Introduction

Our group has previously demonstrated that blood-based tumor markers can be useful clinical outcome predictors for non-muscle invasive urothelial carcinoma of the bladder (UCB) Our aim in this study is to further evaluate the predictive value of CEA, CA 19-9 and CA 125 on disease recurrence and progression. _x000D_

Methods

We prospectively included 328 consecutive patients between February 2008 and August 2014 to measure preoperative serum levels of CEA, CA 19-9 and CA 125 before first transurethral resection of the bladder (TUR). Institutional Ethical Committee approval was obtained prior to this study. Patients diagnosed with pT2 UBC were excluded (42), leaving 286 patients for analysis of recurrence or progression. After first TUR, patients were followed with routine cystoscopy, cytology and ultrasound every 6 months. All patients with non-muscle invasive (NMI) bladder cancer with high-grade disease, previous recurrence, carcinoma in situ (CIS) or T1 received induction and maintenance intravesical BCG.

Results

We found that CEA and CA 19-9 levels were significantly higher in patients who had either tumor recurrence and/or progression compared to those who had no UBC recurrence during follow-up (p=0.02; p=0.03). As we had found previously, however, CA 125 levels did not differ between the two groups (p=0.42). Overall, mean CEA level was 2.1 (0.2-12.8), CA 19-9 was 17.1 (0.4-189.9) and CA 125 was 12.5 (1.2-103.9). In patients who presented tumor recurrence and/or progression, mean CEA was 5.5, mean CA 19-9 was 21.0 and CA 125 was 13.8, while in the non-recurring group, mean CEA was 3.1, mean CA 19-9 was 11.1 and CA 125 was 11.3. Mean follow-up was 4.9 years. Patients were 70.3% males (201); 63.3% (181) of patients had pTa at first TUR. Concomitant carcinoma in situ was present in 25 cases (8.7%).

Conclusions

Biomarkers utilized in routine follow-up of other malignancies, such as CEA and CA 19-9, can also be included in UCB management, since it proved able to distinguish a higher risk group of patients that could be managed accordingly. Future studies may add these blood-based tumor markers to a predictive model and validated in a larger cohort. Although CA 125 was not significantly associated with oncologic outcome, further studies are required before excluding this potential biomarker in UBC.

Funding

none