Abstract
Despite the current debate on the best therapeutic approach, i.e. symptomatic vs intensive strategy, one zoledronate (Zol) infusion is effective in most patients with Paget's disease of bone (PDB), whereas few need retreatment, whose predictors are not well established. We aimed to evaluate long-term efficacy of intensive Zol therapy and predictors of retreatment in PDB. Pagetic complications, clinical and biochemical response to Zol together with frequency of retreatment were retrospectively assessed in forty-seven PDB patients (age, mean ± SD: 72.5 ± 8.9 years, M/F: 24/23; symptomatic/asymptomatic: 16/31). Statistical analysis for retreatment prediction were based on Mann-Whitney U test, Pearson's Χ2 and ROC curve analysis. During seven-year follow-up, all patients achieved pain relief and only one underwent arthroplasty. Bone alkaline phosphatase (BAP) detected three non-responder (6%) and six relapsing (13%) patients needing retreatment. Retreated patients had less old age (66.1 ± 11.2 vs 74.0 ± 7.7 years), higher frequency of polyostotic disease (78% vs 40%) and higher baseline (96.5 ± 24.8 vs 44.9 ± 27.7 mcg/l) and post-Zol nadir BAP levels (24.7 ± 24.1 vs 8.1 ± 4.1 mcg/l) than patients treated once (p < 0.05 for all comparisons). In multivariate analysis both serum baseline and post-Zol nadir BAP significantly predicted retreatment (OR 1.09, 95%CI 1.01-1.17 and 1.29, 1.03-1.62, respectively), with ROC curve analysis showing the greatest accuracies for threshold values of 75.6 and 9.9 mcg/l (sensitivity 88 and 90%, specificity 94 and 86%, AUC 0.92 and 0.93, respectively). Our data in mostly asymptomatic, metabolically active PDB patients treated with intensive Zol therapy show a negligible incidence of pagetic complications and long-term optimal disease control, with BAP being the best predictor of retreatment.