Abstract
High cobalamin levels have previously been associated with short-term cancer risk, including lung cancer. We explored whether levels of cobalamin and/or its binding proteins are useful as a diagnostic tool in patients suspected of non-small cell lung cancer. We included 889 patients referred for fast-track diagnosis of lung cancer to Aarhus University Hospital, Denmark. We analyzed plasma concentrations of cobalamin, transcobalamin, holotranscobalamin and haptocorrin. Information on lung cancer diagnosis was retrieved from a national database. The study cohort showed levels above reference intervals for cobalamin 12%, holotranscobalamin 25%, transcobalamin 9% and haptocorrin 36% (all p-values <.05). We observed no difference in cobalamin or holotranscobalamin levels when comparing patients diagnosed with non-small cell lung cancer (n = 161, 18%) to patients without lung cancer (n = 742, 80%), while transcobalamin showed minor differences. Haptocorrin was significantly higher in those with cancer, mainly among patients with adenocarcinoma (n = 94). A comparison of patients with the highest vs. lowest quartile levels of haptocorrin yielded an adjusted odds ratio for adenocarcinoma of 2.39 (95% confidence interval: 1.26–4.55). However, ROC curve analyzes showed haptocorrin (AUC = 0.55) and total transcobalamin (AUC = 0.56) to be poor diagnostic markers for lung cancer. A high proportion of patients suspected for non-small cell lung cancer showed increased levels of cobalamin-binding proteins. We thereby confirm the association between non-small cell lung cancer and high cobalamin levels and found that haptocorrin was the major underlying factor causing high cobalamin levels. However, none of these biomarkers were of diagnostic use among patients referred for suspected lung cancer.
Acknowledgement
We are grateful for the work performed by Birgitte Holst Folkersen as part of establishing the patient cohort.
Disclosure statement
The authors declare no conflicts of interest.