Institutional experience and pooled survival analysis of chemotherapy based multimodal management in poorly differentiated chordoma

BACKGROUND: Poorly differentiated chordoma (PDC) is an exceptionally rare bone tumor with aggressive behavior and early recurrence. Evidence on the role and integration of chemotherapy remains limited. We aimed to share our institutional management experience and identify clinical prognostic factors in PDC. METHODS: Five consecutive patients with axial PDC treated at our spinal tumor center (June 2021-June 2025) received surgery and anthracycline-ifosfamide (AI) chemotherapy. Neoadjuvant response was assessed by RECIST 1.1 and overall survival (OS) by Kaplan-Meier. We also reviewed histologically confirmed PDC cases reported in the literature with eligible treatment and outcome data, and compared OS by chemotherapy exposure and other factors using log-rank tests. RESULTS: The mean age in our series was 21.4 years; all tumors were mobile-spine PDC with INI-1 loss and brachyury positivity. Neoadjuvant AI-based combination therapy achieved a partial response (PR) in one patient. Median progression-free survival was 13 months and median OS was 21 months. In the pooled cohort (n = 30), there was a female predominance. Prior recurrence was associated with inferior OS. In descriptive comparisons, patients receiving chemotherapy-based multimodality therapy had longer observed OS than those receiving local therapy without chemotherapy, including among recurrent cases. Whereas radiotherapy did not show a comparable survival benefit, this finding should be interpreted cautiously given the likelihood of substantial selection bias and the palliative role of RT in many recurrent cases. CONCLUSIONS: AI-based chemotherapy, particularly in the neoadjuvant setting, was feasible and showed preliminary antitumor signal in selected patients. However, the patient who achieved a partial response also received apatinib; thus, the specific contribution of AI chemotherapy cannot be isolated. In the pooled exploratory analysis, patients who received chemotherapy had a longer observed median OS than those receiving local therapy alone. However, due to the broad definition of chemotherapy and high risk of selection bias, this finding is presented as purely descriptive and hypothesis-generating, and does not imply a meaningful association between chemotherapy and improved OS.