Trophoblast Cell Surface Antigen 2 (Trop-2) Protein is Highly Expressed in Salivary Gland Carcinomas and Represents a Potential Therapeutic Target

Salivary gland carcinomas (SGC) consist of a heterogeneous group of more than a dozen tumors. Although each entity displays recurrent genomic alterations, significant heterogeneity can be found within each group [1, 2]. The majority of early stage tumors (T1-2) can be treated by tumor resection. Additionally, performance of a neck dissection is recommended in high grade carcinomas even in the absence of clinical or radiological evidence of locoregional metastatic nodal spread and in low-grade tumors with suspicion of (loco-regional) nodal involvement [3]. Generally, in advanced-stage tumors (T3-4), high-grade tumors, positive or close margins or in the presence of lymph node metastasis, vascular and perineural invasion, adjuvant radiation therapy is indicated [4, 5]. For adenoid cystic carcinoma and other high-risk pathologies (e.g., high-grade mucoepidermoid carcinoma, salivary duct carcinoma), especially those arising in anatomic sites not amenable to wide resection, such as the nasal cavity and paranasal sinuses, many authors recommend adjuvant radiation therapy to increase local control rates even after complete resection due to the tumors’ aggressive and often perineural growth [4, 6‐10]. However, the role of adjuvant radiation therapy in early-stage tumors without adverse features such as positive margins, perineural invasion and negative neck nodes is contentious, as some authors failed to confirm adjuvant radiation therapy as a prognostic factor [11, 12].

Definitive treatment regimens without surgery consisting of combined chemoradiotherapy with curative intent only come into consideration for patients unable to undergo surgery or for unresectable cancers, such as with skull base involvement [13, 14]. In rare cases, chemotherapy alone can be applied in a neoadjuvant setting to reduce tumor volume for subsequent salvage surgery, but does not play a role in the current standard curative treatment of primary SGC [15‐17]. In case of unresectable, recurrent or metastatic cancer, mainly platinum-based palliative chemotherapy regimens are applied. However, randomized controlled trials are lacking and toxicity is considerable [18]. Currently, more information is gathered about possible molecular targets for a tailored therapy for SGCs [19, 20]. Promising results have been achieved with, for example, Herceptin and NTRK inhibitors in selected patients with primarily unresectable disease [10, 21, 22].

Another interesting target is trophoblast cell surface antigen 2 (Trop-2) which is a transmembrane glycoprotein that is involved in a variety of cell signaling pathways including proliferation, survival, self-renewal, and invasion. Cell signaling mediated by Trop-2 involves the expression of Ki-67, activation of MAPK signaling and cell cycle progression by increasing levels of cyclin D1. Trop-2 expression can induce transcription factor AP-1 leading to upregulation of carcinogenesis associated genes [17, 23]. Expression of Trop-2 is associated with poor outcome, particularly in solid tumors [24]. The potential of Trop-2 as a target for combination therapy using the antibody–drug conjugate sacituzumab govitecan, has already been demonstrated in a number of different tumor entities, such as urothelial carcinoma, squamous cell lung carcinoma and breast cancer [23, 25, 26]. In metastatic triple-negative breast cancer, the FDA has already approved this antibody–drug conjugate [26]. Studies indicate that sacituzumab govitecan mediates anti-tumor responses with varying expression of the Trop-2 protein [27, 28]. Until now, the relevance and distribution of Trop-2 protein expression in SGC has not been addressed (Table 1).

The aim of the current study was to test for the extent, distribution and intensity of Trop-2 protein expression in different entities of SGCs using a bimodal approach of immunohistochemistry and matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) imaging.

This is the first study to determine the protein expression of Trop-2 using different techniques in salivary gland carcinomas. Trop-2 is a transmembrane protein with an extracellular domain and its expression is upregulated in tumors cells relative to normal cells [23]. It is involved in several molecular pathways associated with cancer development and has multiple binding partners. Trop-2 expression can increase the expression of Ki-67 causing Ca2 + mobilization from internal stores.

MAPK signaling is activated by Trop-2 leading to increased levels of phosphorylated ERK1 and ERK2 [17]. Activation of ERK can putatively induce proteasomal degradation of FOXO3a and thereby inhibit cell death [31]. Also, MAPK signaling and cell cycle progression are further stimulated by Ca2 + . Trop-2 expression can increase levels of cyclin D1 and cyclin E, helping to mediate ERK1/2 cell cycle progression. Subsequently, ERK signaling induces AP-1 transcription factor which plays a crucial role in the regulation of tumor-associated target genes during carcinogenesis [32]. AP-1 can induce angiogenesis via VEGF (vascular endothelial growth factor), activate cell proliferation via cyclins and CDKs and cause cell invasion and metastasis via matrix metalloproteinases, podoplanin, Ezrin and CD44. Furthermore, AP-1 can induce epithelial to mesenchymal transition via podoplanin allowing for the nuclear translocation of β-catenin, eventually causing cell growth via downstream effectors [17].

