The prognostic impact of unplanned excision in primary STS of the extremities and trunk wall: Is re-excision always mandatory?

Abstract

e23557 Background: To investigate the outcome of patients affected by soft tissue sarcomas (STS) of the extremities and trunk wall who underwent unplanned excision (UE) and the prognostic impact of microscopic residual tumor at re-excision. Methods: All consecutive patients affected by primary STS of the extremity and trunk wall operated at Fondazione IRCCS Istituto Nazionale dei Tumori from Jan 1997 to Dec 2017 were included and divided in 3 groups: patients primarily resected at our institution (group A), patients undergoing re-excision after macroscopically complete UE (group B) and patients undergoing completion resection after macroscopically incomplete UE (group C). Inverse probability treatment weighted overall survival (OS), crude cumulative incidence of local relapse (CCI-LR) and of distant metastasis (CCI-DM) were calculated and compared. Histological subgroup curves were drawn and multivariable models were performed in group B. Results: 1965 patients were identified; 1079, 679 and 189 in group A, B and C respectively. Median follow-up was 84.41 months. 7-yrs OS was 0.74, 0.84 and 0.81 (p < .001) for group A, B and C respectively; 7-yrs CCI-LR and DM were 0.05 and 0.25, 0.12 and 0.16, 0.14 and 0.29 (p < .001 for both) for group A, B and C respectively. In group B, microscopic residual tumor was present in 26.9% of the surgical specimens, but this figure varied broadly among the histologic subtypes (being the lowest in Solitary Fibrous Tumor, 7.7%, and the highest in Myxofibrosarcoma [MFS], 46.5%). At multivariable analysis, age (for OS only), grade, size and histology were associated to OS and CCI-DM; while age, presence of residual tumor (p < .001) and grade to CCI-LR. Of note, Myxoid liposarcoma (MLPS) and MFS showed the highest CCI-LR in presence of residual disease (18.6% vs 1.3% and 24.7% vs 7.2% respectively). Conclusions: Patients who underwent UE had a risk of LR twice as high as those primarily resected at a referral institution. However, this higher risk was not associated to a higher risk of DM or worse OS. In patients undergoing re-excision after macroscopic complete UE the presence of residual tumor was associated to a higher LR risk, again with no association with DM or OS. Despite the probable presence of a residual selection bias, it is possible that postponing re-excision after macroscopic complete UE doesn’t impact on the outcome. However, higher caution should be exerted for MLPS and MFS, given the higher probability of local relapse when microscopic residual tumor is found in the surgical specimen.