Evaluation of the Expression EGFR, HER2/NEU and the End Effector ERK of the RAS/RAF/MAP Kinase Pathway in Prostatic Adenocarcinoma for a Possible Role as New Target Therapy

Introduction: The alterations of EGFR and HER2/neu as growth factor receptors and the cytoplasmic signal transduction proteins of RAS/RAF/MAP kinases including its end effector molecule (ERK) are important in the carcinogenesis of many tumors. The activation of these protooncogenes in prostate cancer is still under investigation. The aim of this work was to study EGFR, HER2- neu, inactive (non-phosphorylated) and active (phosphorylated) ERK expression in prostatic adenocarcinomas in correlation to the clinical and pathological parameters. Methods: Immunohistochemistry- using tissue microarrays- for EGFR, HER2/neu, non-phosphorylated, and phosphor-ERK, was performed on tissues from 166 patients- with primary prostatic adenocarcinoma with no prior treatment-. The results of different markers expression were correlated with the clinical and pathological parameters and were analyzed statistically. Results: The prostatic tissue showed EGFR, HER2 neu, phosphorylated and non-phosphorylated ERK expression in 8.4%, 1.4%, 78.2%, and 83.4% respectively whether low (patchy) or high expression (diffuse). There were no significant correlations found between patient characteristics and expression of the tested markers. The negative immune reactivity for non-phosphorylated ERK and EGFR- was significantly correlated with high tumor stage (p values 0.03 and 0.01, respectively). Conclusion: EGFR and HER2/neu may play a limited role in prostatic adenocarcinoma as they showed positive expression in a limited number of the examined tissues specifically HER2neu. The expression of non-phosphorylated ERK (mostly weak to moderate) and phosphorylated ERK (mostly moderate to strong)- was appreciated in most cases. Thus, we suggest that anti-EGFR drugs may have a limited role in the treatment of castrate-resistant prostate cancer, but anti-MEK/ERK drugs may have more promising role as a target therapy. It is recommended to perform further molecular testing to elucidate the exact mechanism and significance of these markers.


Introduction
Prostate cancer is the second leading cause of cancer-related death in the United States with decreasing incidence of prostate cancer overall since 2000 but an increasing incidence of wide metastatic disease [1].Metastatic castration-resistant prostate cancer (m CRPC) is incurable with no effective therapy till now beyond hormonal treatment.Thus, there is an urgent need to develop new effective therapy [2].
The genetic changes in EGFR gene-either mutations or amplification-is implicated in the progression of many

Evaluation of the Expression EGFR, HER2/NEU and the
End Effector ERK of the RAS/RAF/MAP Kinase Pathway in Prostatic Adenocarcinoma for a Possible Role as New Target Therapy types of cancers and currently the EGFR inhibitors such as gefitinib and cetuximab are in clinical use for metastatic colorectal cancer and non-small cell lung cancer [3].
Moreover, it is also reported that in prostatic carcinoma, EGFR increased expression correlates with the higher Gleason scores and the advanced stages of the disease [4] and that EGFR activation is associated with metastatic progression and recurrence [5].
The literature revealed a controversy regarding the response of castration-resistant prostate cancer (CRPC) to EGFR inhibitors; while in phase 2 clinical trial, gefitinib showed no response as reflected by PSA level or other objective disease measures in patients with CRPC [6].In a second trial, cetuximab showed a well detected decline in PSA levels in many cases and improved patient free survival in patients with overexpressed of EGFR [7].
Mutations in downstream molecules in the signaling pathway have showed good correlation with the lack of response to cetuximab [8].The Ras/Raf/ MEK/ ERK pathway represents the most important signaling mechanism among all mitogen-activated protein kinase (MAPK) transmission pathways which play a role in signaling cascades and transmit extracellular signals to intracellular targets.Thus, it plays a crucial role in normal cell survival and the development of tumors [9,10].
The MAPK pathways generally include three main kinases, ERK1/2 is one of them with the phosphorylation of ERK can be used as a common endpoint measurement for the activation of this pathway.The extracellular signal-regulated kinases ERK1 and ERK2 are ubiquitous serine-threonine kinases that regulate cellular signaling in both normal and pathologic conditions [9,11].Since ERK1 and ERK2 are very similar, the ERK singular form is used in the current study, although the two subtypes do exist.
Increased activation of the RAS/RAF/MEK/ERK pathway has shown an association with poor prognosis and androgen independence in prostate cancer [3] Increased expression members of MAPK pathway and high levels of phosphorylated ERK1/ ERK 2 were observed in m CRPCs [3,12].So, therapeutic targeting of the MEK/ERK pathway could be a viable strategy for those with m CRPC, and trametinib (the MEK inhibitor) is currently being tested in a phase two trial for patients with m CRPC [3].
This study assesses the immunohistochemical expression of EGFR, HER2, non-phosphorylated and phosphorylated ERK in prostate acinar adenocarcinoma along with the different patients' clinicopathological parameters-in a trial to use the new target therapy of EGFR and/or MEK inhibitors in the management of advanced stage patients especially those with castrate resistant prostate cancer whom failure of other treatment modalities is encountered.

