There remains an urgent need for non-invasive, low cost methods for diagnosis of colorectal cancer (CRC). We undertook a diagnostic accuracy study using faecal haemoglobin (F-Hb), faecal calprotectin (FCP) and urinary volatile organic compounds (VOCs) in patients presenting with lower gastrointestinal symptoms referred via Two Week Wait colorectal pathway.Methods
Of 1850 patients approached, 1016 were recruited prospectively. Of these, 562 with complete colonic investigations returned matched urine and stool samples and were included in the final statistical analysis.Results
The sensitivity and specificity for CRC using F-Hb were 0.80 (95% confidence interval (CI): 0.66–0.93) and 0.93 (CI: 0.91–0.95) respectively. The negative predictive value (NPV) was 0.99 (CI: 0.98–1.0). Using urinary VOCs the sensitivity and specificity were 0.63 (CI: 0.46–0.79) and 0.63 (CI: 0.59–0.67) respectively and the NPV was 0.96 (CI: 0.94–0.98). For those with F-Hb negative CRC (false negatives), adding urinary VOCs revealed the sensitivity of 0.97 (CI: 0.90–1.0) and specificity of 0.72 (CI: 0.68–0.76) with the NPV of 1.0 (CI: 0.99–1.0).Conclusions
Urinary VOCs applied to a F-Hb negative group excludes CRC with the NPV of 1.0. Thus, the addition of urinary VOCs shows promise as a second stage test in investigating symptomatic population.