Introduction

Cancer is a dangerous condition with a high mortality rate worldwide. According to data just released by Iraq's National Cancer Registry, over 31,500 cases of cancer and tumor-related disorders were recorded in 2017-2018. Locally, cancer is thought to be one of the leading causes of mortality, accounting for around 11% of the overall rate [1]. Cancer treatment has not considerably improved despite a decade of unending research, clinical trials, and chemotherapy. Such poor results are attributable in part to tumor cells' invasive nature and resistance to therapy [2, 3]. Salivary gland carcinomas (SGCs) are rare, with a wide range of morphology and clinical behavior [4]. MEC is the most prevalent, accounting for 30% of all malignant tumors in major salivary glands [5]. Adenoid cystic carcinoma (AdCC) accounts for 10% of all salivary gland neoplasms [6]. It has an indolent growth pattern, perineural invasion, and a high prevalence of distant metastases [7]. Polymorphous adenocarcinoma of minor salivary glands (PAC) is uncommon in major glands. It has an invasive growth pattern, morphological variation (thus the word polymorphous), and cytological homogeneity [8]. When cancer cells invade nerves at a distance from the initial lesion, a process called perineural invasion (PNI) has occurred [9]. Oral squamous cell carcinoma (OSCC) [10] and malignancies of the salivary glands are both negatively affected by PNI, with the former being linked to an increased risk of metastasis and a shorter survival time [11]. Many growth factors and signaling pathways should be activated for cancer cells to migrate toward nerves [12]. PI3K/Akt pathway is generally believed to act as a very important signaling pathway of cancer progression, which is closely related to cell proliferation, metabolism, and tumor growth. The activated form of PI3K is also capable of phosphorylating a series of downstream target proteins such as Bad, Caspase 9, p2l, and mTOR which generally promote cell survival [13]. The other proliferative signaling pathways are Cyclin D1 and CDK4 which have been known to increase in aggressive thyroid cancer [14]. It has therefore been suggested that dual inhibition of these pathways extremely inhibit cell growth. Salivary peroxidase is a key scavenger enzyme in the antioxidant system S100 protein is a member of a large family of calcium-binding proteins, it is a soluble protein that was used for the detection of Schwann cells in different tissues some studies demonstrated that S100 protein can be included in several biological events that might be involved in tumor progression [15-18]. PGP9.5 is a cytoplasmic protein known as ubiquitin C-terminal hydrolase L1 (UCHL1), which was discovered in the normal human brain and is specialized for nerves and the neuroendocrine system. Furthermore, increased nerve density can be identified in neurotropic tumors [19]. The neurotrophic factor family includes nerve growth factor (NGF and NGFR). NGF not only influences the proliferation and differentiation of tumors, including head and neck neoplasms, but it may also impact the growth, development, and regeneration of many types of nerve cells in the central and peripheral nervous systems [20, 21]. The nerve growth factor receptor (NGFR) is a transmembrane glycoprotein that is expressed in cells other than neurons and lacks intrinsic tyrosine kinase activity. According to reports, NGFR functions as a tumor suppressor, adversely inhibiting cell growth and proliferation [22]. GAL, a neuropeptide, binds to three G protein-coupled receptors: GALR1, GALR2, and GALR3. Cancer-induced GAL increases neurogenesis and enhances PNI. GAL was linked to tumor-positive lymph nodes and tumor recurrence, both of which are independent predictors of poor prognosis [23]. In this study, it was aimed to investigate the signaling pathway of NGF, NGFR, and Galanin in the invasion and to evaluate PNI in salivary gland carcinomas based on tumor grade and histological subtypes of this biological process.

Martials and Methods

From the archives of Oral and Maxillofacial Pathology/College of Dentistry/the University of Baghdad, Ghazi Al-Harerri Hospital-Medical City, 50 formalin fixed paraffin embedded tissue blocks of surgically removed salivary gland carcinomas of the head and neck from 2000 to 2021, containing (20, 21 and 9 cases of MEC, AdCC and PAC respectively). Information of the patient's age, gender, location, and tumor histology are taken from the histological reports. In this study, MEC is graded using the Armed Forces Institute of Pathology (AFIP) grading method. A fundamental component of a point-based system is the presence or absence of perineural invasion (PNI) [24, 25]. Perzin/Szanto et al. described the histopathological grading approach used in this study for AdCC [26]. Two qualified pathologists examined the tumor's histopathology. In each example, 4 mm of a selected block was used for immunohistochemistry. According to the datasheets for Abcam's and PathnSitu's immunohistochemically detection kits, the samples were deparaffinized in xylene, dried in ethanol, and then processed with a heat-mediated antigen retrieval solution for S100 alone. (Detection of HRP/DAB in Rabbits; Micro-polymer ab 23649, USA) PolyExcel HRP/DAB Detection Systems were utilized for NGF, NGFR, PGP9.5, and PathnSitu, while the IHC Detection Kit was used for S100 and GAL, two steps: after washing for 10 minutes, apply HRP for 20 minutes, and then wash for 30 minutes. The antibodies utilized in this work are summarized in Table 1, along with some of their important features.

