Aims: Colorectal resection (CR) increases plasma VEGF levels which may promote residual tumor growth. This study assessed the effect of perioperative GMCSF on plasma levels of sVEGFR1, Ang-1 and Ang-2 and also the impact of post-GMCSF plasma on in vitro endothelial cell (EC) growth and invasion. Ang-2 increases while sVEGFR1 and Ang-1 impede angiogenesis.
Methods: Fifty-nine CR cancer patients were randomized to 7 perioperative doses of GMCSF or saline for 3days prior and 4days after CR. Blood samples were taken pre-drug (PreRx) and on several postoperative days (POD). Protein levels were assessed and PreRx and POD 5 plasma added to EC cultures after which branch point formation (ECBPF) and invasion (ECI) were measured.
Results: sVEGFR1 levels were significantly higher on POD 1 and POD 5 in both groups but the GMCSF POD 5 level was twice the control value (p=0.002). Ang-2 levels were higher on PODs 1 and 5 in both groups (p<0.05) but the control POD 5 value (vs. GMCSF) was greater (p=0.03). Ang-1 decreases were noted in all (p=not significant, ns). The control group POD 5 ECBPF was 35.8% greater than Pre Rx (p=0.001) while the GMCSF result was 18.0% lower (p=ns); the control POD 5 median percent change from baseline was greater than the GMCSF result(p=0.008). The POD 5 ECI was +12.2% for the control group vs. baseline (p=ns) and -17.2% for the GMCSF group (p=ns): the control median percent change was greater than in the GMCSF group(p=0.045).
Conclusion: CR-related plasma changes are proangiogenic (>Ang-2) and anti-angiogenic (>sVEGFR1); the net effect is promotion of in vitro ECBPF. GMCSF limits the proangiogenic changes (higher POD 5 sVEGFR1 levels and lower Ang-2 elevations, lower POD 5 ECBPF and ECI). The clinical import of these effects is unclear; perioperative GMCSF has anti-angiogenic plasma effects that may limit tumor growth. Further investigation is warranted.