Characteristics and Long-Term Outcome of Surgically Managed High-Grade Extremity Chondrosarcoma

Dedifferentiated chondrosarcoma (DCS) is a highly malignant variant that portends a poor prognosis. Although factors such as clinico-pathological characteristics, surgical margin, and adjuvant modalities likely play a role in overall survival, debate continues with varying results on the importance of these indicators. The purpose of this study is (1) To delineate the characteristics, local recurrence (LR), and survival of patients with intermediate (IGCS), high (HGCS) and dedifferentiated (DCS) chondrosarcoma of the extremity by utilizing detailed cases at one tertiary institution. (2) To assess survival between high grade chondrosarcoma and DCS utilizing a less detailed but large cohort from the Surveillance, Epidemiology, and End Results (SEER) database.

Twenty-six cases of high-grade (conventional FNCLCC grades 2 and 3, dedifferentiated) chondrosarcoma were identified from an ongoing prospective cohort of 630 sarcoma patients managed surgically at a tertiary referral university hospital between 9/1/2010-12/30/2019. A retrospective review of demographics, tumor characteristics, surgical procedure, treatment course, and survival data was performed to determine prognostic factors for survival. An additional 516 cases of chondrosarcoma were identified from the SEER database. Using the Kaplan-Meier method, both the large database and case series were evaluated, and estimated cause-specific survival was calculated at 1, 2, and 5 years.

There were 12 IGCS, 5 HGCS, and 9 DCS patients in the single institution cohort. DCS had a higher stage at diagnosis (p=0.04). Limb salvage was the most common procedure performed in every group (11/12 IGCS, 5/5 HGCS, and 7/9 DCS; p=0.56). Margins included 8/12 wide and 3/12 intralesional for IGCS. For HGCS, there were 3/5 wide, 1/5 marginal, and 1/5 intralesional. A majority of DCS margins were wide (8/9) with only 1 marginal. There was no difference of associated margins between the groups (p=0.85), however there was a difference when margins were classified based on numerical measurement (IGCS: 0.125cm (0.1-0.35); HGCS: 0cm (0-0.1); DCS: 0.2cm (0.1-0.5); p=0.03). The overall median follow-up was 26 months (IQR:16.1-70.8). The time interval from resection to death was lower in DCS (11.5 months (10.7-12.2)), followed by IGCS (30.3 months (16.2-78.2)), and HGCS (55.1 months (32.0-78.2; p=0.047). LR occurred in 5/9 DCS, 1/5 HGCS, and 1/14 IGCS patients. Of the DCS patients only 2/6 who received systemic therapy had LR, while all 3/3 who did not receive systemic therapy had LR. Overall systemic therapy and radiation did not impact incidence of LR (p=0.67; p=0.34). However, patients who had LR were 17.5 times more likely to die within one year (HR=17.5, 95%CI (1.01-303.7), p=0.049), after adjusting for the age at the surgery. There was no correlation with the utilization of systemic therapy, radiation therapy, or margin and overall survival (p=0.63, p=0.52, p=0.74). In the SEER patient cohort, 149 cases (28.9%) were DCS and 367 (71.1%) were HGCS. At final follow-up, 49.6% (n=256) of the cohort had a cause of death due to chondrosarcoma. HGCS was associated with higher chance of 1-year survial (p<0.001), 2-year survival (p<0.001), 5-year survival (p<0.001), and overall survival (p<0.001). Additionally, decreased survival was associated with metastatic disease at presentation (p=0.01). Overall limb salvage was most utilized for both HGCS (76.5%) and DCS (74.3%). In regard to limb salvage vs. amputation, there was no difference in survival at 1 year (p=0.10) or 2 year (p=0.13) between the groups, however those who underwent limb salvage procedure had a significantly better chance of survival at 5 years when compared to amputation (HR=1.49 (1.11-1.99); p=0.002).

High-grade chondrosarcoma remains a fatal disease in many patients, particularly if associated with dedifferentiated subtype. Interestingly, all (100%) DCS patients who did not receive systemic therapy had LR. However, chemotherapy and radiation did not significantly increase survival. In this case series and large database study, HGCS had the smallest surgical margin, but with the longest time interval for both LR and death. Additionally, using the SEER database, DCS and amputation had worse prognosis at the 5-year survival time. Further studies on valuable prognostic influences as well as earlier identification of this rare disease may help in developing better management options. Level of Evidence: III.