Molecular pathways in vulvar squamous cell carcinoma: implications for target therapeutic strategies

Abstract

Background

Additional prognostic factors and personalized therapeutic alternatives for vulvar squamous cell carcinoma (VSCC), especially for advanced stages with poor prognosis, are urgently needed.

Objectives

To review and assess literature regarding underlying molecular mechanisms of VSCC target therapeutic and prognostic approaches.

Methods

We performed a narrative literature review from the inception of the database up to January 2020 limited to English language, organizing knowledge in five main fields: extracellular and intracellular cell cycle deregulation, tumor immune microenvironment, tumor angiogenesis and hormones.

Results

EGFR immunohistochemical overexpression/gene amplification, representing early events in VSCC carcinogenesis, have been correlated with a worse prognosis and led to inclusion of erlotinib in cancer guidelines. p16 expression and HPV positivity are linked to a better prognosis, while p53 overexpression is linked to a worse prognosis; thus, biomarkers could help tailoring conventional treatment and follow-up. The implications of PD-L1 positivity in reference to HPV status and prognosis are still not clear, even though pembrolizumab is part of available systemic therapies. The role of tumor angiogenesis emerges through data on microvessel density, immunohistochemical VEGF staining and evaluation of serum VEGF concentrations. Few data exist on hormonal receptor expression, even though hormonal therapy showed great manageability.

Conclusions

We suggest adding p16, p53 and HPV status to routine hystopathological examination of vulvar biopsies or surgical specimens. Predictive biomarkers for anti-EGFR and anti-PD-1/PD-L1 drugs are needed. Enough preclinical data supporting anti-angiogenic target therapies in clinical trials are existing. Hormonal receptor expression deserves further investigation.

Introduction

In 2020, about 6120 new cases of vulvar cancer (VC) will be diagnosed, determining approximately 1350 deaths in the USA (Siegel et al. 2020). VC constitutes 4% of all gynecological malignancies, with an increasing incidence. The 5-year survival varies from 86.3% for localized forms (stage I/II), to 52.6% for locally advanced forms or with loco-regional diffusion (stages III/IVA), and decreasing to 22.7% for cases with distant metastases (stage IVB) (Cancer Stat Facts 2017). Ninety percent are squamous cell carcinomas (VSCC), among which etiopathological and clinical differences are recognized.

Vulvar intraepithelial neoplasia (VIN), the recognized precancerous lesion for VC, includes the two categories: usual VIN (uVIN) and differentiated VIN (dVIN). Human papillomavirus (HPV) is a recognized cause for VC arising in the context of uVIN, with varying geographical incidence, affecting mainly younger women; uVIN carries a low risk of evolving into an invasive lesion (Pino et al. 2013). According to a recent systematic review and meta-analysis by Zhang et al. (2018), the HPV prevalence in VC tissue was 34% (95% CI 28–39%). Conversely, non HPV-related VCs occur in women ranging from 60 to 80 years and are often associated with lichen sclerosus, squamous cell hyperplasia or dVIN, correlated with a higher invasive malignancy risk (Pino et al. 2013). However, the two etiopathogenic and potentially molecular pathways described did not impact, until now, the clinical management of VC women, even though scientific evidence proved superior survival of women with HPV-positive VC (Zhang et al. 2018; Rasmussen et al. 2018).

In 2016, the National Comprehensive Cancer Network formulated VC treatment guidelines. The standard approach in early stages is foremost surgical, while locally advanced or metastatic cases can be treated with adjuvant, primary or definitive treatments consisting of radiotherapy and/or chemotherapy (Reade et al. 2014). For patients with advanced, recurrent or metastatic disease, the guidelines recommend, with a level of evidence 2B, three target-directed therapeutic agents: erlotinib, an inhibitor of the tyrosine kinase activity of the epidermal growth factor receptor (EGFR), bevacizumab, a monoclonal antibody targeting the vascular endothelial growth factor (VEGF), and pembrolizumab, an inhibitor of the programmed death-1 interaction with its ligand, called PD1-PD-L1 immune checkpoint. Therefore, therapeutic alternatives, most of all for advanced stages with poor prognosis, should be identified. The aim of our study was to review and assess literature on the underlying molecular mechanisms of VSCC carcinogenesis, to identify new relevant markers for prognostic and therapeutic purposes.

