiotrophoblastic giant cells, whereas the mononuclear component was predominantly negative. Interestingly, syncytiotrophoblastic giant cells were also positive when scattered in a seminomatous tissue component. In embryonal carcinoma, 59 cases (63%) were positive with an intense and homogenous pattern. Normal testicular tissue was L1-CAM negative, whereas IGCNU showed a strong expression of L1-CAM in 45% of all cases.
Recently, L1-CAM has emerged as a potential therapeutic target due to its expression on many solid tumours, and only limited expression on normal tissues.6 In vitro and in vivo studies showed efficacy and safety of L1-CAM targeting antibodies by acting via antibody-dependent cellular cytotoxicity or by being labelled with radionuclides.12–23 In a humanised transgenic mouse model of L1-CAM, no adverse effects were observed after the injection of ant-L1-CAM antibodies.24 A first-in-human phase I trial was published by Park et al20 and showed no adverse effects and some objective response after infusing autologous CE7R/HyTK+ CD8+ cytolytic T lymphocytes. However, it cannot be concluded from this study, if the response was due to the L1-CAM therapy or the subsequent salvage therapies.
We observed that L1-CAM expression is markedly enhanced in most GCTs, in 45% of GCNIS but not in normal tissue. As typical for a retrospective study, this investigation is limited by potential biases, such as patient selection and bias of core punching.
To our knowledge, this is the first publication describing L1-CAM expression in GCTs. We conclude that the frequent expression of L1-CAM in testicular seminomas and non-seminomas indicates that L1-CAM could be a promising new therapeutic target in testicular cancer that warrants further functional studies and potentially investigation in clinical trials in the future. Because L1-CAM was never expressed on normal but frequently on cancer cells, further investigation should elaborate the role for L1-CAM as a new ‘neoplastic’ GCT marker.
We thank Martina Storz and Susanne Dettwiler for excellent technical assistance, and Peter Schraml, Tumor Tissue Bank, University Hospital Zurich, for help with tissue microarray construction.
CDF and PKB contributed equally.
Contributors CDF, PB and VT had the original research idea, led the study and performed all secondary dissections. SS, TH, PA and JB helped develop the research idea, contributed to the study and helped structure the paper. TS and HM provided scientific and technical leadership and support within the dissection laboratory during the data collection period.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.