Trop-2 overexpression has been described in a variety of malignant tumors, such as in colorectal, esophageal, lung, ovarian or head and neck cancer [17]. Several clinical trials have been investigating Trop-2 as a potential target. PF-06664178, an antibody–drug conjugate delivering an auristatin microtubule inhibitor, was tested in a phase I study in patients with advanced therapy-resistant solid tumors. Eleven out of 31 patients showed stable disease and no complete or partial response was observed [33]. DS-1062a, an antibody–drug conjugate delivering a topoisomerase I inhibitor and derivate of exatecan, is currently being tested for the treatment of unresectable advanced and incurable NSCLC in an ongoing first in human study [23]. Sacituzumab govitecan (or IMMU-132), an antibody–drug conjugate comprising the active metabolite of irinotecan (SN-38) conjugated to an anti-Trop-2 antibody, has been investigated in several trials. In a phase I study in patients with platinum-resistant urothelial carcinoma, it was well tolerated and three out of six patients had a clinically significant response, with a progression free survival of 6.7–8.2 months [34]. Due to these results, a phase II trial has been initiated. For metastatic NSCLC patients, Sacituzumab govitecan was tested in a phase II study with 54 pretreated patients. A clinical benefit rate, defined as complete or partial response or stable disease for more than four months was observed in 43% of the patients. The median response duration to therapy was six months, with a median overall survival of 9.5 months and median progression free survival of 5.2 months [35]. For SCLC patients, Sacituzumab govitecan has also been evaluated in a phase II study. In 60% of the patients tumor shrinkage was observed and a clinical benefit rate was reported for 34% with a median overall survival of 7.5 months and progression free survival of 3.7 months [25]. However, the most promising results have been achieved for triple-negative breast cancer. Bardia et al. have recently investigated the use of sacituzumab govitecan in 108 cases of refractory triple-negative breast cancer. They reported a response rate of 33.3%, including a complete response in 2.8%, with a median duration of response of 7.7 months and a median progression free survival of 5.5 months [26]. These results have led to the approval of sacituzumab govitecan (TRODELVY, Immunomedics, Inc.) by the U.S. Food and Drug Administration for adult patients with metastatic triple-negative breast cancer who received at least two prior therapies for metastatic disease.

In the current study, among the 91.2% of SGCs with Trop-2 expression, 81.6% exhibited a high to moderate protein concentration in immunohistochemistry. More than half of the cohort consisted of adenoid cystic, mucoepidermoid and salivary duct carcinomas. Thus, the most prevalent tumor entities are well represented in the collective. The rates of Trop-2 expression were particularly high in adenoid cystic carcinomas, salivary duct carcinomas, epithelial myoepithelial carcinomas and mucoepidermoid carcinomas. The results of IHC were validated with MALDI-MS Imaging on 10 exemplary samples. Trop-2 protein expression was detected in 80%. There were two discordant findings regarding Trop-2 expression. One sample showed negative Trop-2 expression in IHC but protein expression in MALDI-MS imaging, and another with negative Trop-2 expression in MALDI-MS imaging showed moderate expression in IHC. A possible explanation for this could be the fact that consecutive TMA sections were used. We have also considered the heterogeneity of protein expression within the tumor. For this purpose, we analyzed all tumors with four tumor biopsies from different tumor sites in TMA format and additionally 20 tumors on full tumor sections. Most of the tumors (89.5%) showed a homogeneous expression of Trop-2. Carcinomas with a high Trop-2 pattern were homogeneously positive even on full section, while undifferentiated or dedifferentiated carcinomas displayed a more heterogeneous Trop-2 expression. This underlines the significance of our TMA-based analysis and allows conclusions to be drawn about the potential therapeutic effectiveness of sacituzumab govitecan in SGCs. The more homogeneously a therapy-relevant biomarker is expressed in the tumor, the more likely it is to be effectively targeted.

In comparably designed studies (e.g., Gatsby study), in which a biomarker (Her2/neu)—also determined on the tumor cells—was combined with an antibody–drug conjugate (trastuzumab emtansine) specifically directed against the biomarker, the very heterogeneous distribution of Her2/neu in gastric cancer was a possible relevant reason for the failure of this study [36]. The same drug combination is effective in breast cancer, possibly because (in contrast to gastric cancer) breast cancer usually shows a very homogeneous expression of Her2/neu.

Given the homogeneous expression of Trop-2 in SGC, future clinical studies will have to verify the relationship between Trop-2 protein expression and sacituzumab govitecan. From today's perspective, however, SGCs seem to be good candidates for a Trop-2-targeted therapy.

In our cohort, Trop-2 expression showed no correlation with the presence of nodal metastases, perineural, vascular or lymphatic invasion and 5-year disease free survival. This is in contrast to the findings of a meta-analysis by Ping et al., who reported a significant association of Trop-2 overexpression and short disease free survival in different solid tumors [24]. Therefore, the role of Trop-2 as a prognostic biomarker in SGCs remains inconclusive.

Limitations of our study could be the retrospective fashion of our analyses. The limited number of cases among each individual tumor type make it difficult to establish correlations with clinical parameters or outcome. However, due to the low incidence of certain SGC subtypes, desirable prospective studies with sufficient participants are particularly challenging. Nevertheless, despite these shortcomings we are convinced that the results of this exploratory study can be followed up in clinical studies with Trop-2 as a target in salivary gland carcinomas.

In summary, we were able to demonstrate that several types of salivary gland carcinoma immunohistochemically express Trop-2 with variable intensity. Since there are currently few systemic treatment options for advanced SGCs, Trop-2 represents a promising target for further clinical studies, such as with sacituzumab govitecan.