Clinicopathological and molecular spectrum of ewing sarcomas/PNETs, including validation of EWSR1 rearrangement by conventional and array FISH technique in certain cases.

Over the years, a wide clinicopathological spectrum has been identified within Ewing family of tumors (EFTs). As these tumors are chemosensitive, their correct and timely identification is necessary. The aims of this study were (1) to present the diverse clinicopathological and molecular profile of EFTs in our settings, (2) to identify a pragmatic approach for diagnosing EFTs, especially for application of ancillary techniques, namely RT-PCR for specific transcripts (EWS-FLI1, EWS-ERG) and FISH for EWSR1 gene rearrangement, in certain cases and (3) to show the utility of tissue microarray in establishing a new FISH test. Fifty-eight EFTs were identified in 38 males and 20 females within an age-range of 1-65 years (median, 16), mostly in lower extremities (14) (24.1 %). Therapeutically, most patients underwent neoadjuvant chemotherapy with subsequent surgery. Histopathologically, diagnosis of EFTs was initially offered in 41/58 (70.6 %) tumors. On review, 59 % tumors showed diffuse pattern, while 41 % displayed rosettes. Immunohistochemically, tumor cells were mostly diffusely positive for CD99 (48/52) (92.3 %); FLI-1 (17/18) (94.4 %); variably for BCL2 (16/18) (88.8 %), synaptophysin (6/20) (35 %), S100-P (2/7) (28.5 %), CD56 (2/5) (40 %), NSE (2/5) (40 %), calponin (3/4) (75 %), EMA (5/24) (20.8 %) and CK (3/24) (12.5 %), the latter two mostly focally. Fifty five tumors were EWS-FLI1 positive, while a single tumor was EWS-ERG positive. Sensitivity for PCR was 61 %. EWSR1 rearrangement was detected by FISH in 12/13 Ewing sarcomas/PNETs. Sensitivity for EWSR1 test was 92.3 % and specificity was 100 %. Thirty-eight tumors, including 14 molecular confirmed EFTs and 21 other tumors were tested for EWSR1 rearrangement. Among 21 unrelated tumors, EWSR1 rearrangement was detected in few myoepithelial tumors, occasional desmoplastic small round cell tumor and an extraskeletal myxoid chondrosarcoma. Further, a tissue microarray with a separate set of 8 EFTs, confirmed at another laboratory was analysed for validation of EWSR1 rearrangement test. 23/28 (82.1 %) tissue cores of the tissue microarray, stained by FISH were interpretable, including EWSR1 rearrangement, detected in 20/28 tissue cores; not detected in 3 liver cores and uninterpretable in 5 (17.8 %) cores. Classical EFTs can be diagnosed with diffuse, membranous CD99 positivity, intranuclear FLI1 positivity and LCA negativity in malignant round cells. In unconventional cases, it is indispensable to reveal the concomitant fusion m-RNA by RT-PCR. In case of negative molecular results, it is necessary to prove EWSR1 rearrangement by FISH. These tests should be interpreted with clinicopathological correlation. Tissue microarrays for FISH are useful during validation of a new test, especially when sarcomas like EFTs show less genetic heterogeneity within tumor cells.