Introduction: Sunitinib malate is an oral multitargeted tyrosine kinase inhibitor of VEGFRs, PDGFRs, KIT, RET and FLT3. To characterize potential biomarkers of sunitinib response, we analyzed plasma levels of a panel of soluble proteins from metastatic CRC (mCRC) patients (pts) in a study of sunitinib monotherapy after failure of standard chemotherapy, clinical results of which have been reported previously.

Methods: In this open-label, multicenter, phase II trial, 82 pts with mCRC previously treated with irinotecan, oxaliplatin and a fluoropyrimidine, with or without bevacizumab (BV), received sunitinib (50 mg/day) in repeated 6-wk cycles consisting of 4 wks on, followed by 2 wks off treatment. Pre-dose plasma samples were obtained on day 1 of the first sunitinib cycle, and on-treatment samples were taken on days 14 and 28 of cycle 1 and on days 1 and 28 of subsequent cycles. Plasma levels of VEGF, sVEGFR-2, sVEGFR-3 and sKIT were analyzed by ELISA.

Results: Mean baseline plasma VEGF levels were 6.9-fold higher in the BV-treated group than in the BV-naïve group (P<0.001), with a significant inverse correlation between time since cessation of prior BV therapy (median, 46 days) and baseline VEGF level (R2=0.61; baseline VEGF was 765 pg/mL vs. 311 pg/mL when time since cessation was below or above the median, respectively). Baseline levels of other plasma proteins did not differ significantly between the BV groups. At the end of the first sunitinib treatment cycle, plasma VEGF levels increased 3.9-fold above baseline on average (range, 0.4-26.1), whereas levels of sVEGFR-2, sVEGFR-3 and sKIT decreased by 45.2%, 40.5% and 27.3%, respectively. For all proteins except sKIT, plasma levels returned to near-baseline during the 2-wk period off treatment and then exhibited changes in subsequent cycles of treatment similar to those seen in the first cycle. In contrast, sKIT levels continued to decline progressively, leveling off after three cycles. Changes in VEGF, sVEGFR-2 and sVEGFR-3 during cycle 1 were significantly greater in BV-naïve pts than in BV-treated pts (P<0.05); changes in sKIT levels did not correlate with prior BV treatment. When pts were stratified based on presence or absence of a decrease in target lesion size (on at least one assessment), decreases correlated significantly with greater changes from baseline in VEGF and sVEGFR-3 levels during cycle 1 (P=0.027 and 0.0001, respectively), but not with changes in sVEGFR-2 or sKIT levels. Analyses of the sub-populations with or without prior BV treatment revealed a similar trend for VEGF in the BV-naïve, but not in the prior-BV cohort.

Conclusions: Changes in the levels of plasma proteins observed in response to sunitinib treatment and the influence of prior BV treatment on these changes suggest that sunitinib may act, at least in part, through modulation of the VEGF and KIT pathways.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA