To establish whether biochemical markers could be used to monitor alendronate (ALN) treatment and predict long-term response in bone mass, we used results from an ongoing, randomized trial of ALN treatment for prevention of postmenopausal osteoporosis (n = 1202). In women treated with ALN (5 mg), change from baseline at month 6 in urine N-telopeptide cross-links of type I collagen (NTX) and osteocalcin (OC) correlated with change from baseline at month 24 in spine, hip, and total body bone mineral density (BMD) [r = -0.28 to -0.31 (NTX) and r = -0.16 to -0.25 (OC), P<0.001]. This corresponded to a 4- to 5-fold greater increase at month 24 in BMD in the tertiles, with the greatest decrease at month 6 in NTX or OC. In women treated with ALN (5 mg) who had a change at month 24 in spine BMD of at least 0%, 86% (NTX) and 79% (OC) had a decrease at month 6 of at least 40% (NTX) or 20% (OC) (sensitivity). The corresponding specificities were 48% (NTX) and 53% (OC). In conclusion, change at month 6 in NTX and OC, in groups of women treated with ALN, indicated the numeric long-term response in BMD within these groups. In individual women, a decrease at month 6, in NTX or OC below the cut-point, validly identified women who responded, on ALN treatment, with a stabilization or an increase in bone mass. However, lack of decrease below the cut-point in NTX or OC could not be used to identify women with a bone loss during ALN treatment.