Data from the Danish cohort (n = 67) of a multicenter trial of oral alendronate in the prevention of postmenopausal osteoporosis were used to evaluate the capacity of the biochemical markers to predict changes in bone mineral density (BMD). A panel of markers were measured: serum N-terminal midfragment osteocalcin (N-MID OC); serum total osteocalcin (total OC); bone-specific alkaline phosphatase (BSAP); serum and urine C-telopeptides of type I collagen (sCL and uCL); urine N-telopeptide crosslinks of type I collagen (NTX); and deoxypyridinoline (dPyr). The correlation between change from baseline at months 3-12 in total OC, N-MID OC, sCL, uCL, and NTX and 2 year response in spine BMD ranged from r = -0.45 to r = -0.78 (p < 0.001), and from r = -0.38 to r = 0.10 (n.s. to p < 0.002) for BSAP and dPyr. Sensitivity and specificity were used to assess the accuracy of change from baseline at month 6 in the biochemical markers for predicting prevention of bone loss in the spine over 2 years. The cutpoints used were a 30% (N-MID OC) or 50% (all other markers) decrease from baseline. Sensitivity levels were 82% (N-MID OC), 98% (total OC), 78% (sCL and NTX), and 89% (uCL). Specificities were 91% (N-MID OC), 59% (total OC), 100% (sCL), 71% (uCL), and 84% (NTX). Positive predictive values were 95% (N-MID OC), 82% (total OC), 100% (sCL), 87% (uCL), and 90% (NTX). In comparison, the predictive capacities of change from baseline at year 2 in hip BMD in predicting prevention of bone loss at the spine were similar: sensitivity, 82%; specificity, 55%; and positive predictive value, 79%. In conclusion, short-term changes in biochemical markers were valid predictors of long-term changes in BMD. Short-term changes in the sensitive biochemical markers revealed a predictive capacity similar to bone densitometry at the hip measured over 2 years. The sensitive biochemical markers offered a fast and valid alternative to bone densitometry for monitoring of alendronate treatment.