Maarten van der Doelen

Chapter 8

reflect less advanced disease. Of importance, the study results may be influenced by the learning curve of our team during the introduction of radium-223 as a new therapy for mCRPC patients. Nevertheless, this was one of the first real-world studies that showed the importance of patient selection for radium-223 therapy. Following the pivotal ALSYMPCA trial, multiple prospective and retrospective studies identified variables associated with OS in patients treated with radium-223 therapy. (45, 46, 48-54) The largest study that assessed variables associated with OS was the international, early access, open-label, single-arm phase 3b trial. In this study, OS was longer in patients with baseline ALP levels less than 120 U/L, hemoglobin levels of 10 g/dL or higher at baseline, good ECOG performance status at baseline, and in patients reporting no pain at baseline. (55) In an exploratory post-hoc analysis of the ALSYMPCA trial additional putative prognostic baseline variables were also assessed. (54) Multivariable analysis in this study showed that older age, poorer baseline ECOG performance status, lower albumin levels, and elevated baseline ALP, PSA and LDH levels were associated with worse OS in the group of patients treated with radium-223. Other, more recent, large retrospective cohort studies confirmed the importance of these prognostic baseline variables. (48, 50, 51) Furthermore, the influence of the extent of bone metastatic disease on OS has been described in several studies. The skeletal tumor burden on baseline bone scintigraphy, measured by the automated bone scintigraphy index, was significantly associated with OS and biochemical response to radium-223 in multivariable analyses. (56-59) Other studies found that quantitative analysis of skeletal tumor burden on baseline Fluoride-18 labeled Sodium Fluoride ( 18 F-NaF) or Choline ( 18 F-Choline) PET/CT was a predictor of OS, whereas visual analysis of bone scintigraphy showed no value in predicting OS. (60-62) However, the quantification of skeletal tumor burden prior to the initiation of radium-223 has not been implemented into daily clinical practice yet. Prospective studies comparing different imaging techniques to quantify skeletal tumor burden and the association with OS in patients who are treated with radium-223 are required. Furthermore, the construction of a prognostic model, such as the model by Halabi et al for patients who receive docetaxel chemotherapy, might aid in selecting the right patients for radium-223 therapy. (63) Previous studies have described that bone-related biochemical parameters are the main prognostic factors for OS in patients with metastatic prostate cancer. (34) ALP has been mostly described, both as a prognostic marker as well as a potential biomarker of response to life-prolonging therapies in patients with bone metastatic prostate cancer. (64-66) Several retrospective series have reported that elevated baseline ALP levels were

208