Discussion
Ewing’s sarcoma remains an enigmatic and particular malignant tumor 100 years after its discovery. Spinal involvement most commonly results from metastasis in advanced stages of the disease, while EWS originating from the spine is rare and extremely rare if the sacrum is excluded [2,3]. Diagnosis could be quite difficult due to an insidious onset, non-specific symptoms and misinterpreted images. Thus, symptoms may not be present until neurological deficits occur and diagnosis suspicion may be after several consultations which increase doubts [7,8]. The average delay from the onset of symptoms to the diagnosis has been reported to be 34 weeks all locations combined [9]. In our case the delay was 48 weeks. That’s why approximately 25% of patients present with metastatic disease at diagnosis; fortunately this was not the case of our patient [3].
Definitive diagnosis requires cytological, immunohistochemical (CD99) and cytogenetic analysis of a pathologic specimen. The translocation involving chromosome 22 is identified in more than 90% of cases, and it is the landmark to differentiate EWS from other small, round blue cell tumors [3,10].
Currently, early diagnosis and multimodal treatment combining surgery, chemotherapy, and local radiation therapy increases the chance of a successful outcome [11]. Indelicato et al [12], in a review have reported a five-year overall survival rate of 71% and local control rate of 89% for non-metastatic spinal and paraspinal EWS. However, when compared to other sites of occurrence, prognosis of EWS of the spine remains worse [13].
Initial chemotherapy and local radiotherapy might be administered before surgery with the aim to shrink bulky and unresectable tumors, to eradicate micrometastases and for acute relief of epidural compression, but it should be noted that there is a variable sensitivity to radiation and chemotherapy due to biological heterogenecity [1].
During the last two decades, the outcome in patients with localized disease has improved through an aggressive surgery known as en bloc vertebrectomy by combining anterior and posterior approach or from a single posterior approach as described by Tomita and al [14].
Boriani et al [6] described three major methods of performing en bloc excisions in the thoracolumbar spine: vertebrectomy if the tumor is confined to zones 4 to 8 or 5 to 9; sagittal resection when the tumor occupies zones 3 to 5 or 8 to 10 and resection of the posterior arch when it is located between the zones 10 and 3 according to their surgical staging system.
However, en bloc resection is a highly demanding procedure that must be carefully planned and the greater surgical risk can be accepted only if it offers a safer result and is performed by specialized surgical and anesthesiology teams [5].
Among the difficulties to be considered is the release of the spinal cord if the tumor is expanding to layer D according to WBB staging system. In this case theoretical safe margin should include the dura in the resection specimen but the cost-to-benefit ratio of such procedure should be carefully evaluated. Boriani and his collaborators have shown that a simple release of the dura without resection can be accepted and without proven consequence on the risk of recurrence [6,13].
Compared to cases where only decompression or lesionectomy was done, patients who underwent en bloc spondylectomy had a lower recurrence rate [8,15]. This is why, whenever possible, surgical en bloc wide resection with an anterior column reconstruction is preferable in order to obtain a better oncological control and a better preservation of the spine biomechanics [3].