P59.27 Complementary Utility of Combined ALK/ROS1 FISH with Immunohistochemistry for ALK/ROS1 Rearrangement Testing in Lung Cancer

Abstract

Testing for EGFR mutations, ALK and ROS1 rearrangements and BRAF mutations are now recommended as the essential minimum predictive biomarkers to be tested in primary non-squamous NSCLC patients presenting with advanced disease. While the ALK D5F3 Ventana immunohistochemistry (IHC) Assay is approved for detection of ALK rearrangements and for targeted therapy, ROS1 IHC only serves as a screening tool for ROS1 rearrangements with demonstration of gene fusions by molecular methods required for confirmation and targeted treatment. Nevertheless, equivocal IHC results, although uncommon, still occur with the D5F3 ALK assay necessitating complementary methods for confirmation. The recent availability of combined ALK/ROS1 break-apart fluorescence in-situ hybridisation (FISH) probes offer the advantage of tissue conservation with ability to analyse both gene alterations in a single tissue section. The aim of the present study was to study the utility of the combined ALK/ROS1 FISH assay as a complement to ALK and ROS1 IHC in routine predictive biomarker testing algorithms. Over a 6-year study period, all EGFR wild type NSCLC samples were subject to automated ALK D5F3 Ventana IHC assay and ROS1 IHC using the D4D6 (Cell signalling technology) clone as a manual lab developed test. Diffuse strong granular staining in tumor cells for ALK and ROS1 protein was interpreted as positive. All cases of equivocal ALK staining and positive ROS1 staining was subject to combined ALK/ROS1 FISH using the FlexISH ALK/ROS1 DistinguISH Probe (Vysis) performed as per manufacturers’ instructions. Presence of ALK/ROS1 break apart signals and/or isolated ALK/ROS1 3’ signals in >15% of interpretable tumour nuclei was considered as positive for ALK or ROS1 rearrangement. The ALK Ventana Assay was performed in 711 formalin fixed paraffin embedded EGFR wild type NSCLC samples including biopsies, lymph node excisions and cell blocks. Of these, 12% (85/711) were positive, 2% (13/711) were equivocal, 75% (539/711) were negative, while in the remaining 11% (76/711), result could not be evaluated due to exhaustion of tumor tissue. FISH analysis in four of the ALK IHC equivocal cases show absence of ALK translocations while no tissue remained for FISH analysis in the remaining equivocal cases. ROS1 IHC was performed in 434 cases of the above cohort of which 7% (30/434) were positive. ROS1 immunopositivity was enriched among ALK IHC positive cases with 43% (19/44) of ALK IHC positive cases also showing protein expression of ROS1. FISH analysis revealed ROS1 translocation in two cases (both showing diffuse strong ROS1 protein expression and negative ALK IHC) while all dual ALK/ROS IHC positive cases tested by FISH showed presence of only ALK rearrangements with non-rearranged ROS1 signals. Tissue adequacy remains the major limiting factor for predictive biomarker testing in lung cancer. The use of combined ALK-ROS FISH allows for excellent demonstration of ALK and ROS1 rearrangements in the same section using the same criteria validated in single gene break apart FISH probes. ROS1 protein is commonly overexpressed in ALK rearranged NSCLC and may represent cross reactivity with ALK or a related protein.