The NGF score was done by Yu et al. [27]. Simpson et al. NGFR scoring was used [28], as well as Galanin score according to Brener et al. [29].

PNI criteria

Perineural invasion was defined as the presence of tumor cells inside a 3-layer nerve sheath or within close proximity to neural structures (constituting at least 33% of the neural circumference) [25]. Two oral pathologists analyzed all of the verified nerve bundles for inconsistencies and the presence or absence of histopathological slides stained with E&H, S100, and PGP9.5 antibodies.

Statistical analysis

Using statistical

Package for Social Sciences (SPSS) version 25, A Chi-square test was used to assess the association between clinical variables. A level of P-value less than 0.05 was considered significant.

Results and Discussion

Distribution of perineural invasion in tissue sections stained with H&E, S100 and PGP9.5 across SGCs subtypes

Twenty-three cases (46%) of the total SGCs samples revealed the presence of PNI in H&E sections, where 8 (40.0%) in MEC, 13 (61.9%) in AdCC, and 2 (22.2%) in PAC and the distribution did not differ significantly between cases (p=0.41). Using S100 immunohistochemically marker, PNI has increased to 25 (50%) cases in total SGCs which 10 (50.0%) in MEC, 12 (57.1%) in AdCC, and 3 (33.3%) in PAC and distribution did not differ significantly between cases (p=0.4). Using the PGP9.5 immunohistochemically marker, PNI was increased to 36 cases (72%) in total SGCs which 15 (75.0%) in MEC, 17 (81.0%) in AdCC, and 4 (44.4%) in PAC and the distribution did not differ significantly between cases (p=0.11). The total number of cases with PNI in (H&E, S10,0, and PGP9.5 collectively) was 39 (78%).

All markers contributed to PNI significantly (p<0.05), as demonstrated in Table 2 and Figures 1, 2, and 3.

Sensitivity and specificity of immunohistochemically markers in the diagnosis of PNI in SGCs

The highest specificity was observed for PGP 9.5 (78.6%) in which the test can truly roll out all negative cases (sensitivity 100% of all tests). The PPV was at its highest for PGP 9.5 (92%) in which there was a 92% probability that cases with a negative screening test result indeed did not have PNI, as presented in Table 3.

Association between PNI and tumor grade of SGCs

PNI was present in 16/20 (80%) MEC, 19/21(90.5%) AdCC, and 4/9(44.4%) PAC. The subset relationship between histological grade and PNI based on carcinoma type showed that there was a significant difference in the PNI distribution for both AdCC (p=0.034), and MEC (0.041), while PNI among PAC cases could not be assessed due to all PAC cases are considered as low grade, as listed in Tables 4, 5, and 6.

Association of neurite outgrowth (NGF & NGFR) and neuropeptide (Galanin) with PNI and histological subtypes

The tumor cells in SGCs tissues with positive NGF, NGFR, and GAL staining were found to have these markers localized either in the cytoplasm or on their cell membranes. The assessment of NGF, NGFR, and GAL showed that for NGF, the majority of cases (76%) had scores of 3 and 4 and only one case did not show NGF, for NGFR score and intensity, and also the majority of cases had a score of 3 and 4 and intensity of the moderate and strong (65% and 80%, respectively). For Gal, (70%) of cases had (+++ and ++++). Both NGF and GAL expression were significantly correlated with PNI status (P0.001 and P0.01, respectively), while there was no relation between NGFR with PNI, as provided in Table 5. Likewise, there was an association between NGF and histological subtypes (P=0.01), as indicated in Table 6 and Figure 4.

Association between Immunohistochemically markers

Results showed a significant association of H&E with S100, PGP9.5, NGF, NGFR, and GAL. S100 was significantly correlated with PGP9.5 and GAL. PGP9.5 was significantly correlated with NGF and GAL. NGFR score is associated with its intensity. Finally, there was a significant correlation between NGF, NGFR, and GAL, as demonstrated in Table 7, 8, 9, 10, 11, and 12.

