In this case report we demonstrate a metatarsal lesion with morphological features in keeping with ‘classic’ Ewing sarcoma, on initial biopsy, with morphological and immunophenotypical changes to an Adamantinoma-like Ewing sarcoma variant post chemotherapy. We considered that this change may be due to one of the following: 1. Emergence/persistence of a more chemotherapy resistant clone; 2. Under sampling of the tumour on the initial biopsy; or 3. Alterations in the tumour morphology &/or immunophenotype caused by chemotherapy.
Chemotherapy response of Ewing sarcoma is graded according to the Paediatric Oncology Groups Study(24)and these changes typically are present as necrosis, haemorrhage, cystic degeneration, calcification, ossification and fibrosis, though alternative grading systems like that by Huvos et al. also exist (25). The grading of chemotherapeutic response has prognostic and therapeutic significance as it plays a role in chemotherapy agent selection for the continuation chemotherapy (24, 26–28). There is also some literature that suggests that the Adamantinoma-like Ewing sarcoma may be associated with a more aggressive clinical behaviour and with poorer outcomes (7).
Other morphological changes in Ewing sarcoma post chemotherapy include: neuronal maturation with rosette formation and gangliocytic phenotype differentiation. Rhabdoid change has also been described though the tumour maintained its immunophenotype (29, 30). Knezevich et al. described loss of EWS/FLi-1 gene fusion in the recurrent tumour (31), while Smith et al. described a different genetic alteration in one of the tumour nodules (32). No case reports of squamous differentiation (including keratin pearl formation) with P40 expression post chemotherapy could be found on a literature search (33).
Adamantinoma–like Ewing sarcoma shows squamous differentiation, cytokeratin expression and P40/P63 expression which is not seen in classic Ewing sarcoma although 25% of classic Ewing sarcomas do express keratins such as CAM5.2, AE1/AE3 and MNF-116 (3, 4). In our case report the original tumour was P63 & P40 negative and only showed positivity for these two markers in the resection specimen (post neo-adjuvant chemotherapy). CK5 tended to mimic this pattern. FISH testing for EWS fusion was positive on the initial biopsy, but the mating partner remains uncertain.
It would hence appear that this phenotype and immunophenotype was not present initially and only appeared post-chemotherapy. Therefore, raising the following questions: 1. Is this a more chemotherapy resistant clone? 2. Was the initial tumour truly under sampled? 3. Is this a chemotherapy induced change?
Additionally if a pathologist is confronted with a metastatic lesion of such a tumour (particularly in an older patient), in the head and neck region, it may cause diagnostic dilemmas with a number of neoplasms that also show a small cell morphology with squamous differentiation such as basaloid variants of squamous cell carcinoma and NUT-midline carcinoma. Thus, adequate immunohistochemical workup and/or molecular confirmation to establish the diagnosis of Adamantinoma-like Ewing sarcoma is advised (17) .
Apart from the unique histological findings of this case, a novel reconstructive strategy was also employed. Reconstruction of the first metatarsal is more critical than the lesser metatarsals given its role in progression through the gait cycle. In the context of trauma, first ray reconstruction using both free and pedicled fibula flaps was described as early as 1991(34). Wang et al. described a series of four patients who had metatarsal reconstruction, two of the first ray and two of the lesser, using pedicled fibula flaps (35). More recently Hilaire (36) described computer assisted virtual surgical planning in a case of first metatarsal reconstruction using a free fibula flap. Malignancy of the metatarsals, however, has limited options in terms of salvage surgery. Management of primary bone tumours of the lesser metatarsals usually consist of ray (37) or below knee amputation (35, 36). Few reports exist on the use of autologous vascularised fibula grafts, either free or pedicled, for forefoot reconstruction. Borthakur et al. described vascularised free fibula transfer post resection of a first ray osteosarcoma (38). Toriyama et al. reported concomitant first and second metatarsal reconstruction following resection for chondrosarcoma using a free vascularised double-barrelled fibula graft (39), and Toma et al. reported a series of six cases of metatarsal primary bone tumours resected and reconstructed with free vascularised fibula graft, including the first and lesser rays(40). To our knowledge, our case is the first description of a pedicled vascularised fibula graft used for the reconstruction of the first metatarsal post resection for malignancy.