Background: Neoadjuvant chemotherapy followed by surgery and adjuvant chemotherapy is the most-frequently employed strategy for patients with high-grade osteosarcoma. However, the contribution of neoadjuvant chemotherapy compared with postoperative adjuvant chemotherapy has not been tested rigorously in patients with nonmetastatic high-grade pelvic osteosarcoma.
Questions/purposes: (1) Does neoadjuvant chemotherapy followed by delayed surgery offer a survival benefit to patients with nonmetastatic high-grade pelvic osteosarcoma compared with immediate surgery and adjuvant chemotherapy? (2) Is the timing of chemotherapy and surgery associated with a difference in the survivorship free from local recurrence and the risk of complications? (3) Is the threshold of 90% necrosis after neoadjuvant chemotherapy appropriate to distinguish responders from nonresponders in patients with pelvic osteosarcoma?
Methods: Between 2000 and 2015, our center treated 112 patients with nonmetastatic high-grade primary pelvic osteosarcoma, of whom 93 underwent tumor resection with chemotherapy. Four patients (4%) were lost to followup before 24 months but were not known to have died; the remaining 89 patients were included in this retrospective study. Based on the timing of surgery and chemotherapy, patients were analyzed in two groups: (1) neoadjuvant chemotherapy followed by delayed surgery and adjuvant chemotherapy (n = 56; mean followup of 61 months, range 27-137 months), and (2) immediate surgery followed by adjuvant chemotherapy (n = 33; mean followup of 77 months, range 25-193 months). The total duration and intensity of chemotherapy was similar in both groups. During the period in question, we generally used neoadjuvant therapy followed by delayed surgery and adjuvant chemotherapy when patients received their biopsies in our center. We typically used immediate surgery with adjuvant chemotherapy when patients initially refused chemotherapy or when they had severe pain or poor walking function. Patients in the neoadjuvant chemotherapy group had a higher proportion of sacral infiltration; other factors such as sex, age and tumor size were well balanced between groups. We compared overall survival and local recurrence-free survival rates between the two groups. We completed univariate log-rank tests and multivariate Cox analyses in all patients to identify factors associated with survival and local recurrence using the Kaplan-Meier method.
Results: No survival benefit was found in the patients treated with neoadjuvant chemotherapy followed by delayed surgery compared with the group treated with immediate surgery and adjuvant chemotherapy. At 5 years, the overall survival (OS) was 42% (95% CI, 33-52) for all patients in this study, 43% (95% CI, 30-56) for the neoadjuvant group, and 40% (95% CI, 25-55) for the immediate surgery group; p = 0.709. With the numbers available, there was no difference in the likelihood of successful limb salvage (five of 56 patients [89%] in the neoadjuvant chemotherapy group versus three of 33 patients [91%] in the immediate surgery group; p = 0.557). The 5-year local recurrence-free survival was 67% (95% CI, 59-76) with no difference between the two groups (68%; 95% CI, 57-78% versus 67%; 95% CI, 52-81; p = 0.595). With the numbers available, there was no difference in survival between patients whose tumors demonstrated more than 90% necrosis; however, only four of 56 patients in the neoadjuvant chemotherapy group demonstrated 90% necrosis.
Conclusions: We found no survival advantage with chemotherapy before surgery compared with immediate surgery in patients with nonmetastatic high-grade pelvic osteosarcoma. The decision on chemotherapy timing should be made for reasons other than survival. A prospective trial is needed to confirm this conclusion.
Level of evidence: Level III, therapeutic study.