The degree of perineural invasion (PNI) into surrounding tissues is a strong predictor of how aggressively a tumor will spread. In PNI, a critical phase is for cancer cells to invade and spread throughout neural tissue. Cancer cells have the ability to secrete neurite outgrowth-encouraging and axonal-directing chemicals, which cause axonal outgrowth toward the tumor [25]. Histopathological reevaluation of H&E slides from study samples confirmed the PNI presence in SGCs in 23/50 instances. Only ten numbers among them, in fact, had PNI in their related histopathology findings. However, identifying the nerve structure in tumor specimens with H&E is sometimes challenging because the invasion causes the nerve tissue to change its histologic form and become difficult to recognize [30]. S100 and PGP9.5 staining confirmed the presence of 25/50 and 36/50, respectively. The number of foci observed differed somewhat between the two markers, which can be ascribed to variances in the sectioning method, poor processing leading to tissue folding or dislodgement, and differences in staining intensity. The detection rate of PNI was increased to (78.0%), which was consistent with earlier research [31]. Our findings demonstrated that, as compared with conventional H&E labeling, both S100 and PGP9.5 immunohistochemically staining might improve nerve bundle identification more than H&E. We showed, for example, that PNI detected by PGP9.5 had greater specificity than PNI identified by S100 staining. White and colleagues [32] further thought that S100 staining had a reproducibility issue. Thus, we supposed that the high PNI incidence proved by Kurtz et al. [33] was due to the inaccurate nerve bundle identification using S100 labeling.

Figure 4: Photomicrograph showing NGF expression in MEC tumor cells (A): High immunoreactivity of few tumor cells (Original magnification 10X), (B): Same case in (A) showing high immunoreactivity of tumor cells. (Original magnification 40X). NGF expression in AdCC tumor cells, (C): High immunoreactivity of few tumor cells (Original magnification 10X), (D): Same case in (A) showing high immunoreactivity of tumor cells. (Original magnification 40X). NGF expression in PAC tumor cells, (E): High immunoreactivity of few tumor cells (Original)

This is the first study to show that PGP9.5 outperforms S100 as a particular marker of peripheral nervous system neural fibers and its applicability in detecting perineural invasion of salivary gland carcinomas. Our findings are similar to previous studies that show that standard H&E labeling has a poor rate of identifying PNI in colon cancer tissue samples and that S100 staining can greatly boost the incidence of PNI detection [34]. The distribution of H&E, S100, and PGP9.5 among SGC subtypes was not substantially different, demonstrating that each staining technique may efficiently detect nerve bundles. Each of the H&E, S100, and PGP9.5 markers significantly contributed to PNI. Our results indicate that IHC might be a useful regular method in surgical pathology laboratories to identify PNI of SGCs. PNI and tumor grade were shown to be significantly related. There was a substantial relationship between PNI and MEC and AdCC grades. As a result, our findings support the ongoing use of perineural invasion as a grading criterion in MEC and AdCC. PNI is an important factor in the survival outcome of MEC patients, according to Lanzel et al. [35], and S100 marker in MEC sections increases the accuracy, ease, and speed of PNI assessment. Katabi et al. found no significant effect of PNI on overall survival, illness specific survival, or recurrence free survival in 2014 [36]. McHugh et al. discovered in 2012 that poor clinical outcomes were related with high tumor grade, advanced stage, perineural invasion, positive surgical margins, and submandibular placement [37]. Many writers have examined the influence of histological patterns on the clinical outcome of AdCC [38, 39]. Some studies have indicated that perineural invasion predicts a poor prognosis [40-42], whereas others have found that perineural involvement has no effect on the outcome [43]. Jang et al. revealed in another investigation that AdCC with PNI proceeded to distant metastasis, but no distant metastasis was detected in individuals without PNI [44]. Other findings suggest that PNI accelerates distant metastatic development, potentially impacting patient outcomes [45]. Polymorphous adenocarcinomas have a low tendency for PNS. According to the prior research, PNI was (52%) [46], while in our findings, PNI is detected in (44.4%). However, PNI may be detected in around 30% of PAC instances [47]. PAC should not be dismissed as a generally low-grade malignancy because it can recur, spread, and even kill. As a result, the WHO decided to rebrand PLGA to PAC. Over a 5- to 10-year period, the local recurrence of PAC has been reported to be 5.3-33%, with an average time interval of roughly 70 months [8]. The assessment of perineural invasion ranges greatly depending on the histological subtype. That is, the increased prevalence of PNI may simply indicate a more aggressive tumor, which is strongly associated with a poor prognosis, as shown in MEC and AdCC. The PNI incidence in the latter might range from 27-82% of cases, with AdCC having the highest tendency for PNI, followed by SCC [48]. Indeed, the existence of several nerves around these tumors and/or the presence of interchanging communication between tumor cells and accompanying nerves might explain this incidence difference [49]. Because of its rarity, polymorphous adenocarcinoma has a disproportionately low number of cases. The expression of tumor cells for NGF and NGFR in the current investigation demonstrated a significant immunoreactivity of cancer cells in SGC subtypes; their expression was not linked with tumor grade. The current investigation found a substantial association between NGF and PNI, indicating that NGF may play an important role in PNI even when tiny nerves are involved. Farah (2018) showed that NGF is unrelated to tumor grade [31]. In malignancies of the oral and salivary glands, NGF may contribute to the initiation of PNI [50]. As previously stated, the PNI development may necessitate an interchanging cross-talk between tumor cells and neighboring neurons, in which NTs and axonal guidance molecules play a key role [51]. Our findings revealed a substantial relationship between NGF and NGFR, overexpression of NGF suggested that SGCs cells produced NGF, and staining intensity was associated with pathological subtypes of various cancers. Researchers have also shown that NGF and its receptor are expressed in AdCC tissue, with the solid type having higher levels of expression than the cribriform-tubular type, which is generally considered to be less aggressive [52, 53]. Previous research found a low positive rate of NGFR immunostaining and no link between NGFR overexpression and PNI in AdCC. However, there were high correlations between NGF/TrkA immunostaining and PNI, confirming the NGF/TrkA system's involvement to PNI in AdCC [54]. Kolokythas et al. (2010) in a study of the role of NGF in OSCC stated that there was an upregulation of NGF and its receptor in areas adjacent to nerves [55]. The most likely reason is that tumor cells and nerves develop together because NGF, produced by AdCC, MEC, and PAC cells, is drawn to NGFR in nerve tissues. Furthermore, the earlier studies show that NGF expression generates greater neovascularization surrounding neurofibers, supplying resources for tumor development, and boosting quicker proliferation of the tumor cells nearby and perineural invasion [56, 57]. H&E staining was used to diagnose PNI, and Frydenlund et al. [58] noted that NGFR-positive staining is associated with PNI in desmoplastic melanoma.

