Small Round Blue Cell Tumours of Bone

Q: Two round cell sarcomas of bone look identical down the microscope. How much can the fusion alone change the prognosis?

A: From 40 per cent to 92 per cent. In the 148-patient global GRACefUl series, three-year overall survival was 92.2 per cent for BCOR::CCNB3-rearranged sarcoma, 78.7 per cent for unclassified undifferentiated round cell sarcoma, and 39.6 per cent for CIC-rearranged sarcoma (p less than 0.0001). They also differ clinically: BCOR::CCNB3 patients are younger (median 17 versus 32), male-predominant, more often bone primaries, with fewer synchronous metastases. This is why no family can be counselled and no outcome quoted until the molecular result returns — and why fresh tissue must leave theatre for sequencing. Caveat: retrospective over 36 years with non-standardised treatment, so the survival gap is confounded by era and regimen as well as biology.

Q: A biopsy shows small round blue cells with strong membranous CD99. Is this diagnostic of Ewing sarcoma?

A: No. CD99 is sensitive but not specific — lymphoblastic lymphoma, synovial sarcoma and some rhabdomyosarcomas express it, and staining intensity also varies in CIC::DUX4 and BCOR::CCNB3 sarcomas. Add NKX2.2: alone it is 80 per cent sensitive and only 84 per cent specific, with false positives in olfactory neuroblastoma, extraskeletal myxoid chondrosarcoma, mesenchymal chondrosarcoma, small cell carcinoma and Merkel cell carcinoma. CD99 and NKX2.2 together reach 98 per cent specificity, fooled in Shibuya's 131 tumours by only one mesenchymal chondrosarcoma and one small cell carcinoma. Molecular demonstration of the fusion remains the reference standard.

Q: EWSR1 break-apart FISH is positive on a round cell sarcoma of bone. Does that settle the diagnosis?

A: No, and this is the current trap. WHO recognises a separate category — round cell sarcoma with EWSR1-non-ETS fusion, including NFATc2 and PATZ1 partners — in which the break-apart probe is also positive because EWSR1 is genuinely rearranged, just not with an ETS partner. Break-apart FISH detects the split; it does not name the partner. Sequencing does. The four WHO categories are Ewing sarcoma, EWSR1-non-ETS fusion sarcoma, CIC-rearranged sarcoma, and sarcoma with BCOR genetic alteration — and "Ewing-like sarcoma" and "PNET" are retired terms.

Q: The panel returns NKX2.2 negative on a round cell sarcoma. What does that tell you?

A: It argues against Ewing sarcoma. NKX2.2 was diffusely nuclear positive in 93 per cent of Ewing sarcomas but negative in both CIC::DUX4 and BCOR::CCNB3 sarcomas in Çetin's ten-year single-centre reclassification. DUX4 positivity then identifies CIC::DUX4 outright. Morphology helps while sequencing is pending: pleomorphism, prominent nucleoli, wide clear or eosinophilic cytoplasm, myxoid or pink hyaline stroma and spindle cell areas all point away from Ewing.

Q: Is it reasonable to give an Ewing-type regimen to any round cell sarcoma of bone while awaiting molecular results?

A: It is the pragmatic default, but know the signal against it. In Çetin's series, BCOR::CCNB3 sarcomas treated on an Ewing regimen showed a histopathological regression response of 30 per cent, against 70 per cent in those given non-Ewing regimens. Treat that as a reason to obtain the molecular diagnosis urgently rather than as established practice: it rests on eighteen patients split between non-randomised arms, with regression response as a surrogate rather than survival.

Q: Which single finding best distinguishes Ewing sarcoma from osteosarcoma on plain radiograph?

A: Tumour matrix. Osteosarcoma produces cloud-like or fluffy mineralised osteoid within the lesion and the soft tissue mass; Ewing produces no matrix, only reactive host sclerosis. Location helps secondarily (diaphysis for Ewing, metaphysis for osteosarcoma) but periosteal reaction patterns overlap completely — onion-skin is not pathognomonic of anything.

Q: A 4-year-old has multiple lytic skull and long bone lesions. What is the first-line non-invasive test?

A: Urinary vanillylmandelic acid and homovanillic acid, together with a full blood count and film. These identify metastatic neuroblastoma and acute leukaemia respectively, and can spare the child an open bone biopsy. Ewing sarcoma is rare under 5.

Q: Which MRI sign favours a Brodie abscess over a round cell tumour?

A: The penumbra sign — a thin rim of relatively high signal on unenhanced T1-weighted images at the margin of the abscess cavity, representing vascular granulation tissue. Supporting features are a sequestrum, sinus tract, involucrum and rim enhancement around a non-enhancing centre.

Q: Why must every biopsy of a suspected Ewing sarcoma include a microbiology sample?

A: Because subacute osteomyelitis reproduces the entire clinical and radiological picture — fever, raised ESR, permeative diaphyseal destruction, lamellated periosteal reaction — and reactive lymphoid infiltrate can be misread histologically. Culture is the only reliable way to exclude it, and missing infection delays curative antibiotic treatment. Inflammatory markers never discriminate between the two.