Pathological fracture due to lytic lesion caused by a myeloid neoplasm with FIP1L1‐PDGFRA

A normally fit 44-year-old male nuclear power plant worker presented with severe hip pain after kicking a ball. Radiological investigations revealed a lytic lesion in the right neck of femur (top). He denied constitutional symptoms. A full blood count showed: haemoglobin concentration 142 g/l, platelet count 217 × 109/l and white blood cell count 8·9 × 109/l with a normal differential count except for an eosinophil count of 1·21 × 109/l. Routine biochemistry, including urate, lactate dehydrogenase and serum protein electrophoresis, was normal. Computed tomography and positron emission tomography scans showed no evidence of other bone lesions or malignancy. A biopsy of the lesion revealed a dense infiltrate of eosinophils admixed with histiocytic and other myeloid cells and accompanied by Charcot–Leyden crystals (bottom). Immunohistochemistry showed no B-cell or T-cell infiltrate. There was no evidence of a plasma cell neoplasm, Langerhans cell histiocytosis or mastocytosis. Fluorescence in situ hybridization of the tumour revealed deletion of CHIC2 (4q12), consistent with a FIP1L1-PDGFRA fusion gene. A subsequent marrow aspirate showed 10% eosinophils but no monotypic B-cell population or immunophenotypically aberrant T-cell population. A trephine biopsy showed eosinophilia with clustered mast cells surrounding immunophenotypically normal B-cell-rich nodules. The bone marrow aspirate was unexpectedly negative for FIP1L1-PDGFRA and analysis from the trephine biopsy specimen failed. The patient was commenced on imatinib 100 mg orally daily with resolution of the eosinophilia and subsequent marrow biopsy has shown resolution of the above features. We are not aware of any previous report of a FIP1L1-PDGFRA-associated chronic eosinophilic leukaemia presenting with an osteolytic lesion. There have been a number of reports of extramedullary acute myeloid leukaemia harbouring a FIP1L1-PDGFRA fusion gene but none presented with pathological fracture due to a lytic bone lesion and, in addition, there was no evidence of a primitive myeloid neoplasm in our patient. There has been one report of a lytic bone lesion in a patient with the hypereosinophilic syndrome, but this was prior to the discovery of the FIP1L1-PDGFRA fusion gene.