Methods

The data research was conducted using the following databases: MEDLINE, EMBASE, Web of Sciences, Scopus, ClinicalTrial.gov, OVID and Cochrane Library querying for all articles related to VSCC from the inception of the database up to January 2020. As provided for vulvar Paget’s disease (Mantovani et al. 2019), we considered literature relevant to the following questions: What are the basic molecular characteristics of VSCC? Does it express biological markers available as therapeutic targets? What are the essential features of VSCC cells which drive the malignant transformation and what are their biological markers?

The studies were identified with the use of a combination of the following text words: ("vulvar" AND "cancer” OR “carcinoma"), AND ("molecular”, OR “biological”, OR “protein”, OR “expression”, OR “gene”, OR “mutation”, OR “target" OR “receptor”). The selection criteria of this narrative review included randomized clinical trials, nonrandomized controlled studies and review articles of VSCC and molecular mechanisms in view of target therapeutic strategies. The entire list of publications was cheeked for duplications. Article not in English language, published prior to 1995, conference papers and reviews, and studies with overlapping information were excluded. Additionally, we reviewed the reference lists to identify other relevant documents.

Figure 1 reports the flowchart.

Fig. 1.

Flowchart––studies selection

Study conception and design involved (G.G), (S.M. F.), (G.M.) and (G.S.). The search strategy was conducted by an independent investigator (G.M.) while (S.M. F.), and (L.D.C.) contributed to full text examination and final selection of the eligible studies. G.M., S.M. F., F.I., S.G., L.T. and L.D.C. contributed the extraction of relevant data. Critical revison was provided by (G.G.), (S.G.), (L.T.), (A.F.), (G.Z.) and (G.S). Final approval was produced by all the authors.

We used the same framework as for vulvar Paget’s disease (Mantovani et al. 2019), organizing and assessing knowledge regarding extracellular and intracellular cell cycle deregulation, tumor immune microenvironment, tumor angiogenesis and hormones.

Results

A total of 1230 studies were initially considered using the detailed search strategies. After the title and abstract evaluation and the exclusion of papers not relevant to the topic, duplicates and non-English literature, 312 publications were selected for full-text lecture, of which 63 articles were considered eligible for the present review. The information driven from their analysis were framed in five fields of interest Fig. 1.

Cell cycle deregulation: extracellular signaling

EGFR is a transmembrane receptor that binds the epidermal growth factor (EGF) and the transforming growth factor-α. EGFR constitutes a proto-oncogene, affecting a number of molecular cascades and possible pathogenetic targets, whose alterations can lead to self-sufficiency from growth signals. Anti-EGFR drugs include monoclonal antibodies such as cetuximab, which block the ligand-binding site, and small molecules (gefitinib, erlotinib), which compete with adenosine triphosphate at the tyrosine kinase-binding site. An abnormal EGFR expression and activity have been reported in many cancers, including lung, head–neck, pancreas and colon cancers (Ciardiello and Tortora 2008). Alterations can be present as constitutively active mutated receptor or as overexpression due to transcriptional/post-transcriptional mechanisms or gene amplification may be present (Grandal and Madshus 2008).

Studies investigating the role of abnormal EGFR functioning in the carcinogenesis of VSCC are continuing since decades and have been recently revised by Clancy et al. (Clancy et al. 2016), giving us the chance to understand the evolution in laboratory testing used to study the receptor alterations, as described below.

