Frequency of HER2 Expression, MMR Deficiency, and PI3KCA Mutation in Pretreated Surgical Specimens of Patients with Esophageal Squamous Cell Carcinoma in Iran Cancer Institute

Background & Objective: Iran is located in the esophageal cancer geographical belt. As multiple genetic alterations are responsible for the molecular pathogenesis of esophageal squamous cell cancer (ESCC), the role and frequency of HER2 expression, MMR deficiency, and PI3KCA mutation are not well defined. Methods: We carried out HER2/neu expression, dMMR/MSI high, and PI3KCA mutation analysis in specimens of patients with ESCC. We accessed archival tissue blocks related to specimens of 68 ESCC cases at the time of surgery following neoadjuvant chemoradiation. These patients underwent surgery during 2013-2018 at the Cancer Institute of Iran affiliated with the Tehran University of Medical Sciences in Tehran. Results: None of the patients showed HER2 expression, dMMR/MSI high, or PI3K mutations. Conclusion: dMMR/MSI-H and PI3KCA mutation and HER2 expression may not be reliable andfrequent targets for systemic therapy in patients with esophageal SCC.


Introduction
Cancer, as a chronic, costly, and life-threatening disease, is one of the most critical health challenges in the world. Esophageal cancer is considered the eighth most common cancer and the sixth most frequent cause of death worldwide. In 2018, according to GLOBOCAN data, the global rate of esophageal squamous cell carcinoma (ESCC) incidence and death were 572,000 and 509,000, respectively (1). As a multifactorial disease, the development of gastric and esophageal cancers is influenced by both genetic and environmental factors. Therefore, its incidence is expected to vary throughout the world. The reports have indicated that the disease is relatively common in Iran. In a report by World Health Organization, esophageal cancer is the ninth most common malignancy in Iran (2). What is often referred to as the "esophageal cancer belt" originates from northern Iran and passes through Central Asia to reach northern central China (3). Reports from the Cancer Registry of the International Agency for Research on Cancer indicated that the prevalence of esophageal cancer in northern Iran was 165 per 100,000 male individuals and 195 per 100,000 female individuals (2). Squamous-cell carcinoma (SCC) is the predominant subtype of esophageal cancer in Iran (4).
The overall 5-year relative survival of esophageal cancer between 2012-2018 in SEER geographic areas is 20.6% (5). Systemic therapy has limited effectiveness in metastatic esophageal cancer, with responses observed in only 20-40% of the population, and a median survival time of around 9 months (6). Based on the literature, 5-year survival in ESCC is low, and cancer metastases, rather than primary cancer, are the main cause of as high as 90% of cancer-related deaths (7). Metastases are the final step of a cellular biological process termed the invasion-metastasis cascade. This process involves enhanced cell mobility and intravasation, transit in lymphatic and blood vasculature, extravasation, and finally growth at a new location (7,8). The identification of the biological process and genetic pathways that cause invasion and metastasis will facilitate the development of new treatment modalities. This study aimed to evaluate some possible pathways and mutations that contribute to the development of invasion and metastasis of ESCC.
Latest advances in gene profiling techniques and the development of metastatic cell models have been crucial in the identification of genes and pathways associated with tumor metastases (9). Therefore, to improve the risk-adapted therapeutic strategies and stratify patients to enroll in future clinical trials on actionable molecular pathways in ESCC, we need the identification of new prognostic or predictive molecular markers. The reports indicated targets to treat malignancies, such as breast, stomach, and esophagus (10,11).
The epidermal growth factor receptor family comprises four homologous molecules, including EGFR, c-erbB-2, cerbB-3, and c-erbB-4. A study suggested that the abnormal activation of kinase activity in these receptors has a pivotal role in the development and progression of ESCC (12). The tyrosine kinase function of these receptors is essential for intracellular signaling and cell transformation in the proliferation, differentiation, and maturation of the embryonic intestine (13,14). Although the detection of gene amplification might be a mainstay factor in decisionmaking for treatment, data are scarce regarding HER2/neu amplification in patients with ESCC.
Phosphatidylinositol 3-kinases (PI3K) are a ubiquitous lipid kinase family that catalyzes the phosphorylation of molecules phosphatidylinositol (PI), PI (4) P, and PI (4, 5) P2 into PI (3) P, PI (3, 4) P2, and PI (3, 4, 5) P3, respectively (15). These products of lipid entity can then activate a variety of downstream events that regulate a wide spectrum of essential cellular processes, including growth and cell cycle progression, apoptosis and migration, metabolism, and vesicular trafficking. This overactivation might result in cell function dysregulation by interrupting apoptosis and inducing proliferation, eventually triggering tumor formation (16).
Microsatellite instability (MSI) is identified by extensive somatic alterations in the nucleotide repeat sequences (microsatellites), which is almost always a reflection of a germline mutation in any of the mismatch repair (MMR) genes, mostly MLH1 and MSH2 (17). In colorectal cancer, MSI is associated with clinicopathological characteristics, such as proximal location and poor differentiation, low metastasis potential, and eventually a better prognosis (18). Based on the previous data about these targets in the treatment of patients with ESCC, the present study attempted to assess the status of MSI phenotype and HER2 expression and mutation in the PI3K pathway in ESCC in Iranian patients.

