| dc.description.abstract | Background. Monitoring patients with renal cancer for recurrence or progression is ineffective, inconvenient and costly. We conducted a retrospective analysis of serum CA 15-3 concentrations in a single-centre patient database and examined its impact on prognosis and follow-up. Material and methods. Serum CA 15-3 levels were studied in 46 patients (26 male, 20 female, median age 61) with histologically confirmed renal cancer. Distribution for stage I/II/III/IV disease was 15/12/4/15, respectively. All but one patient underwent curative nephrectomy or biopsy. Metastases were diagnosed in 26 patients. Six patients developed lung metastases, 6 bony, 3 liver, 4 nodal and 9 patients had deposits in multiple organ systems. Results. Abnormal serum CA 15-3 values were detected in 23% of patients (6/26) with metastatic disease at the time of diagnosis of metastases. Median concentrations for patients with localised and metastatic disease were 17 and 26 U/ml, respectively (Student's t-test p = 0.019, upper limit of reference range 31.3). High marker levels were commoner in patients with lung metastases (4/11), with a median CA 15-3 value of 89 U/ml. In 11 patients with metastases prospectively followed-up, the correlation between a more than two-fold serum marker concentration increase with disease progression yielded sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 62.5, 100, 100 and 50%, respectively, not of statistical significance. The correlation between detection of marker increase and cancer relapse in 31 patients with resected tumours yielded sensitivity, specificity, PPV and NPV of 40, 95, 80 and 77%, respectively, reaching marginal statistical significance (p = 0.05). At a median follow-up of 28 months, there was a trend for better 2-year overall survival in patients with normal versus high baseline marker concentrations (89 versus 53%, Logrank p = 0.05). Among metastatic disease patients, the 2-year overall survival difference (79 versus 60%) reached statistical significance (Logrank p = 0.035). Conclusion. Despite the lack of sensitivity and a poor correlation with tumour burden, monitoring of serum CA 15-3 levels may contribute towards a more effective follow-up and treatment of patients with renal cancer, especially in the subgroup of patients with pulmonary metastases. The indications for prognostic significance of baseline marker levels in terms of overall survival may be interpreted on the basis of aggressive phenotype or tumour-host immunological interactions. | en |
