Randomized trial of marker-directed versus standard schedule zoledronic acid for bone metastases from breast cancer.

Abstract

Listen

Translate article

511 Background: Zoledronic acid (ZOL) reduces skeletal morbidity associated with metastatic bone disease by 30-50%. Current recommendations are for indefinite administration of intravenous 3-4 weekly (w) ZOL 4mg. Over recent years it has become clear that the risk of skeletal morbidity is related to the rate of bone resorption. We have compared a marker directed schedule of ZOL (M-ZOL), using measurements of urinary n-telopeptide of type I collagen (NTX), to a standard treatment schedule (S-ZOL) in patients with bone metastases from breast cancer. Methods: The primary endpoint was skeletal related events (SRE). A non-inferiority study was designed with 80% power and (one-sided) 5% alpha to demonstrate that M-ZOL retained 67% of the efficacy of S-ZOL. This required 1500 patients, assuming a SRE rate of 0.7/year. Following minimisation for known prognostic factors, patients were randomised to receive S-ZOL 3-4 w or M-ZOL (15-16w; 8-9 w or 3-4w if NTX levels were <50, 50-100, >100 nmol/mmol creatinine respectively), with the schedule adjusted according to NTX measured every 16 weeks. The study duration was 24 months. Results: Due to lower than expected recruitment, the study closed in 2009 following recruitment of 289 patients. 90% of patients had received >4 cycles of ZOL or pamidronate prior to randomisation. The median number of ZOL infusions administered to S-ZOL patients was >2x that received on M-ZOL. 46 (32%) S-ZOL and 55 (38%) M-ZOL patients experienced an SRE. The numbers of SRE were 94 and 138 in the S-ZOL and M-ZOL arms, with the excess in SRE being largely due to more patients on M-ZOL experiencing ≥2 SRE. Multivariate analysis adjusting for key minimisation factors and baseline NTX for all SRE showed a hazard ratio for M-ZOL vs. S-ZOL of 1.41 (90%CI 0.98-2.02, p=.12). NTX levels were significantly higher at all time points in the M-ZOL treated patients. Osteonecrosis of the jaw was uncommon with 3 cases with S-ZOL and 1 with M-ZOL. Conclusions: The study is underpowered to demonstrate non-inferiority in SRE outcome between the treatment strategies. However, the results suggest that the adjustment of ZOL schedule based on NTX values alone may represent sub-optimal management.