High levels of EGFR expression in non-HPV-determined VSCCs were initially highlighted with immunohistochemical (IHC) methods showing a correlation with advanced stage, inguinal lymph node (LN) metastases and worse survival, though results are non-definitive and sometimes conflicting. An increasing expression has been demonstrated in the passage from normal epithelium to VC and finally LN metastasis (Johnson et al. 1997; Brustmann 2007). The correlation with the risk of a nodal spread has been outlined and related to disease-free survival (DFS) in 61 total cases (25% vs. 54% in those patients with EGFR expression levels ≥ 90%) (Johnson et al. 1997). However, the probability of not having LN metastases decreased only from 13 to 6% if EGFR expression was considered in addition to classical histopathological parameters (Oonk et al. 2007). Subsequently, adding fluorescence in situ hybridization (FISH) techniques, EGFR gene amplification seemed to characterize a subgroup of HPV-independent tumors linked to decreased survival (Growdon et al. 2008). An increased gene copy number was associated with high tumor stage, number of LN metastases and HPV negativity in a retrospective study analyzing the tumoral tissue of 183 women (Woelber et al. 2012). Finally, no mutation in EGFR gene sequence has ever been described (Trietsch et al. 2015; Palisoul et al. 2017). Several case reports describe the use of EGFR inhibitor drugs in locally advanced or recurrent VC cases, alone or associated with classical chemotherapeutics (Richard et al. 2008; Olawaiye et al. 2007; Bacha et al. 2011; Inrhaoun et al. 2012). In the only phase II clinical trial conducted, 41 patients received erlotinib for a minimum of 28 days and a maximum of 42 days. Patients belonging to the I cohort were candidates for surgery or radiotherapy/chemotherapy, while patients belonging to the II cohort had measurable metastatic disease. A partial response was obtained in 27% of cases, while there was a stable disease in 40% and a progression in 17% (six out of seven patients showing disease progression belonged to the II cohort and were previously multi-treated). Seven patients could not continue the study because of excessive toxicity and several others had grade 3 toxicity. The results raise fundamental questions about whether previous therapies can affect the response to erlotinib, considering gene amplification a possible predictive marker for treatment response (Horowitz et al. 2012).

Cell cycle deregulation: intracellular signaling

The cell cycle is a biological mechanism that proceeds through different phases (G0, G1, S, G2 and M) and is strictly regulated by cyclin-dipendent kinases (important for G1-S transition through phosphorylation of the retinoblastoma protein Rb), activating coenzymes (the cyclins) and several cyclin-dependent kinase inhibitors (p16, p21, p27). The majority of studies in VC focus on the p16-cyclin D/CDK4/CDK6-Rb-pathway (Milde-Langosch and Riethdorf 2003). In parallel, the tumor suppressor gene p53 plays a critical role to preserve DNA fidelity from various insults through the regulation of cell cycle checkpoints, DNA repair, senescence and apoptosis (Blandino and Agostino 2018).

A series, comprising 57 VSCC cases, showed 51% abnormal cyclin D1 expression (associated with a greater depth of invasion) and 37% abnormal pRb (associated with a poor tumor grade), which were additionally found in precursor lesions adjacent to malignant tumors (Rolfe et al. 2001). Only a few cases showed abnormal expression of both proteins; this is consistent with the finding of another study where cyclin D1 IHC staining was different in p16-negative (high expression) and p16-positive (low expression) tumors (Stewart and Crook 2014). Similarly, Choschzick et al. described a moderate or strong cyclin D1 expression in 45.5% of 183 VCs; increased levels of expression were related to positive lymph nodes and negative HPV status (Choschzick et al. 2012).

The loss of p16 expression is frequently seen in a variety of human tumors (for example ovarian cancer) and mainly happens by gene deletion, point mutations or promoter hypermethylation. In HPV-related cancers, such as VC and cervical cancer, the E7 oncoprotein binds and inactivates the tumor suppressor protein Rb, leading to p16 overexpression. After years of single studies giving contrasting conclusions on the prognostic value of p16 overexpression in VC, two recent systematic reviews and meta-analyses stated a possible association with a better prognosis (Cao et al. 2016; Sand et al. 2019). In the meta-analysis by Cao et al., 17 studies, reporting data of 2309 patients, were included. The overexpression of p16 was significantly associated with a lower FIGO stage, absent LN metastasis, decreased patient age (< 55), positive HPV status and a higher overall survival (OS) with a risk ratio of 0.53 (95% CI 0.35–0.80) (Cao et al. 2016). More recently, Sand et al. included 475 VSCC cases (of which 38% were p16 positive) in the analyses of OS. The pooled hazard ratio was 0.40 (95% CI 0.29–0.55) and p16 represented a possible independent prognostic marker for OS considering studies with adjusted analyses for other main prognostic factors (Sand et al. 2019).

