ERBB2 (HER2) fluorescence in situ hybridization (FISH) guides therapy selection for solid tumors, including breast and gastric cancers. While laboratory technologists typically perform the initial enumeration of HER2 FISH, laboratories differ in the extent of pathologist review, ranging from full slide reassessment to review of representative captured images. By performing a retrospective analysis of ∼10,000 HER2 FISH cases scored using breast and gastric criteria, the objective was to define decision boundaries at which technologists' and pathologists’ interpretations converge, enabling a data-driven approach to triaging HER2 FISH sign-out. Results showed that agreement between technologists' and pathologists’ interpretations was highest when technologists’ counts were well beyond the diagnostic cutoffs for the HER2/CEP17 ratio and the number of HER2 signals per cell. Conversely, cases near diagnostic thresholds showed increased discordance, emphasizing the need for full review of the FISH slide in this subset. Individual technologists differed in their percent discordance with pathologists’ final results; however, this was not explained by case volume (experience) alone and did not consistently improve over time. In summary, distinct HER2 FISH technologists HER2/CEP17 ratio and HER2 signal number per cell values were identified where concordance with pathologists results exceeded 99%, suggesting cases in these ranges may require only limited pathologist slide review, potentially allowing for more efficient triaging and sign-out of HER2 FISH.
Extracellular matrix remodeling and angiogenesis are important events in the evolution of cystic and neoplastic lesions. This study aimed to assess and compare the differential expression of extracellular matrix metalloproteinase inducer (EMMPRIN) and angiogenic index in radicular cysts (RCs), dentigerous cysts (DCs), and odontogenic keratocysts (OKCs). Twenty RCs, 20 DCs, and 20 OKCs were evaluated by immunohistochemistry. The expression of EMMPRIN was evaluated in the epithelial component and in endothelial cells of vessels in the fibrous capsule of the lesions. The angiogenic index was determined by microvessel count (MVC) using anti-CD105 immunoexpression. Immunoexpression of EMMPRIN was observed in most RCs (95%) and in all DCs and OKCs. Compared with RCs and DCs, OKCs revealed lower immunopositivity for EMMPRIN in the epithelial component ( P <0.001). In endothelial cells, EMMPRIN expression was higher in OKCs than in RCs ( P =0.010) and DCs ( P =0.007). RCs had the highest angiogenic index, with a statistically significant difference compared with OKCs ( P =0.018). A positive correlation was observed between MVC and immunoexpression of EMMPRIN in endothelial cells of vessels in RCs ( r =0.654; P =0.002). The results suggest an important role for EMMPRIN in the development of RCs, DCs, and OKCs. The higher expression of this metalloproteinase inducer in endothelial cells of vessels in the fibrous capsule of OKCs could be related to the aggressive biological behavior of these lesions. In RCs, EMMPRIN seems to act as a proangiogenic factor.
In recent years, the emergence of newly characterized entities with eosinophilic/oncocytic cytoplasm has posed significant diagnostic challenges, particularly due to the overlapping morphologic features of chromophobe renal cell carcinoma (ChRCC) and renal oncocytoma (RO). In this study, we analyzed the clinicopathological and immunohistochemical (IHC) characteristics of 49 eosinophilic renal tumors to determine whether GATA3 expression could aid in the differential diagnosis of these neoplasms. On the basis of integrated morphologic and IHC evaluation, the tumors were reclassified into the following diagnostic categories: ChRCC (n=15), RO (n=13), unclassified eosinophilic/oncocytic renal cell carcinoma (U-EO-RCC; n=9), low-grade oncocytic tumor (LOT; n=5), and eosinophilic clear cell renal cell carcinoma (E-CCRCC; n=3). In addition, the cohort included single cases of papillary renal neoplasm with reverse polarity (PRNRP), unclassified RCC with MiT family translocation-like features (U-MiTFL-RCC), SDHB-deficient RCC, and FH-deficient RCC. Notably, one of the U-EO-RCC cases was further characterized by the presence of a TSC mutation. All LOT cases, one SDHB-deficient RCC, one FH-deficient RCC, one PRNRP, and one of fifteen chromophobe RCC’s had diffuse/strong staining with GATA3. In contrast, all E-CCRCC (n=3), renal oncocytoma (n=13), and U-MiTFL-RCC (n=1) cases demonstrated complete GATA3 negativity. In conclusion, GATA3 serves as a valuable diagnostic adjunct in the differential diagnosis of emerging eosinophilic renal tumor entities. However, due to its expression across diverse subtypes, GATA3 results must be interpreted within the morphologic context and as part of a comprehensive immunohistochemical panel—including K7, K20, CD117, vimentin, FH, SDHB, and cathepsin K—to avoid potential diagnostic pitfalls and ensure the accurate identification of clinically significant syndromes.
