Long-term outcomes in a high-risk bladder cancer screening cohort

To evaluate the long-term outcomes in patients at high risk of bladder cancer who participated in a bladder cancer screening trial.

Patients who were classified as high risk based on age ≥50 years, ≥10 pack-years (combination of packs of tobacco per day and years of smoking) smoking and/or ≥15 years environmental exposure were enrolled in a one-time screening trial using a nuclear matrix protein 22 (NMP22) assay, between March 2006 and November 2007, at Dallas Veterans Affairs Hospital. We assessed the subsequent detection of smoking-related malignancies (bladder, lung and renal cell carcinoma [RCC]) in these patients up until 31 January 2014. Multivariable regression analysis was used to determine factors associated with bladder cancer diagnosis and survival.

The study cohort included 925 patients, of whom 886 (95.8%) were smokers and 613 (66.3%) had received hazardous occupational exposure. At initial screening, 57 patients had a positive NMP22 test and two had bladder cancer. Another nine patients (1.0%) were diagnosed with bladder cancer during the median follow-up of 78.4 months. The bladder cancers were non-invasive (Ta) and seven were low grade and four high grade. RCC and lung cancer were diagnosed in 10 (1.1%) and 18 patients (1.9%), respectively. A total of 134 patients died, including three from RCC and 12 from lung cancer, but none from bladder cancer. Factors associated with worse overall survival on multivariable analysis were: lung cancer (hazard ratio [HR] 5.06; P < 0.001), microscopic or gross haematuria (HR 1.66; P = 0.006 and HR 2.11; P = 0.02, respectively), and >60 pack-years smoking history (HR 4.51; P = 0.037).

At 6.5 years of follow-up, no patients in this high-risk cohort developed muscle-invasive bladder cancer. Lung cancer, haematuria and >60 pack-years smoking history are independent predictors of mortality. Other-cause mortality is an important consideration in patients undergoing bladder cancer screening.