Abstract
Introduction: A Phase II RCC study (VEG102616) demonstrated activity with a response rate = 34.7% (CI 95%, 28.4 – 40.9) and disease control rate (CR + PR +SD) = 80% [ASCO 2008, #5046]. An analysis of plasma cytokines and angiogenic factors (CAFs) was undertaken with the objective of identifying patients who were the most likely to respond to pazopanib.
Methods: A subset (n=129/217) of available plasma samples were selected based on extremes of tumor shrinkage from baseline. CAF levels were correlated with dichotomized tumor shrinkage (DTS) by logistic regression, with continuous tumor shrinkage (CTS) by linear regression, and with progression‐free survival (PFS) by Cox regression. ROC analysis was performed using a stratified random sample of 97 patients in a training set and 32 pts in a test set.
Results: Elevated levels of HGF, IL‐8, and IL‐6 were significantly correlated with less tumor shrinkage. Lower levels of E‐Selectin and elevated IL‐6 were associated with shorter PFS. ROC curve analysis for HGF resulted in an AUC of 0.668 with an optimal cutoff of 638.5 pg/ml. Using this cutoff, HGF‐based predictions in the test set had an overall accuracy of 53%, positive predictive value of 67% and negative predictive value of 50%.
Conclusion: CAF profiling showed that lower baseline levels of plasma HGF, IL‐6, and IL‐8 are correlated with greater tumor shrinkage in renal cell carcinoma pts treated with pazopanib. Also, elevated levels of E‐Selectin and lower levels of IL‐6 were associated with longer PFS.
| CAF1 . | p‐value . | Fold Change2 . | |
|---|---|---|---|
| Dichotomized | HGF | 0.006 | Fold change: 1.3 |
| Tumor Shrinkage | IL‐8 | 0.036 | Fold change: 3.0 |
| (Logistic regression) | IL‐6 | 0.039 | Fold change: 1.5 |
| TIMP1 | 0.047 | Fold change: 1.2 | |
| Continuous | HGF | 0.029 | Not Applicable ‐ NA |
| Tumor Shrinkage | IL‐8 | 0.048 | NA |
| (Linear Regression) | IL‐6 | 0.031 | NA |
| PFS (Cox Proportional | E‐Selectin | 0.017 | NA |
| Hazards Model) | IL‐6 | 0.005 | NA |
| CAF1 . | p‐value . | Fold Change2 . | |
|---|---|---|---|
| Dichotomized | HGF | 0.006 | Fold change: 1.3 |
| Tumor Shrinkage | IL‐8 | 0.036 | Fold change: 3.0 |
| (Logistic regression) | IL‐6 | 0.039 | Fold change: 1.5 |
| TIMP1 | 0.047 | Fold change: 1.2 | |
| Continuous | HGF | 0.029 | Not Applicable ‐ NA |
| Tumor Shrinkage | IL‐8 | 0.048 | NA |
| (Linear Regression) | IL‐6 | 0.031 | NA |
| PFS (Cox Proportional | E‐Selectin | 0.017 | NA |
| Hazards Model) | IL‐6 | 0.005 | NA |
The following CAF analysis were performed: Angio‐10 (VEGF, VEGF‐R2, ICAM‐1, TIE‐2, PLGF, TIMP‐1, HGF, FGF basic, E‐Selectin, and IL‐8) and Cyto‐8 (IL‐1 beta, IL‐2, IL‐5, IL‐6, IL‐8, IL‐10, IL‐12, TNF alpha) BIOCHIPS (Biomarker Workstation, Decision Biomarkers, Inc). Only the significant results were displayed above.
Fold change = (median CAF level for group with less tumor shrinkage) / (median CAF level for group with greater tumor shrinkage).
Citation Information: Mol Cancer Ther 2009;8(12 Suppl):A11.
RSS Feeds