Patient samples
The study was approved by the Ethical Committees at Armed Forces Hospital (AFH), Muscat, Oman with reference number AFMS-MREC006/2020.This study included 166 prostatic acinar adenocarcinoma cases that were collected from Pathology archives starting from the beginning of 2007 till the end of 2018.

Inclusion criteria
Patients with available clinical information along with available tissue material in the blocks were included in this study.The clinical information that was extracted including; patient age, signs and symptoms, PAS level before and after treatment.The histopathological parameters were assessed including; the Gleason grade, combined Gleason score, tumor percentage, tumor stage and WHO grade group.Then, the clinical risk group was determined.As the general patient condition can affect the patient therapeutic plan, any patient with organ failure was also mentioned.As this research is targeting the castrate resistant patients, therapy that was taken was extracted in brief including its type that may be hormonal, radiological, or surgical or if any chemotherapy had been taken along with patient response by follow up PSA level to determined patient with remission and relapse or recurrence after curative surgery.

Exclusion criteria
Any patients with unavailable or minimal tissue material in the paraffin blocks-were excluded from this study.

Tissue microarray (TMA) construction
Revision of hematoxylin and eosin-stained slides of the 166 prostatic acinar adenocarcinoma cases that were extracted from the Pathology archives at Sultan Qaboos University and Armed Forces Hospital, Muscat, Omanwas done.Selection of the blocks that showed available tissue material was done.Mostly, 2 blocks for each patient were retracted (thus if the tissue material of one core was lost during preparation or was not enough, there would be another core).Control specimens were used as negative marker control including 20 cases of non-neoplastic prostatic tissue.As well as 5 cases of invasive duct carcinoma which were Her2-neu positive (score 3) and they were positive for both inactive (non-phosphorylated) and active (phosphorylated) ERK (positive control for these markers).Also, 3 cases of proliferative endometrium (as positive control for EGFR) were included.Many other non-neoplastic tissues were used for tissue microarray mapping including 3 cases of tonsillar tissue, 3 cases of colonic mucosa, 3 cases of kidney, 3 cases of skin, 3 cases of testis and 3 cases of thyroid tissue.
Then, preparation of the recipient tissue microarray paraffin block is done by cutting two or three cylindrical cores, each about 1 mm from the selected area at the donor paraffin blocks (area of non-necrotizing invasive malignancy with good cellularity-were selected by examining the corresponding slides) using Manual Tissue Arrayer MTA-1 from Estigen OU, Tiigi 61b, 50410 Tartu, Estonia.Finally, we had prepared six TMA paraffin blocks, each one showed 90 malignant and control cylindrical tissue cores arranged in specific way (according to the prepared map) to be able to code the malignant cores.

Immunohistochemistry and interpretation
Tissue sections (5 μm) were mounted on amino-acetyl silane-coated glass slides (Starfrost, Berlin, Germany), Sections are kept in hot oven over the night along with xylene for dewaxing, then descending grades of alcohol and distilled water for rehydration.Then, application of the primary antibodies for each of the studied markers (EGFR, HER2-neu, non-phosphorylated and phosphorylated ERK)-was done guided by the provided Ventana Benchmark protocol using Utra system automated monostainer (Ventana Medical Systems) (Table 1).Examination of the positive and negative control samples-was done initially followed by examination of neoplastic cores in corresponding to the designated map.
Regarding the histopathological features of the cases, the commonest and the highest Gleason pattens were 4 representing (42.1 and 45.1 %) respectively along with combined Gleason Score 7 presenting (31.9%).Table 2 showed all patients clinical and histopathological features.

Markers expression in the studied cases and their correlation with the patients and tumor criteria Regarding EGFR
Most prostatic carcinoma cases showed negative staining (91.5%).The positive cases mostly showed low expression representing about 4.9% with less evident cases showing high expression about 3.5%.(Table 3, Figure 1).No significant correlation was found between EGFR expression with any of the patient or tumor criteria.On the other hand, a significant correlation was found between negative EGFR expression and high tumor stage (Figure 2).

Regarding Her2-neu
Almost all prostatic carcinoma cases showed negative staining (98.6%) and only 2 cases representing (1.4%) showed positive staining.Thus, a statistical correlation between Her2-neu expression and the patient's or tumor criteria could not be done (Figure 1).

