Abstract 2391: All patients with metastatic breast, colorectal and prostate carcinoma have circulating tumor cells

Abstract

Abstract Background: Presence of EpCAM+ cytokeratin+ nucleated circulating tumor cells (CTC) in metastatic carcinoma patients is associated with poor survival and may be used to guide treatment. Assessment of treatment targets on these CTC holds the promise of a liquid biopsy. However the proportion of patients in which a sufficient number of CTC are detected in 7.5 mL of blood is not sufficient. In this study we estimate the number of CTC in blood of patients with metastatic disease, explore the relationship with survival and determine whether assay modifications and/or increases in sample volume are needed to achieve the required increase in number of CTC detected. Methods: EpCAM+CK+DNA+CD45- CTC enumeration was performed with the CellSearch® system in 7.5 mL of blood of 836 patients with metastatic breast, colorectal and prostate cancer patients (Cristofanilli, NEJM 2004, Cohen, JCO 2008, De Bono, CCR 2008). EpCAM+, CD45-, Nucleic acid+ CTC were enumerated in 100ul of NH4CL lysed blood from 140 metatatic cancer patients and compared to CTC as enumerated with the CellSearch system. Image analysis for automated CTC enumeration and modeling of CTC frequencies was performed in Matlab. Results: The median number of CTC in 7.5mL of blood metastatic breast, colorectal and prostate is 5 and in 39% of patients no CTC are detected. The best fit of the CTC frequency distribution was used to extrapolate the sample volume to 5 liters of blood and predicted that 99% (95% confidence interval, CI, 95-99.8%) of patients had at least 1 CTC before initiation of therapy, which decreased to 97% (95% CI 87-99.5%) after the first cycles of therapy. One hundred CTC per liter of blood are present in ∼80% of patients, 1 CTC per ml of blood in ∼40% of patients and 10 CTC per ml of blood in ∼20% of patients. The median survival of patients with CTC is reduced by 6.6 months for each tenfold CTC increase. CTC definitions that do not include a requirement for presence of DNA, EpCAM, cytokeratin 8,18,19 (CK) or absence of CD45 are less predictive of survival, than the strictest definition requiring a CTC to be EpCAM+CK+DNA+CD45-. Comparison of CTC detected by flow cytometry to the CellSearch CTC method on 140 patients shows that improvements in EpCAM recovery may increase the average number of detected CTC by 6.5 fold (95% CI 5.6-7.3), yet the fits predict that this will only reduce the number of patients with 0 CTC from 39% to 21% (95% CI 13-31%). Conclusions: EpCAM+CK+DNA+CD45- CTC are present in blood of all patients with metastatic breast, colorectal and prostate cancer. The predicted strong relation with survival suggests the importance of this phenotype for metastasis. To use CTC as a liquid biopsy for the majority of patients, the CTC yield needs to be improved 100-1000 fold. This requires a dramatic increase in sample volume which may be achieved by in vivo flow cytometry, or through processing an apheresis product. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2391. doi:1538-7445.AM2012-2391