Would warning of osteonecrosis of the jaw (ONJ) have induced a change in the administration schedule of zoledronic acid (ZA)? A report from Toledo, Spain.

e19596 Background: ZA therapy has been associated with the development of ONJ. In Nov 8th, 2005, osteonecrosis of the jaw was elevated to a safety warning by Spanish Drug Agency (SDA). The aim of this study was to investigate if that warning has had any effect in the pattern of ZA administration in a general hospital. Methods: We retrospectively analysed database using pharmacy, medical and nurse claims data from the Oncobass integrated claims database for all pts diagnosed with cancer who had received treatment with ZA between Jan 1, 2002, and Dec 31, 2011. Delivered doses of ZA were stratified in 2 groups: used before the SDA warning and after the SDA warning. Main variable: time between two consecutive doses (TTCD). Secondary variables: age, sex and diagnosis. Descriptive statistical analysis were applied for both groups. Comparisons of means of TTCD before and after the warning were performed with Student’s test, or the Mann-Whitney test, as necessary. Statistic significance: p value < 0.05. Results: We identified 761 patients (374 (49.2%) male; 387 (50.8%) female), median age 64 y (25 – 92) who were delivered 7301 doses [mean: 9.6 doses/patient (d/p)]: 2,353 (32.2%) before the SDA warning and 4948 (67.8%) after. Pathologies: breast cancer (BC) 257 patients (pts) (33,77%), 2,918 doses (40%), mean 11.4 d/p; multiple myeloma (MM) 167 pts (21.94%), 2001 doses (27%), 12 d/p; prostate adenocarcinoma (PA) 129 pts (16.95%), 1,510 doses (20.7%), 11.7 d/p; lung cancer (LC), 112 pts (14.72%), mean 4.2 d/p, 465 doses (6.4%); other 96 pts (12.61%), 407 doses (5,5%), 4.2 d/p. 3,126 (43%) doses were delivered to male and 4,149 (57%) to female. TTCD was significantly increased after ths SDA warning [32.12 vs 33.87 days (d); p< 0.001; CI 95% (-0.79)- (-0.90)]. By pathologies, differences were significant only for MM [35.8 vs 44.4 d; p<0.001; CI 95% (-10.19)-(-6.99)] and for PA [28.67 vs 32.03; p < 0.001; CI 95% (-5.08)-(-1.63)], but not for the other pathologies. Median d/p was unchanged before and after warnings for all pathologies except for MM (13 d/p vs 8.3 d/p). Conclusions: In our environment, ONJ warnings have not changed substantially the ZA schedule pattern, except for MM.