Regarding phosphorylated ERK
The phosphorylated ERK1/ERK2 positive immunoreactivity-was cytoplasmic and nuclear in most cases with few cases showing only nuclear stain or only cytoplasmic stain.While non-phosphorylated ERK immune reactivity was only cytoplasmic.Also, the expression of phosphorylated ERK was mostly moderate to strong stain, while, the expression of nonphosphorylated ERK was mostly weak to moderate stain.Thus, in the present study, we don't segregate cases by stain site whether cytoplasmic or nuclear or intensity.We categorized the cases into either low expression (patchy) if the positive reactivity affects less than 60% of the neoplastic cells and high expression (diffuse) if the positive reactivity affects 60% or more of the neoplastic cells.
Most of the cases showed positive cytoplasmic and nuclear staining, either patchy (48.5%) or diffuse (29.7%) (Table 3, Figure 1).There was no significant correlation between phosphorylated ERK expression and the patient criteria.While the only significant correlation found as regards tumor criteria was between the negative expression of phosphorylated ERK and the high stage of the tumor (Figure 2).

Regarding non-phosphorylated ERK
Most of the cases showed positive cytoplasmic staining, either patchy (61.5%) or diffuse (21.9%) (Table 3, Figure 1).No significant correlation was found between non-phosphorylated ERK expression and the patient's or tumor criteria (Figure 2).

Correlation of the expression of different markers
The correlation of each marker with the rest of the markers showed no significant correlation (Figure3).
Both, intensity of the stain (mild, moderate, or strong) and the extent of the positive neoplastic cells-were evaluated by two different investigators, if interobserver variation encountered, one more investigator was included.As each case was presented by two or three cores, if different stain was found among the examined cores for the same case, an average was considered.Regarding EGFR interpretation: Negative result was considered if less than 1% of the cells were positive.Then categorization of the cases was done into 3 groups: Low expression (1-10%), Moderate expression (10-50), High expression (more than 50%) [13].
In this study, the low (1-10%) and moderate expression (10-50%) groups-were categorized as patchy / low expression group along with the high expression group (more than 50%) for statistical proposal.Any membranous and / or cytoplasmic stain was interpreted as positive [14,15].
Regarding HER2/NEU, only complete strong membranous stain in more than 10% of the cells-was considered positive as in breast carcinoma cases [16].
Regarding ERK (both non-phosphorylated and phosphorylated), The negative stain was considered if no staining or less than 10% stained cells are encountered.Positive stain was considered if more than 10% of neoplastic cells were positive.Then positive results were further divided into low expression / patchy if less than 60% of the neoplastic cells were positive and high expression (diffuse) if more than 60% of the cells were positive [17].
The final interpreted immunohistochemical markers results were analyzed along with the collected patient data and the tumor histopathological features to find out any signification correlation.The patient consent was not applicable as there was no direct patient communication.

Statistical analysis
There were 166 retrospective cases underwent cross section study.SPSS program version 25 was used for study analysis.The mean, maximum and minimum and SD (or Median and IQR for non-parametric data) were used for the quantitative data.While counts and percentage were considered for the qualitative data.For comparing the quantitative variables between different groups, the Mann Whitney U test and Kruskal Wallis tests-were used.While, Chi-square test (or Fisher Exact test) was used to compare qualitative data between different groups.Statistically significant P value was considered if less than or equal to 0.05.

Clinicopathologic characteristics
In the present study, regarding the patient clinical characteristic features, most of the studied cases were older than 70 years (63%), with high serum PSA level exceeding 20 ng/ml (66.6%) and with bone metastatic disease (46.5%).Thus, most cases were considered clinically in the high-risk group (75.9%) (depending on the high serum PSA, high Gleason score and high WHO grade group as well as high tumor stage).

