New Imaging System Offers Standardized Imaging Framework for Assessing Bone Metastases Response
A new imaging framework, RECIBM, standardizes the assessment of treatment response in bone metastases by integrating MRI, PET, SPECT, CT, and radiography into a unified classification system. RECIBM demonstrated strong agreement with existing response criteria, reclassified 19% of patients, and showed the best correlation with overall survival and bone progression-free survival, suggesting it may provide a more accurate and clinically meaningful tool for evaluating treatment outcomes. The study was published in the Radiology journal by Tingting X. and colleagues.
Longitudinal patient clinical databases and imaging databases of patients followed up between October 2017 and September 2023 were analyzed. The novel RECIBM scoring system was constructed through multi-disciplinary consensus among oncologists, radiologists, and nuclear medicine specialists; thereafter, the tool was assessed with clinical patient data. The design of the study included two cohorts: paired cohort that included patients whose advanced imaging made it possible to assess them on all four sets of criteria, i.e., RECIBM, University of Texas MD Anderson Cancer Center criteria, PERCIST and EORTC criteria.
Another cohort included patients who had undergone technetium 99m methylene diphosphonate bone scintigraphy, and their prognostic performance was evaluated using MD Anderson and RECIBM criteria. Agreement between criteria was quantified with Cohen kappa, prognostic performance for overall survival and bone progression-free survival was estimated with Harrell C-indices and multivariable Cox models.
Key findings:
- In general, the examined population of patients had a median age of 59.0 years with an interquartile range between 52.0 and 68.3 years [IQR, 52.0-68.3 years]. Besides, 59 people of this group were male.
- High statistical concordance of the RECIBM-defined responses classification was found with all existing criteria based on excellent Cohen's kappa values varying within the range from 0.80 to 0.98 (all P < 0.001).
- The application of the unified RECIBM approach helped to reclassify precisely 19% of the patient sample through correcting the clinical status of response for 16 people out of the total number of 84.
- In the paired imaging subset, RECIBM obtained the highest Harrell C-index for OS of 0.807 (in comparison with the range of older approaches – 0.771 to 0.782; P = 0.10-0.14).
- Moreover, a better baseline C-index for BPFS was observed using the new framework (C-index of 0.849 compared to the range 0.799-0.833 in the case of the existing approaches; raw P = 0.03-0.26; adjusted P = 0.088-0.256).
- Finally, multivariable Cox regression analyses proved that the RECIBM-defined response significantly predicted longer OS (hazard ratio, 0.07 [95% CI: 0.03, 0.19]; P < 0.001) and excellent bone progression-free survival (hazard ratio, 0.03 [95% CI: 0.01, 0.08]; P < 0.001).
It was found that RECIBM-based classifications had high agreement with conventional classifications, reclassified 19% of patients and correlated with overall survival and bone progression free survival. The impressive retrospective statistics presented above is necessary to form the empirical basis for modern oncological practice. It is proved that combining the separate data from imaging studies can efficiently solve problems in diagnosis of skeletal cancers.
Reference:
Xu, T., Yu, X., Chen, F., Wang, W., Cheng, K., Liu, Y., Wang, Y., Peng, D., Huang, Z., Zhang, C., Qiu, L., Liu, Z., Mou, L., Lei, L., Qi, C., Zhang, J., Li, B., Xie, L., Dai, T., … Chen, Y. (2026). Response Evaluation Criteria in Bone Metastases: Performance and Association of Response Classifications with Survival Outcomes. Radiology, 320(1). https://doi.org/10.1148/radiol.260907
