An IHC Screen For EphA3 Positive FFPE Tumors

Eph receptors are the largest subgroup of the receptor tyrosine kinases (RTK). Unlike other RTK, these function principally during development. Quiescent in postembryonic tissues, Eph expression by adult tissues is abnormal and implicated in tumor initiation, invasion and metastasis.

Aberrant EphA3 expression is seen in solid and hematologic tumors, particularly advanced stage lymphoproliferative and myeloproliferative diseases. In this regard, targeting EphA3 may constitute a novel treatment for hematological malignancy.

With clinical utility in mind, i.e., patient selection for anti-EphA3 therapy, a panel of commercial and proprietary (KaloBios) antibodies was screened by Western Blot for reactivity to recombinant Eph receptors (EphA3, EphA4, EphA5, EphA7, EphB2, EphB3).

Those with EphA3 binding selectivity were further screened (QualTek) by immunohistochemistry (IHC) using formalin-fixed paraffin-embedded (FFPE) normal and diseased human bone marrow (NLBM, AML) as well the LK63 pre-B-ALL cell line from which EphA3 was originally isolated.

Different tissue pretreatments were used for antigen/epitope retrieval of EphA3, including steam-heating in citrate-based or Tris & chelator-based buffers, subsequent/or protease digestion, or neither.

Following each antigen/epitope retrieval procedure, antibody reactivity for EphA3 was assessed by light microscopy using enzymatic biotin, tyramide and polymer-based detection techniques. In each instance, the location of the EphA3/antibody complex was visualized with 3,3-diaminobenzidine that precipitates a discrete insoluble reaction product in presence of enzyme (HRP). Nuclei were counterstained with hematoxylin to assess cell/tissue morphology.

From this screen, one antibody (with an epitope in the cytoplasmic CT-domain of EphA3) was chosen for further assay optimization based on its selective reactivity towards LK63 and AML and low selective reactivity towards NLBM.

Assay optimization included, amongst other aspects, evaluation of EphA3 expression in a panel of NLBM and hematopoietic tumors (200+ specimens inclusive of acute & chronic leukemia, peripheral T-cell & B-cell neoplasm, Hodgkin & non-Hodgkin lymphoma, multiple myeloma, myelodysplasic syndrome, myeloproliferative neoplasm) by a board-certified hematopathologist.

In NLBM, the frequency of EphA3+ immature-blast cells (CD34+ or CD117+) was insignificant. Less than 10% CD34+/CD117+ cells were EphA3+.

Elevated EphA3 expression (percentage EphA3+ nucleated cells) was observed in most hematopoietic tumors.

For example, in multiple myeloma, tumor cells were typically EphA3+/CD138+ plasma cells. For AML, leukemic CD34+ or CD117+ blasts/initiating cells were typically EphA3+. Some CD138- plasma cells or leukemic CD34-/CD117- cells were also EphA3+.

Correlation was made between EphA3 expression and specific tumor maturation stages, differentiation status and/or tumor aggressiveness.

Tumors in blast crisis presented elevated EphA3 expression, e.g., CML in accelerated or blastic crisis but not chronic phase.

Elevated EphA3 expression was noted in pre-B-ALL & pro-B-ALL (early pre-B-ALL) rather than mature B-cell ALL.

EphA3 expression for some peripheral B-cell neoplasms correlated well with tumor grade: high grade, poorly differentiated (typically aggressive) B-cell lymphomas or follicular lymphomas were EphA3+. A similar relationship was noted for non-Hodgkin lymphoma (Hodgkin lymphoma was EphA3-).

Preclinical screening also provided evidence for EphA3 expression by stroma/fibroblast (mostly lymphoma) and vasculature/endothelium, further rationale for development of reliable tools for profiling EphA3 in hematologic tumors and other malignancies.

Using well-characterized normal and diseased FFPE bone marrow biopsies, this IHC assay for EphA3 has subsequently been validated (QualTek) to provide data that is not directly available from routine histopathology review and that supports use of the assay for profiling EphA3 in specific hematologic tumors and for patient selection in early Phase clinical trial/s of an anti-EphA3 monoclonal antibody (KB004, KaloBios).

Beyond this, EphA3 targeted therapy with KB004 is anticipated for treatment of solid tumors. Recent genome-wide surveys identify EphA3 amongst the most frequently overexpressed genes in pancreatic, colon and lung carcinoma, melanoma and glioblastoma.