Supplementary Figure S1; Supplementary Table S1.

3040 Background: Baseline (BL) pVEGFA level has been investigated as a predictive biomarker for the clinical efficacy of anti-VEGF therapy with Bv in 13 trials (7 indications).Three studies, AVADO (breast cancer [BC]), AVAGAST (gastric cancer), and AVITA (pancreatic cancer [PC]), reported that a high BL pVEGFA level was associated with prolonged PFS or OS, but contrasting data were obtained in other Bv clinical trials. Therefore, pVEGFA as a potential predictive factor of Bv activity in specific oncological indications and individual pts remains poorly understood. Methods: To investigate the potential value of BL pVEGFA level (from EDTA-plasma) as a predictive biomarker for Bv activity, sensitivity analyses were carried out on pVEGFA assay data from 13 pivotal Bv trials. In all trials, a VEGFA assay with preferred sensitivity toward the short VEGFA isoforms was used (IMPACT platform, currently used for pt stratification in the MERiDiAN BC trial). Results: In the AVADO study, a high BL pVEGFA level was associated with improved PFS in pts treated with Bv 7.5mg/kg but not Bv 15mg/kg. Additionally, while BL pVEGFA levels were predictive of OS and PFS in the AVITA trial, these results were not replicated in a second PC trial (CALGB 80303). Parallel sensitivity analyses of the IMPACT VEGFA assay and assays that recognize all VEGFA isoforms similarly, suggested that preferential detection of the short isoform of VEGFA was not required to assess outcome correlations (AVADO trial). Sensitivity analysis of pVEGFA data from the AVADO trial also indicated an 11% difference in Spearman's rank correlation between different VEGFA assays, which resulted in a loss of the interaction p-value. Finally, longitudinal analysis of VEGFA level in plasma samples from healthy donors showed substantial intra-pt variability in pVEGFA concentration over time. Conclusions: These data suggest that pVEGFA level is not a robust predictive biomarker for Bv activity; stratification of pts based on a single BL measurement of pVEGFA is unlikely to be successfully implemented in clinical practice. Further assessment of this hypothesis awaits the prospective analysis of pVEGFA data from MERiDiAN.

Abstract Background: Fecal occult blood testing (FOBT) is the recommended first line screening for the detection of colorectal cancer (CRC). To improve the detection of CRC we evaluated serum markers and combinations of serum markers as an alternative approach. Methods: Applying Lasso Regression, a specialized form of penalized logistic regression, we selected six markers for an evaluation in a collective of 857 patients including 301 CRC patients, 143 patients with adenoma, 266 healthy controls and 147 disease controls. For each marker and marker combination the performance was assessed. Results: We tested a total of 22 biomarkers for the detection of CRC from serum. Of these six markers were selected for a marker combination by Lasso Regression. Included were the well-known tumor markers CEA and CYFRA21-1 as well as novel markers or markers that are less routinely used for the detection of CRC: ferritin, osteopontin, anti-p53 and seprase. CEA showed the best sensitivity of all markers with 43.9 % at 95% specificity, followed by seprase (42.4%), CYFRA21-1 (35.5%), osteopontin (30.2%), ferritin (23.9%) and anti-p53 (20.0%). When these markers were combined a sensitivity of 72.3% was reached at a corresponding specificity of 95% and of 62.1% at 98% specificity. Focusing on more screening relevant stages, UICC stages 0-III, reduced the sensitivity slightly to 68.0% and 53.3%, respectively. In a sub-collective where matched stool samples were available (75 CRC cases and 234 controls) the sensitivity of the marker combination was comparable to fecal immunochemical testing (FIT) with 82.4% and 68.9% vs. 81.8% and 72.7% at 95% and 98% specificity, respectively. Conclusion: When six markers were combined to detect CRC from serum, the combination reached a performance that was comparable to FIT. This provides a novel tool for CRC screening to trigger a follow-up colonoscopy for a final diagnosis. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2738.

Abstract Purpose: Fecal occult blood testing is recommended as first-line screening to detect colorectal cancer (CRC). We evaluated markers and marker combinations in serum as an alternative to improve the detection of CRC. Experimental Design: Using penalized logistic regression, 6 markers were selected for evaluation in 1,027 samples (301 CRC patients, 143 patients with adenoma, 266 controls, 141 disease controls, and 176 patients with other cancer). The diagnostic performance of each marker and of marker combinations was assessed. Results: To detect CRC from serum samples, we tested 22 biomarkers. Six markers were selected for a marker combination, including the known tumor markers CEA (carcinoembryonic antigen) and CYFRA 21-1 as well as novel markers or markers that are less routinely used for the detection of CRC: ferritin, osteopontin (OPN), anti-p53, and seprase. CEA showed the best sensitivity at 95% specificity with 43.9%, followed by seprase (42.4%), CYFRA 21-1 (35.5%), OPN (30.2%), ferritin (23.9%), and anti-p53 (20.0%). A combination of these markers gave 69.6% sensitivity at 95% specificity and 58.7% at 98% specificity. Focusing on International Union against Cancer (UICC) stages 0–III reduced the sensitivity slightly to 68.0% and 53.3%, respectively. In a subcollective, with matched stool samples (75 CRC cases and 234 controls), the sensitivity of the marker combination was comparable with fecal immunochemical testing (FIT) with 82.4% and 68.9% versus 81.8% and 72.7% at 95% and 98% specificity, respectively. Conclusions: The performance of the serum marker combination is comparable with FIT. This provides a novel tool for CRC screening to trigger a follow-up colonoscopy for a final diagnosis. Clin Cancer Res; 16(24); 6111–21. ©2010 AACR.