Study Design
This retrospective study was carried out on the esophagectomy samples obtained from patients with ESCC at the time of surgery. The patients attended the Cancer Institute of Iran affiliated with the Tehran University of Medical Sciences in Tehran, Iran, during 2013-2018.

Inclusion and Exclusion Criteria
In this study, patients without gender and age restrictions were evaluated. We included specimens with SCC histology in surgical specimens of patients after neoadjuvant chemoradiotherapy. Those with complete response to preoperative treatment were kept out due to a lack of any malignant tissue to perform adequate testing.

ESCC Cases and Specimens
First, the records of patients with esophageal cancer extracted from the archives of the Radiation Oncology Department of Imam Khomeini hospital in Tehran, information about the stage of the disease, grade and survival rate, neoadjuvant treatment of tumor, location based on distance from incisors according to endoscopy or endoscopic ultrasound or computed tomography scan, and other characteristics were registered in the designated data forms. The TNM Classification of Malignant Tumors staging was determined according to the surgical pathology reports. In this study, we utilized the archived formalin-fixed paraffin-embedded blocks from the tumor tissues of 68 ESCC patients at the time of surgery.
The used samples included the tissues prepared during surgery from patients with esophageal cancer. The standard and routine method used in the pathology section was employed to prepare the tissue, which briefly includes several steps, namely the preparation of appropriate initial tissue incision, fixation, and processing with a tissue processor, performing appropriate incisions using a microtome, and preparing slides, dying slides, and assembling them. The tumor site on the slide was marked under a microscope and matched with the paraffin block.
Most of the cases received neoadjuvant chemoradiation and chemotherapy. In this setting, only those with incomplete responses to neoadjuvant therapy could be evaluated because there was no remnant tumor in cases with complete response, and there was no access to the preneoadjuvant biopsy samples. Gender, primary-tumor site, histological grade and the extent of normal tissue infiltration, lymphatic and venous invasion, and surgical margin and pathological stage, tumor regression grade (TRG), The sections from tissues were then de-paraffinized and incubated in a pressure cooker containing citrate buffer for retrieval of antigens. In the next step, the activity of endogenous peroxidase was inhibited by a 3% hydrogen peroxide solvent. The sections were then exposed to polyclonal primary antibodies against HER2/neu (1:500, A0485, DAKO, Denmark). In the following step, the sections were incubated in a secondary biotinylated antibody of an LSAB+ peroxidase Kit (DAKO, K0690, Denmark), and then incubated with streptavidin HRP (DAKO, Denmark), and were counterstained using hematoxylin dye. Immunohistochemical analyses of HER2 expression describe the positivity of tumor cells and their staining pattern.
Deoxyribonucleic acid (DNA) isolation and polymerase chain reaction (PCR) reactions were required to detect the presence of mutations in PI3K. The paraffin block was selected for two cuts from the marked tumor area using a scalpel and inserted into a microtube for each tissue sample. New blades were used to cut each tissue to prevent the carry-over effect. QIAGEN Gene Read DNA FFPE Kit (QIAGEN, Germany) was utilized to extract DNA based on the instructions. After extraction, DNA quality was evaluated by absorbance ratios at 260/280 nm. Table 1 shows primer sets and amplification programs. Amplification was performed on a LightCycler 96 System (Roche, Basel, Switzerland). The PCR products were sent to Kowsar Biotech Company (KBC®, Tehran, Iran) to sequence amplified gene fragments using Sanger sequencing. The results included electropherogram files provided by the company to the researcher. Finally, the presence of a gene mutation was determined using gene alignment on the peak scanner software and comparing the obtained sequence with the standard PIK3CA sequence.
