Systematic review with meta-analysis: volatile organic compound analysis to improve faecal immunochemical testing in the detection of colorectal cancer.

Abstract

Faecal immunochemical test (FIT) is emerging as a valid test to rule-out the presence of colorectal cancer (CRC). However, the accuracy of FIT is dependent on the cut-off applied. An additional low-cost test could improve further detection of CRC. To evaluate the efficacy of combined FIT and volatile organic compounds (VOC) in the detection of CRC within symptomatic populations. Systematic reviews on the diagnostic accuracy of FIT and VOC, for the detection of CRC, were updated. Meta-analyses were performed adopting a bivariate model for sensitivity and specificity. Clinical utility of combined FIT and VOC was estimated using Fagan's nomogram. Post-test probability of FIT negatives was used as a pre-test probability for VOC. The pooled sensitivity and specificity of FIT at 10µg/g faeces, for the detection of CRC, were 0.914 (95% confidence interval [CI]=0.894-0.936) and 0.783 (CI=0.850-0.696), respectively. For VOC, the sensitivity was 0.837 (CI=0.781-0.881) and the specificity was 0.803 (CI=0.870-0.712). The area under the curve for FIT and VOC were 0.926 and 0.885, respectively. In a population with 5% CRC prevalence, the estimated probability of having CRC following a negative FIT was 0.5% and following both negative FIT and VOC was 0.1%. In a FIT-negative symptomatic population, VOC can be a good test to rule-out the presence of CRC. The estimated probability reduction by 0.4% when both tests being negative offers adequate safety netting in primary care for the exclusion of CRC. The number needed to colonoscope to identify one CRC is eight if either FIT or VOC positive. Cost-effectiveness and clinical accuracy of this approach will need further evaluation.