Discussion
This study incorporated 166 cases of prostatic adenocarcinoma for whom an immunohistochemical study were done for EGFR, HER2-neu, and the end effector ERK (both non-phosphorylated and the phosphorylated) of the cell signaling Ras/Raf/MEK/ERK pathway.All clinical and histopathological features were also analyzed with especial concentration on the cases showed castrate resistant prostate cancer or recurrence especially if associated with organ failure, for whom a new safer therapy is needed.This study was searching for the expression level of EGFR, HER2-neu and ERK for a possible role of EGFR inhibitors and/or MEK inhibitors as a therapeutic target for castrate resistant prostatic adenocarcinoma.
This study showed that most prostatic adenocarcinoma neither expresses EGFR nor HER2-neu (only 8.4% and    of those who were treated by hormonal therapy followed by radical prostatectomy [14].Di Lorenzo et al. [4] have found significant associations between EGFR overexpression and poor prognostic indicators as higher Gleason score, perineural invasion, more tissue involvement by carcinoma, and disease recurrence, thus proving a strong prognostic significance of expression of EGFR in prostatic cancer.On contrary, this result didn't find any significant correlation between the positive expression of EGFR and patient criteria or the unfavorable prognostic features of the tumor.This could be due to the few numbers of positive cases (8.4%).In concordance with our results, Back et al. [18] found no significant association of EGFR expression with other clinicopathologic parameters except its inverse correlation with androgen receptor expression.
Although the results of the current study do not support that EGFR or HER2-neu are driving molecular changes in prostatic cancer, yet the role of EGFR as a prognostic biomarker can't be ignored as Cathomas et al. [7] found that targeting EGFR resulted in a well-detected PSA decline in many cases, and improved PFS in patients with overexpression of EGFR.Further molecular studies are recommended to assess any significant pathological EGFR variant thus, patients can get benefit from EGFR tyrosine kinase inhibitors.
In the present study, most prostate cancer cases showed either patchy or diffuse positive expression for both phosphorylated and non-phosphorylated ERK with 78.2% and 83.4% respectively, In Raf/MEK/ERK signaling pathway, ERK1 represents the immediate downstream target of druggable MEK1/2 which is druggable with trametinib (an approved therapeutic agent for melanoma) [18].In concordance with our result, Nickols et al. [2] have found that patients with castration-resistant prostate cancer have higher levels of phosphorylated ERK1/2 compared to patients with untreated primary prostate cancer.Thus, therapeutic targeting of the Raf/MEK/ ERK pathway could be a valuable treatment for patients with castration-resistant metastatic prostate cancer.This hypothesis is under ongoing phase II trial tests [18].Moreover, up-regulation of ERK1/2 signaling has been shown to be one such mediator in resistant clones of previously cetuximab-sensitive cell lines (Anti EGFR tyrosine inhibitor) [19].
The main limiting factor of this study was that lack of performing molecular studies to investigate the association of positive IHC expression with molecular abnormalities and gene amplifications.Identification of the underlying gene amplification may also help in the proper selection of patients that can benefit from anti-EGFR or anti-MEK therapies.
Another important issue that limits this research is that the studied markers were expressed cytoplasmic with or without nuclear staining and segregation of nuclear stained cases wasn't done.Smith et al. [20] found that metastatic and castrate resistant prostate cancer showed increased nuclear localization rather than cytoplasmic localization of P-ERK.They suggested that p-ERK enters the nucleus in cancer cells to promote proliferation, while staying in the cytoplasm of the normal cells with no significant function effect as normal cells express lower levels of nuclear pore complex proteins and the nuclear transport factors, Thus P-ERK dissociated from nuclear entry, which is a rate-limiting step and thus cytoplasmic p-ERK positivity may not suggest functional activity of P-ERK and may not correlate with bad prognosis.Tanaka et al. [21] In the present study, we didn't segregate our cases with only nuclear staining as most cases were showing both nuclear and cytoplasmic stain.Extended larger studies correlating the IHC expression with the molecular study are needed concerning only nuclear P-ERK stain.
In conclusion, this study demonstrated that the immunohistochemical expression of EGFR and HER2neu are low in prostatic adenocarcinoma even in the castrateresistant cases.But even though, as castrate-resistant patients had no more optional therapy, the study of EGFR on both IHC and molecular levels may suggest benefit from the new EGFR tyrosine kinase target therapy.On the

Figure 3 .
Figure 3. Correlation between the Different Markers Expression; (A) EGFR with non-phosphorylated ERK and phosphorylated-ERK and (B) non-phosphorylated ERK with phosphorylated-ERK.

Figure 2 .
Figure 2. The Correlation between the Different Studied Immunohistochemical Markers with the Clinical and Pathological Features of the Studied cases (A) with EGFR, (B) phosphorylated-ERK and (C) with non-phosphorylated ERK

Table 1 .
The Staining Protocol and Antibodies information for Ventana Benchmark® Ultra system Automatic Stainer ERK

Table 2 .
[4] Summary of Clinical and Histopathological Features of the Studied Samples 1.4% were positive respectively).In concordance with our result, Back et al, 18 reported no amplification of the EGFR or HER2 genes in their studied prostate cancer specimens.On the other hand, Di Lorenzo et al.[4]found EGFR expression in 41.4% of non-metastatic prostatic carcinoma treated with radical prostatectomy and 75.9%

Table 3 .
The Immunohistochemical Results of the Examined Markers hand, non-phosphorylated ERK and phosphorylated ERK were appreciated in most studies' cases.This may suggest a promising role of MEK inhibitors.Taking into consideration the need for extended studies that concern only the nuclear ERK stain and correlate the IHC expression with molecular tests.Immune Expression of EGFR, HER2/NEU, ERK and Phosphorylated ERK in Prostate Carcinoma in post-chemotherapeutic samples is a potential predictor of the prognosis of neuroblastoma.Pediatr Surg Int.2021;37(2):287-91.https://doi.org/10.1007/s00383-020-04806-w.This work is licensed under a Creative Commons Attribution-Non Commercial 4.0 International License. other