BACKGROUND: In breast cancer, α(ν)β(3) and/or α(ν)β(5) integrins are overexpressed in both endothelial and tumour cells. Radiolabelled peptides based on the Arg-Gly-Asp (RGD) sequence are radiopharmaceuticals with high affinity and selectivity for these integrins. The RGD-dimer peptide (E-[c(RGDfK)]2) radiolabelled with (99m)Tc has been reported as a radiopharmaceutical with a 10-fold higher affinity for the α(ν)β(3) integrin compared with the RGD-monomer. Ethylenediamine-N,N'-diacetic acid (EDDA) is a hydrophilic molecule that may favour renal excretion when used as coligand in the (99m)Tc labelling of hydrazinonicotinamide (HYNIC) peptides and can easily be formulated in a lyophilized kit. AIM: The aim of this study was to establish a biokinetic model for (99m)Tc-EDDA/HYNIC-E-[c(RGDfK)]2 prepared from lyophilized kits and evaluate its dosimetry as a tumour-imaging agent in seven healthy women and three breast cancer patients. MATERIALS AND METHODS: (99m)Tc labelling was performed by adding sodium pertechnetate solution and 0.2 mol/l phosphate buffer (pH 7.0) to a lyophilized formulation containing E-[c(RGDfK)]2, EDDA, tricine, mannitol and stannous chloride. The radiochemical purity was evaluated using reverse-phase high-performance liquid chromatography and instant thin-layer chromatography on silica gel analyses. Stability studies in human serum were carried out using size-exclusion high-performance liquid chromatography. In-vitro cell uptake was tested using breast cancer cells (MCF7, T47D and MDA-MB-231) with blocked and nonblocked receptors. Biodistribution and tumour uptake were determined in MCF7 tumour-bearing nude mice with blocked and nonblocked receptors, and images were obtained using a micro-SPECT/PET/CT. Whole-body images from seven healthy women were acquired at 0.5, 1, 3, 6 and 24 h after (99m)Tc-EDDA/HYNIC-E-[c(RGDfK)]2 administration with radiochemical purities greater than 94%. Regions of interest were drawn around the source organs at each time frame. Each region of interest was converted to activity using the conjugate view counting method. The image sequence was used to extrapolate the (99m)Tc-EDDA/HYNIC-E-[c(RGDfK)]2 time-activity curves of each organ to adjust the biokinetic model and calculate the total number of disintegrations (N) that occurred in the source regions. N data were the input for the OLINDA/EXM code to calculate internal radiation dose estimates. In three breast cancer patients with histologically confirmed cancer, static images were obtained at 1 h in the supine position with hands placed behind the head. RESULTS: (99m)Tc-EDDA/HYNIC-E-[c(RGDfK)]2 obtained from lyophilized kits demonstrated high stability in human serum and specific cell receptor binding. The biodistribution data from mice showed rapid blood clearance, with both renal and hepatobiliary excretion, and specific binding towards α(ν)β(3) integrins in the MCF7 tumours. In women, the blood activity showed a half-life value of 1.60 min for the fast component (T1/2α = ln2/26.01) and half-life values of 1.0 h for the first slow component (T1/2β = ln2/0.69) and 4.03 h for the second slow component (T1/2γ = ln2/0.16). Images from patients showed an average tumour/heart (blood) ratio of 3.61 ± 0.62 at 1 h. The average equivalent doses calculated for a study using 740 MBq were 4.9, 6.2, 20.7, 34.5 and 57.0 mSv for the liver, intestines, spleen, kidneys and thyroid, respectively, and the effective dose was 6.1 mSv. CONCLUSION: All absorbed doses were comparable to those known from most of the (99m)Tc studies. (99m)Tc-EDDA/HYNIC-E-[c(RGDfK)]2 obtained from kit formulations showed high tumour uptake in patients with malignant lesions, making it a promising imaging radiopharmaceutical for targeting site-specific breast cancer. The results obtained in this study warrant further clinical studies to determine the specificity and sensitivity of (99m)Tc-EDDA/HYNIC-E-[c(RGDfK)]2.