Regarding p53, in HPV-related cancers, the oncoprotein E6 promotes rapid degradation of p53. Conversely, the majority of p53 mutations result in a dysfunctional protein and its accumulation in cancer cells. According to a previous review, p53 mutations and p53 protein overexpression were found in 35–68% of VCs and in 13–61% of VIN, suggesting a role in early carcinogenesis (Knopp et al. 2009). Indeed, recent studies focusing on mutational analysis confirm p53 as the cancer-related gene more frequently mutated in VSCC (Palisoul et al. 2017). p53 IHC overexpression has been considered a feature of the HPV-independent pathway (Pino et al. 2013; Gadducci et al. 2012; van der Avoort et al. 2006) and a correlation with a worse prognosis has been provided in the first systematic review and meta-analysis on this topic (Sand et al. 2019). The latest studies on gene sequencing and somatic mutation profiling of VC lead to the same results. Trietsch et al. (2015) reviewed the literature reporting genetic and epigenetic changes in VSCC and cancer precursor lesions; p53 mutation frequency was 28/171 (16%) among HPV-positive and 109/361 (30%) among HPV-negative tumors. In a recent retrospective molecular evaluation of 43 cases, p53 mutation were significantly associated with viral infection absence (9% vs. 62%). In the HPV-negative group, patients with mutations in p53 tended to have a poorer prognosis, although not reaching significance (Weberpals et al. 2017). Similar percentages of p53 mutations related to HPV status (8% in HPV-positive cases vs. 76% in HPV-negative cases) are reported in a study, comprising 72 VSCCs patients, in which p53 mutated cancer patients definitively had a worse prognosis with a 31% mortality rate (Kashofer and Regauer 2017). Recently Nooij et al. identified, through next generation sequencing, p53 mutations in 66% of HPV-negative and 29% of HPV-positive specimens and proved a substantial agreement with immunohistochemistry (IHC). In this research, prognosis was related to HPV status and not to p53 (Nooij et al. 2017). Recently, in 81 VSCCs, HPV-negative and positive cases showed similar p53 mutation rates (41% and 46%, respectively). The presence of p53 mutations was not correlated with the stage or grade of the disease, or with progression-free survival (Zięba et al. 2018). Mutated p53 is a challenging targetable biomarker (Blandino and Agostino 2018). However, adding p16, p53 and HPV status to classical histopathological features may provide further indications to adapt standard therapy with the aim of improving treatment outcomes.

Tumor immune microenvironment

The tumour microenvironment is a very important and hot topic in cancer research within the past few years. The crucial role of immune microenvironment in carcinogenesis is supported by scientific evidence. Tumors grow within an intricate network of epithelial cells, vascular and lymphatic vessels, cytokines and chemokines, and infiltrating immune cells. Different types of infiltrating immune cells, expressing different types of receptors with stimulating or inhibiting functions, have different effects on tumor progression. PD-L1 is a member of the B7 costimulatory/coinhibitory family, expressed in a wide range of cells including tumor cells and antigen-presenting cells. PD-L1-PD-1 interaction interferes with T cell-mediated signal transduction. The first successful off-label treatment with pembrolizumab, an immunotherapy agent for immune checkpoint blockade, was recently described in a patient with recurrent vulvar carcinoma showing complete clinical remission after two cycles (Shields and Gordinier 2019). Pembrolizumab is an anti-PD1 monoclonal antibody that has shown efficacy and good tolerance in various human tumors, mainly non-small cell lung cancer (NSCLC) and melanoma. In KEYNOTE-028, a nonrandomized phase Ib trial, pembrolizumab was administered to 475 patients with PD-L1-positive advanced solid tumors until disease progression or unacceptable toxicity; among them, 18 patients had VSCC (Ott et al. 2019). Objective response rate (the primary end point) ranged from 0% in pancreatic cancer to 33% in NSCLC. The median OS was of 3.9 months (2.8–5.5 months) in vulvar tumors. Tumor immune microenvironment, PD-L1 expression and tumor mutational burden were considered to predict treatment response.

