Prognostic impact of EGFR expression and immunohistochemistry-based "molecular classification" in bladder cancer

Recent genomic studies emphasize the necessity of molecular classification to reflect diverse clinical and pathological characteristics of bladder cancer. Immunohistochemically bladder cancer can be classified into molecular subtypes, including basal, luminal, and p53-like subtypes. Epidermal growth factor receptor (EGFR) is frequently expressed in basal-type bladder cancers and is associated with poor prognosis. In our study, 88 urothelial carcinoma cases were retrospectively analyzed, molecularly subtyped using CK5/6, GATA3, p16 immunohistochemistry and examined for EGFR expressions as well as clinical and histopathological features. Tumor cell scores ≥20 % considered positive, classifying cases as luminal (GATA3-positive), basal (CK5/6-positive), double-positive (both-positive), or double-negative (both-negative). Further division of luminal and basal cases was based on p16 status: luminal-p53 or basal-p53 (p16-positive) and luminal-non-p53 or basal-non-p53 (p16-negative). Among the cases, 4 (4 %) were double-negative, 48 (55 %) luminal-non-p53, 21 (24 %) luminal-p53, 5 (6 %) basal-non-p53, 3 (3 %) basal-p53, and 7 (8 %) double-positive. Our findings revealed that basal-non-p53 type bladder cancer is associated with poor prognosis, muscle invasion, and high-grade cytology. Basal-p53 and double-negative types exhibited less aggressive features compared to basal-non-p53 types, with associations observed with lamina propria invasion and high-grade cytology. Luminal-p53 type demonstrated higher recurrence rates. Luminal-non-p53 type displayed the least aggressive characteristics, often associated with papillary histopathology. EGFR expression was found to be high in basal-non-p53 type and was further correlated with adverse prognostic indicators, lamina propria invasion, and high-grade cytology. The identification of molecular subtypes and EGFR expression through immunohistochemistry, alongside traditional bladder cancer classifications, enhances tumor behavior prediction and supports effective clinical management.