Prognostic Significance of L1CAM Expression in Endometrial Carcinoma

Introduction
Endometrial Carcinoma (EC) is a malignant tumor of the epithelial lining of the endometrial cavity in the uterus, and it seriously threatens the health of women with high mortality rate [1]. It is regarded as one of the most prevalent cancer types among women globally and the most common cancer of the female genital tract[2]. Moreover it is the fifth most common cancer among women worldwide [3]. EC is the most common gynaecologic cancer in the developed countries and it ranked as the second common malignancy following cancer cervix in the developing countries [4]. It is considered that there are 319,500 cases of EC per year [5] and it is responsible for about 76,000 deaths per year worldwide [6].

EC is the third gynaecological cancer after ovary and cervix cancers, and it represents the 13th most known cancer in Egypt. Egypt, although having lower incidence of EC as compared with other Middle East countries, has shown an increase in the incidence over the last 12 years as it contributes to 31.4% of female genital tract malignancies according to Cancer Pathology Registry, Cairo University and National Cancer Institute, with increased mortality rate and accounts for 1.6% of total cancers in female [7]. 

The majority of EC cases occurred in postmenopausal women, but a growing proportion of younger women are also being diagnosed with EC; specifically, approximately 25% of women with EC are 

premenopausal, and 5% are diagnosed before the age of 40. All things considered, the peak incidence occurred between the ages of 60 and 70[8]. Low parity, nulliparity, early menarche, late age at menopause, age over 55, ovarian diseases like polycystic ovary syndrome (PCOS), tamoxifen therapy, chronic liver disease, obesity, and exogenous hormone use are some of the risk factors for EC [3].

Traditionally, EC has been categorised into a dualistic paradigm based on its clinical, molecular, and biological characteristics. Type I or endometroid ECs, which include 80% of cases, while type II or non-endometroid ECs are including papillary serous carcinoma, clear cell carcinoma, and carcinosarcoma [9]. 

Endometrial hyperplasia, obesity, hormone receptor positivity, and excess oestrogen have all been linked to type I (endometroid) carcinomas. Type I EC are moderately to well differentiated[10]. The majority of histologic types in type I EC are of lower grade, with a 5-year disease-free survival rate of more than 85% so women with type I EC had a favourable prognosis[11]. Type II includes less common serous, clear cell, undifferentiated carcinoma and carcinosar-coma-types and generally considered to be estrogen independent[12].  Type II EC were linked to poorly differentiated cancer and an atrophic endometrium [13]. The Cancer Genomic Atlas (TCGA) first introduced molecular classification of EC using whole genome sequencing in 2013, resulting in four distinct subtypes. These subtypes have been further studied clinically and found to translate into prognostic outcomes. Copy-number low: Nonspecific Molecular Profile (NSMP) subtype is defined by the lack of molecular subtype expression that defines each of the other three groups; prognostic outcomes are less clear within this subgroup, probably because of a high degree of population heterogeneity; further molecular stratification of the NSMP subgroup is required to better classify this group and aid in directing the proposal of adjuvant treatment. This understanding has led to additional research efforts on alternative molecular classifiers than what developed by the TCGAs [14]
L1CAM (also known as CD171) is one of the first neural adhesion molecules to be identified, it plays a crucial role in the maturation of the nervous system.  L1CAM was discovered in 1984 as a novel cell surface antigen expressed in the mouse central nervous system [15]. Because it is crucial for neuronal migration, different-iation, nerve outgrowth, axon guidance, fasciculation of axons and dendrites, myelination, and synaptogenesis, L1CAM is involved in the development of the central nervous system [16].

     L1CAM was shown to be expressed more frequently in both the primary tumor and the metastases of different kinds of cancers. By creating an immunosuppressive tumor microenvironment and increasing their resistance to endogenous death and drug-induced apoptosis, its expression gives cancer cells more survival [15]  L1CAM was frequently linked to a poor prognosis and is expressed in a variety of solid maligna-ncies. L1CAM has been demonstrated to maintain the aggressiveness of ovarian cancer (OC) tumors by promoting cell invasion, proliferation, and resistance to apoptosis. It is also necessary for the growth and spread of OC cells within the peritoneum. Lastly, there is now evidence linking L1CAM activation to OC chemoresistance [17]. A considerable risk of cancer-related mortality appears to be present in colorectal carcinomas with high expression of L1CAM, even at an early stage of the disease. Furthermore It had a connection to metastasis[18].

The prognostic importance of elevated L1CAM expression in EC is still up for debate. A minor non-endometrioid (serous, clear-cell differentiation) component, an unfavourable epithelial/mesenchymal transition, or a hidden aggressive neuroendocrine features may be connected to the identification of L1CAM in endometrioid endometrial cancer.[19]. It has been demonstrated that having a positive L1CAM is highly correlated with a poor prognosis and aggressive EC. Nonetheless, It has been shown that L1CAM expression was related to a bad prognosis, but only in women with endometroid EC and not in non-endometroid EC patients[20]. 

Material and Methods
Patients selection criteria:
This is a retrospective study included 52 formalin fixed, paraffin embedded EC tissue blocks.  These tissue blocks were collected from Minia University's patho-logy department archive during the period between April 2021 and April 2023.

Cases included 50 cases of endometroid adenocarcinoma and two cases of non-endometroid EC. All cases have been examined for L1CAM expression. Two cases of non-endometroid type are exclu-ded from the statistical analysis and interpreted separately. 

