Expression of HER2/neu in gastric adenocarcinoma and its correlation with serum HER2/neu level and E-cadherin expression.

AIM: : To assess HER2/neu expressions and correlate with E-cadherin and Serum HER2 level in gastric carcinoma. METHOD: 31 gastric biopsies and 1 resected specimen were taken in the study with patient details and stained with H and E for histopathological details following Lauren's classification. Immunohistochemistry for HER2 and E-cadherin expression was conducted followed by serum HER2/neu ELISA. RESULT: Adenocarcinoma with 61% diffuse, 29% intestinal, and 10% other type were observed with predominant HER2 immunoexpression in intestinal-type than in diffuse-type adenocarcinoma. Other observations marked 44% as 3+/positive and 56% as 2+/equivocal in intestinal type while 26% cases as 3+/positive, 69% as 2+/equivocal, and 1% as 1+/negative were observed in diffuse type. The data presented 33% membranous positivity and 67% both membranous + cytoplasmic positivity in intestinal type while 2% showed membranous positivity, 47% both membranous + cytoplasmic, and 42% only cytoplasmic positivity in diffused type. On comparing the localization pattern of HER2 and E-cadherin, 25% of cases showed membranous staining while 50% of cases showed membranous with cytoplasmic staining for both. No cytoplasmic HER2 staining as well as no any staining for E-cadherin was shown by 6% cases. CONCLUSION: Thus, it can be concluded that cytoplasmic expression of HER2 in gastric adenocarcinoma (mainly diffuse type) may be due to shedding of its extracellular domain, leading to loss of membranous E-cadherin expression on immunohistochemistry. The loss of membranous expression of E-cadherin and increased serum HER2 ELISA were correlated well with these findings.

Artificial intelligence technologies empowering identification of novel diagnostic molecular markers in gastric cancer.

In recent clinical practice the molecular diagnostics have been significantly empowered and upgraded by the use of Artificial Intelligence and its assisted technologies. The use of Machine leaning and Deep Learning Neural network architectures have brought in a new dimension in clinical oncological research and development. These algorithm based software system with enhanced digital image analysis have emerged into a new branch of digital pathology and contributed immensely towards precision medicine and personal diagnostics. In India, gastric cancer is one of the most common cancers in males as well as in females. Various molecular biomarkers are associated with gastric cancer development and progression of which HER2 protein, a transmembrane tyrosine kinase (TK) receptor of epidermal growth factor receptors (EGFRs) family is of prime importance. The EGF receptor expression in gastric cancer is linked with its prognostics and theragnostics. These expressions are assessed by immunohistochemistry (IHC) and molecular techniques such as Fluorescence in-situ hybridization (FISH), as per recommendations for HER2 targeted immunotherapy. These have motivated the software giants like Google Inc. to produce innovative state of art technologies mimicking human traits such as learning and problem solving skill sets. This field is still under development and is slowly evolving and capturing global importance in recent times. A literature search on PubMed was performed to access updated information for this manuscript.