Patients and Methods: We examined 45 peripheral blood samples and 30 bone marrow samples from patients with a variety of nonneoplastic conditions using nested RT-PCR for CK-19, We also examined bone marrow and peripheral-blood samples from 23 patients with primary breast cancer and peripheral-blood samples from 37 patients with metastatic breast cancer. The number of CK-19 transcripts was estimated in positive specimens by competitive PCR and normalized to the number of ABL transcripts as an internal control for the quality and quantity of cDNA, RT-PCR results were compared with the numbers of CK-19-positive cells detected by immunocytochemistry,

Results: Analysis of samples from patients without cancer enabled us to define an upper limit for the background ratio of CK-19 to ABL transcripts (1:1,000 for blood samples and 1:1,600 for bone marrow samples). Using these figures as cut-off points, elevated CK-19: ABL ratios were detected in peripheral-blood samples of 20 of 37 (54%) patients with metastatic breast cancer and in bone marrow samples of 14 of 23 (61%) patients with primary breast cancer. Only three of 23 (13%) primary breast cancer peripheral-blood samples and none of the control samples were positive by these criteria, Only two of 23 patients (9%) with primary breast cancer showed immunocytochemically detectable cells in the blood; 10 of 23 (43%) showed immunocytochemically detectable cells in the bone marrow,Of 36 patients with metastatic breast cancer, eight (22%) showed positive events.

Conclusion: Quantitative RT-PCR for CK-19 detects a percentage of patients with breast cancer and may enable the progression or regression of the disease to be monitored. (C) 1999 by American Society of Clinical Oncology.