There is an unmet need to develop imaging methods for the early and objective assessment of breast tumors to therapy. 3'-Deoxy-3'-[18F]fluorothymidine ([18F]FLT)-positron emission tomography represents a new approach to imaging thymidine kinase activity, and hence, cellular proliferation. We compared graphical, spectral, and semiquantitative analytic methodologies for quantifying [18F]FLT kinetics in tumor and normal tissue of patients with locally advanced and metastatic breast cancer. The resultant kinetic parameters were correlated with the Ki-67 labeling index from tumor biopsies. [18F]FLT accumulation was detected in primary tumor, nodal disease, and lung metastasis. In large tumors, there was substantial heterogeneity in regional radiotracer uptake, reflecting heterogeneity in cellular proliferation; radiotracer uptake in primary tumors also differed from that of metastases. [18F]FLT was metabolized in patients to a single metabolite [18F]FLT-glucuronide. Unmetabolized [18F]FLT accounted for 71.54 +/- 1.50% of plasma radioactivity by 90 minutes. The rate constant for the metabolite-corrected net irreversible uptake of [18F]FLT (Ki) ranged from 0.6 to 10.4 x 10(-4) and from 0 to 0.6 x 10(-4) mL plasma cleared/s/mL tissue in tumor (29 regions, 15 patients) and normal tissues, respectively. Tumor Ki and fractional retention of radiotracer determined by spectral analysis correlated with Ki-67 labeling index (r = 0.92, P < 0.0001 and r = 0.92, P < 0.0001, respectively). These correlations were superior to those determined by semiquantitative methods. We conclude that [18F]FLT-positron emission tomography is a promising clinical tool for imaging cellular proliferation in breast cancer, and is most predictive when analyzed by graphical and spectral methods.