Descriptive studies regarding infiltrating immune cells in VSCC are rare. CD4+ and CD8+ lymphocytes have been shown within carcinoma nests and in the stroma, without correlation with prognosis (Sznurkowski et al. 2011a). The lack of a prognostic correlation was confirmed in a subsequent study investigating the number of intratumoral CD8+ and Foxp3+ lymphocytes (respectively, cytotoxic effector cells and regulatory cells with an immunosuppressive function), next to HLA class I and indoleamine 2,3-dioxygenase (IDO) (Jong et al. 2012). The latter is an intracellular enzyme involved in the kynurenine pathway, currently considered an important immune escape mechanism in cancer that may represent a mechanism of resistance to anti-PD-1/anti-PD-L1 therapy. IDO expression, but not Foxp3+ infiltration, was associated with significantly worse OS among VSCC patients and was identified as an independent prognostic factor (Sznurkowski et al. 2011b). High intraepithelial GrB+ infiltrates (considering both lymphocytes expressing granzyme B and innate immunity as natural killer cells CD56+) were correlated with longer OS in non-metastatic VSCC cases. Moreover, high intraepithelial CD56+ infiltrates were correlated with longer OS in patients with metastatic dissemination (Sznurkowski et al. 2014).

We found six studies (Table 1) investigating PD-L1 expression in VSCC cancer and immune cells, using different methods of IHC interpretation to address PD-L1 positivity. In the study by Hecking et al. on 103 patients with primary VSCC, almost 10% were classified as PD-L1 positive (staining intensity ≥ 2 in ≥ 5% of tumor cells). PD-L1 expression was significantly related with CD3+ (T cells), CD20+ (B cells) and CD68+ (monocytes/macrophages) intra-tumor immunocytes; moreover, PD-L1 expression was found to occur more often in HPV-negative and p16-negative cancers and resulted to be an independent prognostic factor for worse recurrence-free survival (HRD 3.029, 95% CI 1.228–7.471) (Hecking et al. 2017). Palisoul et al. observed a statistically significant difference between primary and metastatic PD-L1 prevalence (29% vs. 7%, p < 0.05) using an identical threshold considering both intensity of staining and percentage of positive stained tumor cells (Palisoul et al. 2017). Sznurkowski et al. found that PD-L1 positivity (≥ 5%) of cancer cells (32%) was more frequent in p16-negative tumors that resulted to be more infiltrated (Sznurkowski et al. 2017a). Only PD-L1 positivity of peritumoral immune cells (61%) was found to be an independent favorable prognostic factor for OS, as recently described for other cancer types (Zhao et al. 2017). The same researchers described in another work the impact of p16 status on immune infiltrate (Sznurkowski et al. 2017b). p16-negative tumors were more infiltrated by intraepithelial CD8+, CD4+ and GrB+ cells than p16-positive tumors. Anyway, the correlation of the tumor immune microenvironment characteristics with the HPV status remains controversial. Choschzick et al. (2018) found no association of tumoral PD-L1 expression with HPV status or OS of 55 VSCC patients (72.7% PD-L1-positive cancers using a 1% threshold for staining on tumor cells). These results are similar to those reported in a previous study by Howitt et al. (2016). The authors identified a genetic base for PD-L1 expression in copy number gain of the encoding genes, failing anyway to find a correlation with p16, intended as a surrogate biomarker of HPV infection. Similarly, Chinn et al. (2019) found no significant difference in PD-L1 expression on tumoral cells in dVIN-associated versus HPV-associated VSCC (75% vs. 63% using a 1% threshold). In conclusion, further efforts are needed to better understand the implications of PD-L1 positivity in reference to HPV status and prognosis, despite clinical applications havingg already shown interesting outcomes.

Table 1.

