Early diagnosis of recurrent breast cancer with FDG-PET in patients with progressive elevation of serum tumor markers.

ABSTRACT

BACKGROUND: The aim of this work is to assess the diagnostic value of positron emission tomography (PET) with 18F-fluorodeoxyglucose (FDG), in the early detection of tumour recurrence in already treated breast cancer patients in apparent complete remission and with a progressive elevation of tumour markers CEA and/or CA 15.3 without any other clinical or instrumental signs of relapses. METHODS: The author studied 45 women (mean age 58+/-12, range 35-80 years) with histological diagnosis of breast cancer who underwent a tumour marker-guided whole body FDG-PET. All patients were in remission, without any other clinical or instrumental signs of relapses, except for the progressive elevation of CA 15.3 and/or CEA, tested during the follow-up. FDG-PET results were controlled by pathology when histological sampling was possible, by other conventional imaging modalities (US, X-rays, CT, MRI) and/or by clinical follow-up up to 12 months at least. RESULTS: FDG-PET findings were evaluated in 38 patients 27 resulted positive. Among these 27 PET positive patients 24 were true positive and 3 false positive. Tumour marker guided FDG-PET was also able to discover 3 unknown neoplasms not visualized by other modalities. PET revealed 54 sites of intense focal FDG uptake. The anatomical distribution of these sites was 19 skeleton, 18 lymph node basins, 5 liver, 5 pelvic region, 1 lung, 1 pericardium, 1 pleura, 1 contralateral breast, 2 peritoneum and 1 thyroid bed. Forty-eight of these 54 sites of FDG accumulation were confirmed to be metastases. FDG-PET resulted negative in 11 patients and only in 2 of them the other diagnostic modalities were able to discover metastatic lesions; we had 9 true negative and 2 false positive RESULTS. On the basis of our investigation the performances of tumour marker guided FDG-PET per patient are as follows sensitivity 92% (24/26), specificity 75% (9/12), positive predictive value 89% (24/27), negative predictive value 82% (9/11), accuracy 87% (33/38). CONCLUSIONS: This study demonstrated the clinical utility of tumour marker-guided PET in the follow-up of breast cancer patients. This diagnostic approach allowed to modify the clinical management in those patients in whom a tumor relapse or unexpected primary neoplasm was discovered.