Plasma Protein Biomarkers in Advanced or Metastatic Colorectal Cancer Patients Receiving Chemotherapy With Bevacizumab or Cetuximab: Results from CALGB 80405 (Alliance)

Abstract

Purpose: CALGB 80405 compared the combination of first-line chemotherapy with cetuximab or bevacizumab in the treatment of advanced or metastatic colorectal cancer (mCRC). Although similar clinical outcomes were observed in the cetuximab-chemotherapy group and the bevacizumab-chemotherapy group, biomarkers could identify patients deriving more benefit from either biologic agent.

Patients and methods: In this exploratory analysis, the Angiome, a panel of 24 soluble protein biomarkers were measured in baseline plasma samples in CALGB 80405. Prognostic biomarkers were determined using univariate Cox proportional hazards models. Predictive biomarkers were identified using multivariable Cox regression models including interaction between biomarker level and treatment.

Results: In the total population, high plasma levels of Ang-2, CD73, HGF, ICAM-1, IL6, OPN, TIMP-1, TSP-2, VCAM-1, and VEGF-R3 were identified as prognostic of worse progression-free survival (PFS) and overall survival (OS). PlGF was identified as predictive of lack of PFS benefit from bevacizumab [bevacizumab HR, 1.51; 95% confidence interval (CI), 1.10-2.06; cetuximab HR, 0.94; 95% CI, 0.71-1.25; Pinteraction = 0.0298] in the combined FOLFIRI/FOLFOX regimens. High levels of VEGF-D were predictive of lack of PFS benefit from bevacizumab in patients receiving FOLFOX regimen only (FOLFOX/bevacizumab HR, 1.70; 95% CI, 1.19-2.42; FOLFOX/cetuximab HR, 0.92; 95% CI, 0.68-1.24; Pinteraction = 0.0097).

Conclusions: In this exploratory, hypothesis-generating analysis, the Angiome identified multiple prognostic biomarkers and two potential predictive biomarkers for patients with mCRC enrolled in CALGB 80405. PlGF and VEGF-D predicted lack of benefit from bevacizumab in a chemo-dependent manner. See related commentaries by Mishkin and Kohn, p. 2722 and George and Bertagnolli, p. 2725.

Figure 1.. Patient CONSORT diagram.

Bev: bevacizumab; Cetux: cetuximab.

Figure 2.. Prognostic biomarkers.

Forest plots of hazard ratios (HRs) illustrate the prognostic association of biomarkers for both overall survival (A) and progression-free survival (B) in the overall population. All 10 were negative prognostic indicating that higher biomarker levels are associated with higher HR or favors lower expression.

Figure 3.. Kaplan-Meier curves of PlGF and treatment for overall survival (Top) and progression-free survival (Bottom).

For illustrative purposes, PlGF levels were dichotomized as “high” or “low” relative to the median. P-values from the test for interaction of the marker as a continuous measure are reported.

Figure 4.. Kaplan-Meier curves showing the effects of VEGF-D levels and chemotherapy on overall survival (Top panels) and progression-free survival (Bottom panels).

For illustrative purposes, VEGF-D levels were dichotomized as “high” or “low” relative to the first quartile. P-values from the test for interaction of the maker as a continuous measure are reported. Patients with VEGF-D levels in the lowest quartile receiving FOLFOX as the chemotherapy derived more benefit from bevacizumab (the interaction p-value for the marker as a continuous measure was 0.0097), as shown in the middle bottom figure.

Figure 5.. Predictive effect of VEGF-D.

Forest plots of VEGF-D quantiles in the FOLFOX-treated subpopulation for overall survival (A) and progression-free survival (B). Patients with VEGF-D levels in the lowest quantile derived benefit from bevacizumab.