Selected studies on PD-1/PD-L1 expression in VSCC

Author	Number of cases	Laboratory	Results
Chinn et al. (2019)	20 VSCC	PD-L1 membrane staining in ≥ 1% of tumor cells regarded as positive IDO cytoplasmatic staining in ≥ 1% of tumor cells regarded as positive	63% of HPV-associated VSCC PD-L1 positive 75% of dVIN-associated VSCC PD-L1 positive PD-L1 immune cell expression paralleled tumoral expression
Choschzick et al. (2018)	55 VSCC	PD-L1 membrane staining in ≥ 1% of tumor cells regarded as positive	73% PD-L1 positive Correlation of PD-L1 positivity with low tumor stage No correlation with HPV status or OS
Hecking et al. (2017)	103 VSCC	Evaluation of immune infiltrate: CD3+, CD20+, CD68+, Foxp3+, CD163+ PD-L1 membrane staining intensity ≥ 2 in ≥ 5% of tumor cells regarded as positive	23% PD-L1 positive Correlation of PD-L1 positivity with immune infiltrate No correlation with Tregs and TAMs (inhibitory) Correlation of PD-L1 positivity with HPV/p16 negativity and worse RFS (independent prognostic factor)
Sznurkowski et al. (2017a, b)	84 VSCC	Evaluation of immune infiltrate: CD8+, CD4+, Foxp3+, CD56+, CD68+, GZB+ PD-L1 membrane staining in ≥ 5% of cells regarded as positive	32% PD-L1 positive in cancer cells 61% PD-L1 positive in peritumoral immune cells Correlation of PD-L1 positivity in cancer cells with immune infiltrate (CD4+, CD8+, Foxp3+, CD68+) Correlation of PD-L1 positivity in peritumoral immune cells with Foxp3+ Tregs Correlation of PD-L1 positivity in cancer cells with p16 negativity No correlation with HPV status Correlation of PD-L1 positivity in peritumoral immune cells with favorable OS (independent prognostic factor)
Palisoul et al. (2017)	51 VSCC (24 primary, 27 metastatic)	PD-L1 membrane staining intensity ≥ 2 in ≥ 5% of tumor cells regarded as positive	18% PD-L1 positive PD-L1 expression lower in metastatic sites (7% vs. 29%)
Howitt et al. (2016)	23 VSCC	FISH to determine genetic status of CD274 and PDCDILG2 IHC to determine PD-L1 (modified H score 0–300 considering intensity and percentage of staining)	No association between p16 status and the genetic category of the tumor (coamplification, cogain, polysomy, disomy) Highest median PD-L1 expression among cases with coamplification

VSCC vulvar squamous cell carcinoma, PD-L1 programmed death-ligand 1, IDO indoleamine 2,3-dioxygenase, HPV human papillomavirus, dVIN differentiated vulvar intraepithelial neoplasia, OS overall survival, Tregs regulatory T lymphocytes, TAMs tumor associated macrophages, RFS recurrence free survival, GZB granzyme B, Foxp3 forkhead box protein P3, CD274 gene encoding PD-L1, PDCDILG2 gene encoding PD-L2

Tumor angiogenesis

Angiogenesis is a complex biologic process that has a key role in tumor growth, progression, invasiveness and metastases and is amenable to direct targeting (Stephenson et al. 2013). The vascular endothelial growth factor-A (VEGF-A) is the best-known and most studied growth factor involved. Antiangiogenic agents include bevacizumab, a humanized monoclonal antibody, and aflibercept, a chimeric protein produced by recombinant DNA technology, which binds to VEGF-A (Chellappan et al. 2018). After the GOG 240 trial results, bevacizumab has been FDA approved for patients with persistent, recurrent or metastatic cervical cancer (Tewari et al. 2014). Given the etiological and histological similarities with VSCC, investigations on the role of VEGF in this neoplasm seem mandatory, starting from preliminary data. Moreover, the therapeutic regimen cisplatin/paclitaxel/bevacizumab has already been included, in NCCN guidelines, among the systemic therapies available for recurrent/metastatic VSCC, based on extrapolation of randomized phase III trial data in cervical cancer (Tewari et al. 2014; Rosen et al. 2017). The most frequently investigated parameters were microvessel density (MVD), IHC VEGF staining along with other hypoxia markers and evaluation of serum VEGF concentrations in patients affected by VSCC Table 2. Studies including specimens of pre-neoplastic lesions, such as LS and VIN, showed an increasing and progressing expression of angiogenesis indicators in evolving LS and higher grade VIN or VSCC compared to normal controls (Saravanamuthu et al. 2003; Lewy-Trenda et al. 2005; Raspollini et al. 2007; Li et al. 2012). Qureshi et al. (1999) did not find a correlation between MVD and stage, survival or pattern of invasion in 50 VSCCs. Conversely, in most cases an impact on prognosis is evident. Small series reported a significant correlation between MVD, VEGF IHC staining and poorer OS (Obermair et al. 1996). Subsequently, Hantschmann et al. (2005) described a high MVD in 29% of VSCCs and a correlation with TGF-α expression, outlining its role in promoting angiogenesis; tumor with both features tended toward a reduced DFS, although not significant. Not only high vessel number, but also increased vessel size and other vessel characteristics (shape and staining intensity) appeared related to prognosis (Näyhä and Stenbäck 2007). In the most numerous study comprising 158 VSCCs patients, a high vascular density, measured through CD34 immunostaining and Chalkley counting, correlated significantly with tumor diameter and depth of invasion. VEGF IHC expression has been closely related with poor differentiation (Dhakal et al. 2013). Only one study reported that the median serum VEGF concentration, in 41 patients affected by VSCC, was higher compared to 130 controls, showing a correlation with tumor stage and grade; the association with worse DFS and OS was not an independent prognostic factor (Hefler et al. 1999).

