Pretargeting techniques that are based on the sequential administrations of bispecific antitumor/antimetal chelate antibodies (BS-MAbs), a blocker to saturate the anti-chelate binding sites of the BS-MAb still present in the circulation, and the radiolabeled chelate are suitable to increase tumor-to-normal tissue contrasts and enable positron emission tomography (PET) as an imaging method. As demonstrated in the nude mouse model, a combination of pretargeted immunoscintigraphy and PET markedly improved the detection of tumor xenografts. With the presented preliminary clinical trial, we attempted to assess the efficacy of pretargeting and PET for breast cancer localization in patients. The BS-MAb used for pretargeting was synthesized from the F(ab')(2) fragments of the anti-MUC1 MAb 12H12, which reacts with the vast majority of breast tumors, and the F(ab') fragments of an anti-gallium (Ga) chelate MAb via a mixed functional chemical linker. For labeling of the Ga-chelate, we used the short-lived positron emitter Ga-68 (t(1/2), 68 min; beta(+), 88%). The dose and time schedule of pretargeting was deduced from previous animal experiments. Ten patients with biopsy-proven, primary breast carcinoma were infused with 10 mg of the BS-MAB: Eighteen h later, they received i.v. injections of 10.7 mg of a blocker and, 15 min later, 9.6 microg of the Ga chelate labeled with 230-300 MBq of (68)GA: PET imaging was started 60-90 min after injection of the (68)Ga chelate. Average tumor-to-blood and tumor:normal breast tissue ratios were 0.9 and 3.0 at 1 h postinjection. Tumor uptake amounted to approximately 0.003% iD/g corresponding to a standard uptake value of approximately 2. Blood clearance of the (68)Ga chelate showed a t(1/2) beta of approximately 100 min. Fourteen of 17 known lesions, averaging 25 +/- 16 mm in size, were clearly visualized as foci of increased activity with PET. No false-positive but three false-negative readings were obtained. An enhanced, bilateral activity uptake in the whole breast parenchyma, found in 4 of the 10 patients, compromised the recognition of these tumor sites. Although the shedding of the MUC1 antigen and the comparatively low tumor affinity of the BS-MAb, common to all anti-mucin MAbs, proved not to be optimal for increasing tumor:tissue ratios with a pretargeting technique, PET imaging offered better sensitivity for the detection of breast cancer at low tumor contrasts than conventional immunoscintigraphy. This could be demonstrated by the clear visualization of tumor sites 10 mm in size, which contrasted only by a factor of 2 from surrounding normal breast tissue.