In this study, there were three hotspot mutations in the PIK3CA gene, including E542K and E545K in exon 9 and H1047R in exon 20 by Sanger sequencing after PCR amplification. As shown in Figure 1, the sequences of GAA at 1624, GAG at 1634 in exon 9, and CAT at 3139 and 3140 in exon 20 were wild-type (normal), and no point mutation was recorded.
DNA isolation and PCR reaction were required to assess the MSI status. QIAGEN Read TM DNA FFPE Kit was used to extract tissue DNA, and the extraction process was performed according to the kit instructions. After extraction, DNA quality was evaluated by absorbance ratios at 260/280 nm.
For performing the PCR reaction, suitable primers were used to amplify the number of copies of five human DNA microsatellite sequences based on the Bethesda panel of five microsatellite markers. Forward and reverse primers for detection of the amplifying sequences NR27, BAT26, BAT25, NR24, and NR21 were requested from Macrogen (Geum Chun-Gu, Seoul, Korea) with their already defined sequences (19). The PCR reaction was performed in a LightCycler Nanomachine (Roche Diagnostics GmbH, Mannheim, Germany).
The melting graphs were evaluated, and the samples were delivered for capillary electrophoresis if acceptable. Electrophoresis results were analyzed by peak scanner software (version 1.0; Applied Biosystems, CA, USA). Detected alterations in two or more markers were considered MSI-H, while gene mutations in only one marker were considered as MSI-L. PCR protocol: 50 cycles, 95˚C for 15 minutes, 95˚C for 20 seconds, 60˚C for 20 seconds, 72˚C for 20 seconds, and 72˚C for 5 minutes

Statistical Analysis
There was no power analysis for this study, and all accessible specimens underwent laboratory testing to detect desired mutations or expressions. The quantitative variables were presented using central tendency parameters. The qualitative variables were expressed using frequency and proportion. Median survival values in terms of months were expressed using the Kaplan-Meier method. SPSS software version 20 (SPSS Inc., Chicago, Ill., USA) was used for statistical tests.

Results
This study examined 68 patients with ESSC and one patient with gastroesophageal junction SCC. Table  2 shows the characteristics of the patients. Almost all patients underwent esophagectomy. Two patients did not undergo esophagectomy, including one with partial gastrectomy and one who was biopsied with suspicion of interval metastases. According to the operative reports, 17 (25%) and 49 (72%) patients had intermediate (ypT1-2 or N0) and poor (ypT3-4 or N+) pathologic responses to the preoperative treatment, respectively. In other words, based on the modified Ryan system, among those who underwent tumor resection, there were 5 (7.6%), 34 (51.2%), and 27 (40.9%) patients with TRG 1, 2, and 3, respectively. Moreover, 5 (7.4%) patients had local recurrence, and 36 (52.9%) patients had distant metastases. The liver was the most common site for distant metastases in 18 (26.5%) subjects.
The median follow-up of the surviving patients was 57 months using the reverse Kaplan-Meier method. The median overall and disease-free survival were 25 and 18 months, respectively. Accordingly, the 5-year rates of overall survival and disease-free survival were 24% and 17%, respectively. In this study, 68 samples were undergone to IHC evaluation, and none were observed to be HER2 overexpressed. This study examined 39 out of 68 patients for PI3K mutation and MSI status. None of the patients had high dMMR/MSI or PI3K mutations.