Table 2.

Selected studies on angiogenesis in vulvar squamous cell carcinoma (VSCC)

Author	Number of cases	Laboratory	Results
Dhakal et al. (2013)	158 VSCC	IHC for CD34 to measure tumor vascularity trough Chalkley counting method IHC for (HIF)-1α and VEGF	High CD34 Chalkley count significant correlation with larger tumor diameter, deep invasion and (HIF)-1α VEGF significant correlation with poor tumor differentiation High (HIF)-1α expression better DSS
Li et al. (2012)	25 VSCC, 10 VIN, 12 healthy controls	IHC for (HIF)-1α, GLUT-1, CA-9, VEGF (hypoxia markers)	VEGF strong expression 60% VIN, 25% VSCC, significant difference with controls and VIN vs. VSCC Expression of endogenous hypoxia markers might be involved in malignant progression
Raspollini et al. (2007)	10 LS not evolving, 8 LS evolving to VSCC, 10 LS adjacent to VSCC	IHC for VEGF and COX2 IHC for CD34 for fMVD	Mean MVD 30% evolving LS and 24% adjacent LS, significant difference only with not evolving LS VEGF 75% evolving LS and 100% adjacent LS, significant difference COX2 63% evolving LS and 30% adjacent LS, significant difference
Näyhä and Stenbäck (2007)	42 VSCC	IHC for CD34, computer-assisted microvessel count	High vessel number and increased vessel size significant indicators of poor survival All vessel characteristics together (vessel number, size, shape and staining intensity) significant prognostic factor
Hantschmann et al. (2005)	75 VSCC	IHC for TGF-α, c-erbB-2 and factor VIII antigen	65% TGF-α positive > 50% tumor cells, 47% c-erbB-2 overexpression 29% high MVD Correlation TGF-α overexpression and MVD, in tumors with both DFS tended to be reduced
Lewy-Trenda et al. (2005)	31 VSCC, 28 VIN III, 10 VIN II, 12 VIN I	IHC for VEGF	VIN I and VIN II expression 1+/2+ and focal VIN III expression 2+/3+ and diffuse VSCC expression 3+ and diffuse
Saravanamuthu et al. (2003)	31 LS, 13 VIN III, 10 VSCC	IHC for vWF for MVD	MVD continuous significant increase from normal epithelium to VIN III to VSCC
Hefler et al. (1999)	41 patients with VSCC, 130 controls	VEGF serum concentrations	Median serum concentrations higher in patients with VSCC, significant difference Correlation with stage and grade Associated with DFS and OS, but not independent prognostic factor
Qureshi et al. (1999)	50 VSCC	IHC for vWF for MVD	No correlation with stage, survival or pattern of invasion
Obermair et al. (1996)	25 VSCC	IHC for VEGF IHC for factor VIII antigen for MVD	High MVD more frequent in tumors with moderate/strong VEGF expression High MVD worse OS Strong VEGF expression worse OS

IHC immunohistochemistry, (HIF)-1α hypoxia-inducible factor 1-alpha, VEGF vascular endothelial growth factor, DSS disease specific survival, VIN vulvar intraepithelial neoplasia, GLUT-1 glucose transporter 1, CA-9 carbonic anhydrase IX, LS lichen sclerosus, MVD microvessel density, COX2 cyclooxygenase 2, TGF-α transforming growth factor-α, c-erbB-2 human epidermal growth factor receptor 2, DFS disease-free survival, vWF von Willebrand factor, OS overall survival

Hormonal background

Hormonal receptors are commonly expressed in a variety of human tissues and represent an easily targetable biomarker for low-cost therapies with a good tolerability profile, as is commonly known for breast cancer, endometrial cancer and ovarian cancer.

