Results
Our query identified 1303 children and young adults who met inclusion criteria, consisting of 103 (7.9%) patients with CCS, 221 (17.0%) with epithelioid sarcoma, and 979 (75.1%) with synovial sarcoma. Within the entire cohort, 683 (52.4%) patients were male and 890 (68.3%) were white. The median age at diagnosis was 18 years (range 0-25 years). The most common tumor location was in the extremities and trunk, found in 980 (75.2%) patients. There were 1256 (96.4%) patients who underwent primary surgical resection for local control. Of these patients, 183 (14.6%) received neoadjuvant local radiation, 357 (28.4%) received adjuvant local radiation, 11 (0.9%) received intraoperative local radiation, and 34 (2.7%) received both neoadjuvant and adjuvant local radiation. Radiation was the sole treatment for local control in 47 (3.6%) patients.
Patient demographics and tumor characteristics are compared by histologic subtype in Table 1. Larger tumors were seen in 510 (52.1%) patients with synovial sarcoma as compared to 34 (33.0%) patients with CCS and 71 (32.1%) patients with epithelioid sarcoma. Conversely, 72 (7.4%) patients with synovial sarcoma had distant metastasis, compared to 15 (14.6%) patients with CCS and 26 (11.8%) patients with epithelioid sarcoma. Primary surgical resection with local radiation was the most common method of local control for synovial sarcoma, in 499 (51.0%) patients, while surgery alone was more common for CCS (72 patients, 69.9%) and epithelioid sarcoma (157 patients, 71.0%).
Comparing the incidence of lymph node sampling procedures between histologic subtypes in Table 2, there were significantly fewer patients with synovial sarcoma who underwent lymph node sampling (p<0.001). Additionally, only 21 (2.2%) patients with synovial sarcoma had recorded lymph node metastasis, and this number was significantly fewer than the incidence of lymph node metastasis observed in 15 (14.6%) and 33 (15.0%) patients with CCS and epithelioid sarcoma, respectively (p<0.001, p<0.001). Relative rates of lymph node sampling for each histologic subtype remained the same across the 12-year study period.
Unadjusted overall survival differed by histological subtype (Fig. 1). Patients with clear cell sarcoma had a 5-year OS of 56.1%, and this outcome was significantly inferior compared to survival of patients with synovial or epithelioid sarcoma (p<0.001, p=0.001, respectively). There was no significant difference between the 79.5% 5-year OS for patients with synovial sarcoma and the 73.9% 5-year OS for patients with epithelioid sarcoma. When comparing KM estimates of 2-year OS by lymph node status for each histologic subtype (Fig. 2), patients with positive lymph nodes had significantly inferior survival, regardless of the subtype (p<0.001).
Table 3 shows the adjusted hazard ratios for each histological subtype based on lymph node sampling status. Lymph node sampling was associated with a 65% reduction in the risk of death for patients with CCS (p=0.010). In contrast, patients with synovial sarcoma who underwent a lymph node sampling procedure had an associated 60% increase in the risk of death as compared to those who did not have lymph nodes sampled (p=0.007). There was no significant difference in overall survival by lymph node sampling status in children and young adults with epithelioid sarcoma.
For all patients, regardless of histologic subtype, there was a two-fold increase in the risk of death for those with lymph node metastasis (p<0.001) after adjusting for confounding factors, as shown in Table 4. Histologic subtype did not significantly modify this association.