Validation of prognostic scoring and assessment of clinical benefit for patients with bone sarcomas enrolled in phase I clinical trials

Background: We sought to validate the Royal Marsden Hospital (RMH) and MD Anderson Cancer Center (MDACC) prognostic scoring systems for the selection of bone sarcoma patients for phase I clinical trials and to identify additional risk factors related to survival. Patients and Methods: We retrospectively reviewed the baseline characteristics and outcomes of 92 bone sarcoma patients who were referred to MDACC's Phase I Clinical Trials Program. Results: Ninety-two patients with Ewing sarcoma (N = 47), osteosarcoma (N = 22), chondrosarcoma (N = 16), and other tumors (N = 7) were evaluated; 78 were enrolled in at least 1 of 43 different phase I trials. The median overall survival (OS) was 8.8 months (95% confidence interval [CI] = 6.8-13.7 months). Independent factors that predicted shorter survival were male sex, > 2 metastatic sites, > 3 previous therapies, hemoglobin level < 10.5 g/dL, platelet count > 200 x10³/L, creatinine level ≥1.3 mg/dL, and lactate dehydrogenase level >ULN. Patients with good RMH scores (0-1) had longer OS than patients with poor RMH scores (2-3) (HR = 5.8, 95% CI = 2.9-11.0; P < 0.0001), as did patients with low MDACC scores (0-1) as compared to patients with higher MDACC scores (2-4) (HR = 3.2, 95% CI = 1.9-5.6; P < 0.0001). Conclusion: The RMH prognostic score can be used to predict the OS of bone cancer patients referred for phase I trials. The MDACC score added no value to the RMH score and therefore does not have a role in assessment of patients with bone tumors. Patients with advanced bone sarcomas should be considered for phase I trials.