Although estrogen receptors (ER) have been described in normal vulvar epithelium (Schwartz 1990) and vulvar lichen sclerosus (Taylor et al. 2008), the role of hormonal receptors in VSCC carcinogenesis has not been investigated deeply. Zannoni et al. previously conducted a study on 34 patients affected by VSCC to evaluate the expression of the two ER isoforms (ERα and ERβ) along with the progesterone receptor (PR). Values indicating the mean percentage of stained cells and the staining intensity were multiplied to calculate a receptor score. ERβ was the predominant receptor in normal epithelium and shifted from a mainly nuclear to a mainly cytoplasmatic staining in the transition to cancer, showing a reduction in protein expression. ERα staining was completely lost in neoplastic epithelium (Zannoni et al. 2011a). A subsequent research showed that the expression of cytoplasmatic ERβ was associated with grade 3 neoplasms; cases with high expression showed lower DFS and OS with potential use of ERβ as a therapeutic target (Zannoni et al. 2011b). In a recent retrospective molecular analysis on VSCC, a high expression of ER, PR and AR (staining intensity ≥ 1+ in ≥ 10% of cancer cells) was revealed in about 11%, 4% and 4% of 142 samples, respectively. Although not widely studied, the great manageability of hormonal therapy makes it necessary to investigate this topic for a possible application among VC systemic therapies.

In summary, there are multiple potential therapeutic targets that could emerge, enhancing studies on the different molecular pathways of VSCC (Table 3).

Table 3.

Potential therapeutic target for the fields of interest

Drug	Type of antibody	Mechanism of action	Mutation	HPV related/non-HPV related	Prognostic value (yes, no)
Cell cycle deregulation/extracellular signaling—molecule target: EGFR
Cetuximab Gefitinib, Erlotinib	Mouse/human chimeric monoclonal antibody Small molecule	Ligand-binding domain of the EGFR TK inhibitors	High levels of EGFR expression	Yes/yes	Yes
Cell cycle deregulation/intracellular signaling—molecule target: cyclin-dipendent kinases (CDK)
NA	NA	NA	High cyclin D1 expression Inactivation pRb Overexpression of p16 p53 mutations or p53 protein overexpression	Yes/yes	Yes
Tumor immune microenvironment—molecule target: PD1
Pembrolizumab	Humanized IgG4 monoclonal antibody	Binds PD1 and inhibits PD-L1-PD-1 interaction, interfering with T cell-mediated signal transduction	Abnormal PD-L1 expression	Yes/yes	Not clear
Tumor angiogenesis—molecule target: VEGF-A
Bevacizumab Aflibercept	Humanized IgG1 monoclonal antibody Recombinant fusion protein consisting of vascular endothelial growth factor (VEGF)-binding portions from the extracellular domains of human VEGF receptors 1 and 2, fused to the Fc portion of the human IgG1 immunoglobulin	Inhibits VEGF-A Inhibits VEGF	Abnormal VEGF expression	Yes/yes	Yes
Hormonal background—molecule target: ERα and ERβ
Estrogen/progesterone	NA	NA	ERβ reduction ERα lost	NA	NA

Conclusions

EGFR immunohistochemical overexpression/gene amplification represents early events in VSCC carcinogenesis. These events have been correlated with a worse prognosis leading to inclusion of erlotinib in cancer guidelines. However, clear response biomarkers have not been identified yet.

Our literature review showed that p16 expression and HPV positivity are linked to a better prognosis, while p53 overexpression is linked to a worse prognosis. We suggest adding these biomarkers to routine histopathological examination of vulvar biopsies or surgical specimens, to help tailor conventional treatment and follow-up.

The implications of PD-L1 positivity in reference to HPV status and prognosis remain unclear. We propose to intensify research to identify “responders” to immunotherapy.

The role of tumor angiogenesis emerges through data on microvessel density, immunohistochemical VEGF staining and evaluation of serum VEGF concentrations. Enough preclinical data supporting anti-angiogenic target therapy use in clinical trials exists; however, few data are available on hormonal receptor expression. We believe that hormonal therapy needs to be considered as a valuable field of investigation.

Funding

None.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.