Introduction. Membrane-bound dipeptidyl peptidase 5 (DPP5) has been linked to tumor invasion and metastasis processes. Its recent isolation as a soluble protein from plasma may indicate a potential role as a biomarker and prognostic factor for tumors, and as a potential therapeutic target. This study aimed to measure the plasma levels of DPP5 and the related member of the family, DPP4, in healthy subjects and cancer patients, and to determine their clinical significance.

Methods. Enzyme-linked immunosorbent assays (ELISAs) for measuring plasma DPP5 and DPP4 were developed and clinical correlations were made.

Results. ELISAs were validated to be sensitive and specific. In the 747 plasma samples studied, mean DPP5 and DPP4 levels were 0.51±0.30 and 4.65±6.37 μg/mL, respectively. Plasma DPP5 levels were highly correlated with DPP4 levels. Plasma DPP5 and DPP4 levels were significantly lower in cancer patients compared with healthy subjects (4.38 versus 5.65 μg/mL μg/mL, p<0.001 for DPP4; 0.46 versus 0.66 μg/mL, p<0.001 for DPP5). DPP5 and DPP4 levels were higher in younger patients and those with earlier-stage cancers, but also in metastatic cancers; stage III cancer patients had the lowest plasma levels. Cancer patients with low DPP4 levels had shorter survival than those with higher DPP4 levels (p=0.001). There was also a trend toward shorter survival in cancer patients with low DPP5 levels in various cancer types, with the exception of colorectal cancers. An in vitro model indicated that approximately 80% of DPP5 released from the cells into conditioned medium is free (not associated with membranes), and serine protease inhibitors interfere with release of DPP5 into medium.

Conclusions. This study demonstrates that plasma DPP5 and DPP4 levels are reproducible parameters to correlate with disease status, with lower levels noted in cancer patients compared with healthy subjects; but higher levels in metastatic cancers compared with non-metastatic cancers. These findings support a prognostic role for plasma DPP5 and DPP4 levels and they should be considered in attempts to use DPP5 as a therapeutic target to hinder cancer progression. An explanation of low levels of plasma DPP5 in cancer despite previously-reported high level of local expression remains to be determined.