Our result, proved an association between NGFR score and its intensity which is in line with a previous study, in which the intensity of staining by NGFR correlated to the PNI presence. Neuropeptides may be secreted by tumor cells, in which their expression was suggested to affect cellular proliferation and correlated with the clinical course of the tumor [59]. These neuropeptides were considered as growth factors that upon their binding to their corresponding type of receptors GPCRs would contribute to tumor aggressiveness, as these receptors were found to be overexpressed by tumor cells inducing tumor invasion, angiogenesis, and metastasis [60]. Two-thirds of SGCs cases were of high score value of GAL. Despite the failure of the current study to produce a significant relation of GAL expression with tumor grade and histological variants, which is in a line with the previous study (Farah, 2018) [31], but yet previous studies mentioned in general a positive relation of peptides level with tumor stage and prognosis [61]. This discrepancy may be attributed to the scoring system that was used in addition to the previously mentioned limitations relevant to the studied variables in the present study. Perineural invasion status also showed a significant relation in the present study with GAL expression in H&E, PGP9.5, and S100 sections. While Farah (2018) [31] proved that GAL is not associated with H&E, S100 except with PGP9.5. As mentioned before, GAL is released by nerves as well as tumor cells [62]. As a feedback mechanism, tumor cells were found to release GAL that enhances neuritogenesis [23]. One of the studies revealed that the GAL expression was related to MMP2 and MMP9 expression [63], a matter that may point to an additional indirect role of GAL in enhancing PNI. Several studies have demonstrated that the GAL expression can be changed according to cell type, hormonal, and growth factors including NGF and NGFR, which was even found in the present study as a significant relationship between the mentioned markers, which are consistent with earlier research [31]. According to our findings, immunoreactivity for NGF, NGFR, and GAL was moderate at best, with a greater proportion of positive instances seen in high-grade MEC, AdCC, and PAC overall suggesting that AdCC, MEC, and PAC cells expressed NGF, NGFR, GAL, and the staining intensity correlated with pathological subtypes of different malignancies.

Conclusion

The results of this study indicate that NGF, NGFR, and GAL, together with the other pathological variables such as perineural invasion, are linked to the invasion and metastasis of AdCC, MEC, and PAC, and that these parameters may serve not only as prognostic markers, but also as potential therapeutic targets in these diseases.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Authors' contributions

All authors contributed to data analysis, drafting, and revising of the paper and agreed to be responsible for all the aspects of this work.

Conflict of Interest

The author declared that they have no conflict of interest.

ORCID:

Alyaa Mohammed

https://orcid.org/0000-0002-4577-4945

Ban Al-Drobie

https://orcid.org/0000-0002-9796-9322

Bashar H Abdullah

https://orcid.org/0000-0001-9983-1661

HOW TO CITE THIS ARTICLE

Alyaa Kadhim Mohammed, Ban Al-Drobie, Bashar H. Abdullah. Perineural Invasion in Salivary Gland Carcinomas in Relation to Tumor Grade and Histological Subtypes. J. Med. Chem. Sci., 2023, 6(7) 1469-1486

https://doi.org/10.26655/JMCHEMSCI.2023.7.1

URL: http://www.jmchemsci.com/article_160877.html