Discussion
At the global level, the mortality of esophageal cancer is projected to increase among male and female subjects, more evident in less developed countries (20). There is wide geographical variation in the incidence of ESCC throughout the world. This fact might indicate that ESCC has a heterogenous and diverse pattern in its molecular and clinical manifestations (21). Various molecular pathways lead to cancer in various populations. The evaluation of mutations in diseaserelated genes is one of the practical tools for identifying diseases and predicting their probability of occurrence. Additionally, to have an opportunity for individualized treatment, identifying molecular characteristics of the patients' tumor is of utmost importance to refine outcomes and keep down the exposure to unnecessary toxicities.
Some known oncogenic pathways can be observed by referring to the Kyoto Encyclopedia of Genes and Genomes pathway database (22). However, this database has not registered esophageal cancer yet, indicating the importance of further research on the molecular status of esophageal cancer.
The analysis of IHC studies showed that 8.6% of the cases were positive for the expression of HER2 (23). In a few studies that have so far reported the frequency of HER2 expression in ESCC, there is a considerable discrepancy between 0% and 64% (20)(21)(22). This variability might have resulted from differences in the patient selection based on ethnicity and genetic background, preoperative treatment status or stage of presentation, the variations in opted IHC techniques or antibody sources, or even various criteria for evaluating the expression. This study used IHC to detect HER2 expression; however, the detection of HER2 gene amplification using fluorescence in situ hybridization might increase accuracy and make useing of anti-HER2 targeted therapies more beneficial.
The frequency and the clinical relevance of MSI in ESCC are not well recognized. There are discrete and even contradictory MSI studies in the ESCC population. The reported rate of MSI in ESCC largely differs among the previous investigations from 0% to 40% (24). However, relatively high levels of MSI-L status were seen in ESCC, compared to those of colorectal or gastric cancers (25). The MSI is considered a predictive marker of tumor response to immunotherapy and can be essential to examine the status of this genetic instability mechanism in metastatic ESCC. In this study, there were no cases of MSI. Two standard recommended methods are there to detect tumor MMR deficiency, namely PCR and loss of MMR protein expression by IHC (26). The PCR methods and criteria used to determine the MSI phenotype are pentaplex and HT17 repeat. A 17 mononucleotide repeat of HSP110 (HT17) is critical for correct splicing of this chaperone and improving current standard molecular methods to detect MSI in colorectal cancer (27). The pentaplex PCR detection of MMR protein expression was used in this study.
The PIK3CA mutation has been reported with various frequencies, and its association with prognosis has not been consistent in available studies. In ESCC, although PIK3CA mutation has been detected in 2% to 12% of studied cases, its prognostic or predictive role is still unknown (10). Several reasons might explain the difference between the observations of the present study and other studies. First, in the current study, the Sanger sequencing method was used for the mutational analysis of PIK3CA. A comparison study by Arsenic et al. showed that next-generation sequencing has superior sensitivity in detecting PIK3CA mutation over the Sanger sequencing technique (28).
In addition to the above-mentioned issues, lack of access to the tissue samples before neoadjuvant therapy (due to the initial referral of patients to other medical centers) led to the exclusion of samples with a complete response to preop treatment from the study that roughly comprised 30% of all ESCC patients. However, the omission of these samples could not be a causative factor for such low rates of HER2 overexpression, PI3K mutation, and MSI-H status in the present study, compared to those of other studies. The present data indicated that none of the patients with ESCC had HER2 amplification and/or MSI-H status or PI3KCA mutation in Iran; nevertheless, this topic in ESCC requires further investigation.

Conclusion
The results of the current study indicated that dMMR/MSI-H and PI3KCA mutation and HER2 expression are unreliable, frequent targets for systemic therapy in patients with ESCC. Other investigations are encouraged to detect other applicable mutations .