Renal oncocytoma (RO) is a benign renal epithelial tumor that may closely mimic chromophobe renal cell carcinoma (chRCC), particularly in limited biopsy specimens. In addition, the expanding spectrum of oncocytic renal neoplasms further complicates the differential diagnosis. This study included 251 renal epithelial tumors, of which 250 were included in statistical analyses. The cohort comprised RO, chRCC, papillary RCC, clear cell RCC, collecting duct carcinoma, multilocular cystic renal neoplasm of low malignant potential, and a single case of eosinophilic vacuolated tumor (EVT). Tissue microarrays were constructed from formalin-fixed, paraffin-embedded tissue, and immunohistochemical expression of ZEB1, CK7, CD117, AMACR, CD10, GATA3, SDHB, Cathepsin K, and FH was assessed using H-score analysis. Statistical evaluation included 1-way ANOVA, receiver operating characteristic (ROC) curve analysis, and binary logistic regression. RO demonstrated significantly higher ZEB1 expression compared with all other tumor subtypes ( P <0.001), while ZEB1 expression was absent in chRCC. The single EVT case, excluded from statistical analysis, also showed no ZEB1 expression. In contrast, CK7 expression was highest in chRCC and minimal in RO. ROC analysis demonstrated excellent diagnostic performance for both markers (ZEB1 AUC=0.941; CK7 AUC=0.945), and combined analysis showed strong discriminatory power, with an overall classification accuracy of 95.5%. ZEB1 represents a promising complementary marker in the differential diagnosis of RO and chRCC. In combination with CK7, it provides a simple and practical immunohistochemical approach that may improve diagnostic accuracy, particularly in limited biopsy specimens. Further validation across the expanding spectrum of oncocytic renal neoplasms is warranted.
In colorectal carcinoma (CRC), deficient DNA mismatch repair (dMMR) is an established predictive and prognostic biomarker and an essential screening marker for Lynch syndrome. Immunohistochemistry (IHC) for MLH1, PMS2, MSH2, and MSH6 is routinely used to assess expression of mismatch repair (MMR) proteins. Strong, homogeneous nuclear staining in tumor cells supports MMR proficiency (pMMR), whereas the absence of tumor nuclear staining in the presence of intact internal controls supports dMMR. Loss of protein expression typically occurs in heterodimeric pairs, most commonly MLH1/PMS2 or MSH2/MSH6, although isolated loss of PMS2 or MSH6 may also occur. A clinically important interpretive pitfall is the punctate or dot-like nuclear staining pattern observed with the VENTANA anti-MLH1 M1 clone. This aberrant pattern represents MLH1 loss but may be misinterpreted as retained MLH1 expression, particularly by pathologists unfamiliar with its significance. We evaluated 24 paired biopsy-resection CRC cases, including 11 MLH1/PMS2-deficient cases in which biopsy specimens showed punctate MLH1 staining with the M1 clone. In the corresponding resection specimens, 3 distinct MLH1 staining patterns were identified: pattern 1, complete loss of nuclear staining throughout the tumor; pattern 2, punctate staining in superficial tumor cells with complete loss in deeper tumor regions; and pattern 3, punctate staining throughout the tumor. Repeat MLH1 IHC using the BioCare Medical G168-15 clone confirmed that punctate nuclear staining with the VENTANA M1 clone corresponds to true MLH1 protein loss. Four cases were initially misinterpreted by the original pathologists, underscoring the clinical impact of this artifact. Our findings indicate that dMMR staining patterns, particularly aberrant MLH1 staining, are more susceptible to interpretive error than conventional retained staining patterns. Recognition of punctate MLH1 staining as a manifestation of MLH1 loss is essential to prevent false MMR-proficient interpretation and to ensure appropriate molecular workup, Lynch syndrome triage, and therapeutic decision-making.