The available clinicopathological data were obtained from the pathology reports of the cases and from patient's data files. These data include: Patient age, menopausal state, tumor size, site, grade, histological sub-types, lymphovascular invasion (LVI), tumor necrosis, tumor infiltration of the cervical stroma.  Patients and tumor characteristics were listed in table (1). The histopathological classification of the tumors was performed according to the WHO 2014 classification of endometrial tumors[21]. Cases were graded according to 2009 FIGO grading criteria, using the 3-tier system. Then binary FIGO grading system was applied, in which  FIGO grade1 and 2 tumors  are categorized as low grade and FIGO grade 3 tumors as  high grade [22].

Immunohistochemistry:
Five µm sections were prepared on positive charged slides for immunohistochemical staining using the primary antibody for L1CAM (Rabbit Monoclonal antibody 100 ul concentrated (1ul/ml). According to the manufacturer data sheet (BIOSS ANTIBDIES Company) utilizing the avidin biotin-peroxidase complex method with diaminobenzidine (DAB) chromogen detection system. Tissue sections were first deparaffinized and rehydrated on the positively charged slides. After that, the endogenous peroxidase was inhibited by submerging it in a 3% hydrogen peroxide solution and waiting 30 minutes for it to incubate. For antigen retrieval, the slides were submerged in a citrate buffer solution (pH 6) twice for ten minutes each at 750 W. 
The slides were treated by UV block to prevent non-specific background staining.  Primary antibody L1CAM was then added, and tissue sections were incubated for 1 hour at room temperature (dilution 1:100). After removing the extra reagent, the slides were gently washed for five minutes with buffer solution. Subsequent biotinylated antibody was then applied and maintained on each slide for half an hour. DAB substrate and chromogen. The Positive control for L1CAM was human kidney tissue.

Interpretation of immunohistochemical staining: 
L1CAM was expressed mainly in the cell membrane. Occasional weak cytoplasmic expression was detected in some cases. The percentage of positive tumor cells determined the score for L1CAM expression (score 0 = 0%, score 1 = 1–10%, score 2 = > 10–50%, and score 3 = > 50%). Tumors were identified as L1CAM positive if >10% (score 2 and 3) of the epithelial tumour cells exhibited membranous L1CAM staining. [23].

Statistical analysis:
Statistical analysis was conducted using the Statistical Package for Social Sciences (IBM SPSS software version 25), that used to analyze the data. Clinicopathological 
characteristics will be described by descriptive analysis which includes the means, standard deviations (SDs), median. For qualitative data, the data were reported as both numbers and percentages, and either the Fisher`s exact test or the Chi-square test was used to assess them. A p-value of 0.05 or less was considered significant. 

Results
In this study, regarding endometroid subtype cases, 19 out of 50 cases (38%) showed positive L1CAM expression, while 31 cases (62 %) showed negative L1CAM expression. Association between L1CAM expression and clinicopathological data for cases of endometroid type is shown in table (2). 

As regard to tumor grade, a statistically significant association was found between L1CAM expression and tumor grade (p = 0.01). 12 /18 (66.7%) high grade EC cases showed positive L1CAM expression, while 7/32 (22%) low grade EC cases showed positive expression. Also, statistically significant association between L1CAM expression and myometrial invasion (p = 0.001) was detected, as 12/17 (70.6%) cases that showed infiltration more than half of the myometrium showed positive L1CAM expression. In addition, statistically signi-ficant association was found between L1CAM expression and LVSI (p < 0.04).  7 out of 11 cases (63.6%) that showed LVSI showed positive L1CAM expression while only 12 out of 39 cases (30.8%) without LVSI had positive L1CAM expression. Figure 1 (A-E). Furthermore, a statistically significant positive association was found between L1CAM expression and both menopausal state and cervical stromal invasion (p= 0.02, p= 0.002 respectively). No significant associations were found between L1CAM expression and patients’ age, tumor necrosis, tumor site and tumor size (p = 0.7, p = 0.4, p = 0.4 and p = 0.2 respectively).

Regarding non-endometroid cases, there are two non-endometroid cases are included in this study. One case of serous subtype and one case of clear subtype. Concerning serous subtype case, patient characteristics showed post-menopausal state, tumor grade III, tumor stage III, invasion of less than one half of myometrial thickness, positive cervical stromal invasion, No lympho-vascular invasion, and absent necrosis. For L1CAM expr-ession, this case showed high positive L1CAM expression (>50%). As to clear cell subtype case, patient characteristics showed post-menopausal state, tumor grade III and tumor stage IV, invasion of more than one half of myometrium, positive cervical stromal invasion, positive lympho-vascular invasion and presence of necrosis. For L1CAM expression, high positive L1CAM expression (>50%) was detected. Figure 2 (A&B). 

In our study, risk stratification scheme according to (ESGO), (ESTRO/ ESP) guidelines for EC cases revealed that 17 cases were classified as low risk, 15 cases were classified as intermediate risk, 9 cases were intermediate to high risk, 7 cases were high risk, and two cases were advanced risk. A significant association was found between L1CAM expression and higher risk groups (p = 0.003).  85% and 100% of high risk and advanced risk group respectively were L1CAM positive, while only 17.6 % of low-risk groups were L1CAM positive, as shown in table (3).