BACKGROUND: The prognostic relevance of disseminated cytokeratin-positive (CK+) tumor cells in the bone marrow of patients with different types of carcinoma has been demonstrated in several studies. In this prospective study, the frequency and prognostic value of CK+ tumor cells was investigated in the bone marrow of 55 consecutive patients with metastatic renal cell carcinoma (M1 RCC) in comparison with 256 M0 RCC patients from a previous study. METHODS: Aspiration of bone marrow from the anterior iliac crest was performed immediately before tumor resection in RCC patients. Cytospins were made and stained by immunocytochemistry using the APAAP (alkaline phosphatase-antialkaline phosphatase) protocol and monoclonal antibodies CK2 and A45-B/B3. Twenty-seven patients with no evidence of any malignant disease served as a control group. RESULTS: CK+ tumor cells were detected in 42% (23 of 55 patients) of the M1 patients and 25% (63 of 256 patients) of the M0 patients (P <.01). No CK+ cells (0 of 27 patients) were detected in the control group. In the M1 group, CK- patients demonstrated a trend toward a better outcome compared with CK+ patients (log-rank test, P = .19). This difference was significant when applying a higher threshold (0-2 CK+ cells vs. > or = 3 CK+ cells; P <.05). On multivariate analysis, the detection of > or = 3 CK+ cells in the bone marrow was found to be an independent prognostic factor (P <.001). CONCLUSIONS: The results of the current study indicate that disseminated CK+ cells play a role in the biology of tumor spread of RCC, and that their immunocytochemical detection can be useful in assessing the prognosis of patients with M1 disease. Copyright 2006 American Cancer Society.
BACKGROUND: The prognostic relevance of disseminated cytokeratin-positive (CK+) tumor cells in the bone marrow of patients with different types of carcinoma has been demonstrated in several studies. In this prospective study, the frequency and prognostic value of CK+ tumor cells was investigated in the bone marrow of 55 consecutive patients with metastatic renal cell carcinoma (M1 RCC) in comparison with 256 M0 RCCpatients from a previous study. METHODS: Aspiration of bone marrow from the anterior iliac crest was performed immediately before tumor resection in RCCpatients. Cytospins were made and stained by immunocytochemistry using the APAAP (alkaline phosphatase-antialkaline phosphatase) protocol and monoclonal antibodies CK2 and A45-B/B3. Twenty-seven patients with no evidence of any malignant disease served as a control group. RESULTS: CK+ tumor cells were detected in 42% (23 of 55 patients) of the M1 patients and 25% (63 of 256 patients) of the M0 patients (P <.01). No CK+ cells (0 of 27 patients) were detected in the control group. In the M1 group, CK- patients demonstrated a trend toward a better outcome compared with CK+ patients (log-rank test, P = .19). This difference was significant when applying a higher threshold (0-2 CK+ cells vs. > or = 3 CK+ cells; P <.05). On multivariate analysis, the detection of > or = 3 CK+ cells in the bone marrow was found to be an independent prognostic factor (P <.001). CONCLUSIONS: The results of the current study indicate that disseminated CK+ cells play a role in the biology of tumor spread of RCC, and that their immunocytochemical detection can be useful in assessing the prognosis of patients with M1 disease. Copyright 2006 American Cancer Society.