Toshifumi Ozaki, Keiko Hayakawa, Hiroyuki Kawashima, Naofumi Asano, Itsuo Watanabe, Tsukasa Yonemoto, Takahiro Goto, Yoshihiro Nishida, Hiroyuki Futani, Robert Nakayama, Makoto Endo, Kenichiro Hamada, and Eisuke Kobayashi
Background Clear cell chondrosarcoma is an extremely rare chondrosarcoma subtype; thus, its treatment outcomes and associated factors have not been widely studied. Knowing more about it is potentially important because clear cell chondrosarcomas are often misdiagnosed as other benign lesions and subsequently treated and followed inappropriately. Questions/purposes (1) What are the patient- and tumor-related characteristics of clear cell chondrosarcoma? (2) What proportion of patients with clear cell chondrosarcoma initially had a misdiagnosis or a misleading initial biopsy result? (3) What is the survivorship of patients with clear cell chondrosarcoma free from death, local recurrence, and distant metastasis, and what factors are associated with greater survivorship or a reduced risk of local recurrence? Methods Between 1985 and 2018, 12 Japanese Musculoskeletal Oncology Group (JMOG) hospitals treated 42 patients with a diagnosis of clear cell chondrosarcoma. All 42 patients had complete medical records at a minimum of 1 year or death, and were included in this multicenter, retrospective, observational study. No patients were lost to follow-up within 5 years of treatment but four were lost to follow-up greater than 5 years after treatment because their physicians thought their follow-up was sufficient. Clinical data were collected by chart review. The median (range) follow-up period was 69 months (2 to 392). In general, when a possibly malignant bone tumor was found on imaging studies, the histological diagnosis was made by biopsy before initiating treatment. Once the diagnosis had been made, the patients were treated by surgery only, complete resection if technically possible, because chondrosarcomas are known to be resistant to chemotherapy and radiotherapy. Unresectable tumors were treated with particle-beam radiation therapy. When patients with chondrosarcoma were referred after unplanned surgical procedures with inadequate surgical margins, immediate additional wide resection was considered before local recurrence developed. This diagnostic and treatment strategy is common to all JMOG hospitals and did not change during the study period. Primary wide resection was performed in 79% (33 of 42) patients, additional wide resection after initial inadequate surgery in 12% (five of 42), curettage and bone grafting in 5% (two of 42) patients, and radiotherapy was administered to 5% (two of 42). Surgical margins among the 40 patients who underwent surgery at JMOG hospitals were no residual tumor in 93% (37 of 42) of patients, microscopic residual tumor in 2% (one of 42), and macroscopic residual tumor or state after curettage or intralesional excision in 5% (two of 42). The oncological endpoints of interest were 5- and 10- year overall survival, disease-free survival, survival free of local recurrence, and survival free of distant metastases; these were calculated using the Kaplan-Meier method and compared using the log-rank test. Risk ratios with their respective 95% confidence intervals (CIs) were estimated in a Cox regression model. The Bonferroni adjustment was used for multiple testing correction. Results The sex distribution was 74% men and 26% women (31 and 11 of 42, respectively), with a mean age of 47 ± 17 years. Eighty one percent (34 of 42) of tumors occurred at the ends of long bones, and the proximal femur was the most common site accounting for 60% (25 of 42). The mean size of the primary tumors was 6.3 ± 2.7 cm. Definite pathologic fractures were present in 26% (10 of 42) and another 26% (10 of 42) had extraskeletal involvement. None had metastases at presentation. Twenty four percent (six of 25) tumors in the proximal femur were misdiagnosed as benign lesions and treated inadequately without biopsy. Twenty nine percent (10 of 35) patients had initial misdiagnoses by biopsy and core needle biopsies had a greater risk of resulting in inaccurate histological diagnoses. The study patients' 5- and 10-year overall survival rates were 89% (95% CI 74 to 96) and 89% (95% CI 74 to 96), respectively; 5- and 10- year disease-free survival rates 77% (95% CI 58 to 89) and 57% (95% CI 36 to 75), respectively; 5- and 10-year local recurrence-free survival rates 86% (95% CI 68 to 95) and 71% (95% CI 49 to 86), respectively; and 5- and 10-year distant metastasis-free survival rates 84% (95% CI 67 to 93) and 74% (95% CI 53 to 88), respectively. Notably, bone metastases (17%, seven of 42) were as common as pulmonary metastases (14%, six of 42); four patients developed both bone and pulmonary metastases. The difference between 10-year overall survival rates and 10-year disease-free survival indicated very late recurrence more than 5 years after the initial treatment. After controlling for multiple comparisons, the only factor we found that was associated with local recurrence-free survival was initial treatment (positive margin versus primary wide resection) (risk ratio 8.83 [95% CI 1.47 to 53.1]; p = 0.022 after the Bonferroni adjustment). Additional wide resection reduced the risk of local recurrence. Conclusions The femoral head was the most common location of clear cell chondrosarcoma and had a high risk of misdiagnosis as common benign lesions that resulted in initial inadequate surgery and a consequent high risk of local recurrence. Immediate additional wide resection should be considered in patients who had initial inadequate surgery to reduce the risk of local recurrence. Because clear cell chondrosarcoma can recur locally or distantly in the bones and lungs in the long term, patients should be informed of the risk of very late recurrence and the necessity of decades-long with surveillance for local recurrence and lung and bone metastases. Level of evidence Level IV, therapeutic study.