Solidifying liquid biopsies: can circulating tumor cell monitoring guide treatment selection in breast cancer?

Abstract

Almost 150 years ago, T.R. Ashworth first described the presence of epithelial cells in the blood of a woman with metastatic breast cancer that were similar in appearance to her primary tumor cells. Indeed, many patients with a variety of solid tumors, including breast cancer, have detectable cancer cells circulating in the bloodstream, so-called circulating tumor cells (CTCs). CTCs represent a rare cell population in the blood, typically less than 10 cells/mL compared with 1 million WBCs/mL. However, the detection of CTCs within a routine blood specimen provides an opportunity to monitor cancer noninvasively, in essence a liquid biopsy. While dramatic technological advances have recently transformed the analytic capabilities of both CTCs and free plasma DNA analyses, clinical trials based on the first-generation technologies are now providing important insight into how such blood-based monitoring might be integrated into clinical management. In 2004, a seminal study by Christofanilli et al evaluated the prognostic significance of CTCs in metastatic breast cancer, utilizing the Veridex CellSearch assay. The CellSearch assay involves fixation of cells within a blood specimen, followed by immunomagnetic capture of CTCs on the basis of cell surface expression of the epithelial cell adhesion molecule (EpCAM) and staining and scoring of captured cells for expression of epithelial keratins. Women with metastatic breast cancer and elevated CTCs ( 5 cells per 7.5 mL of blood) were found to have a worse progression-free survival (PFS) and overall survival (OS) than women with CTCs below this detection threshold. The prognostic significance of elevated CTCs was subsequently further confirmed in breast as well as prostate and colon cancer. However, expert guidelines from national organizations do not currently support the routine use of CTCs for prognostic assessment, given the unproven clinical benefit of such measurements. In the article that accompanies this editorial, Smerage et al present results from the SWOG S0500 clinical trial, testing whether adjustment of chemotherapeutic regimens based on CTC measures of treatment response can lead to improved clinical outcome in women with metastatic breast cancer. CTCs were measured by CellSearch assay before the start of first-line chemotherapy in women with metastatic breast cancer (N 595).Ofthese,319(54%)womenhadelevatedCTCs( 5cellsper7.5mL of blood) at baseline. The 123 (20.7%) women who had detectable CTCs at baseline and continued to have elevated CTCs despite one cycle of chemotherapy were randomly assigned to either continuation of the same treatment or to an alternative second-line chemotherapy. Both the first-line and second-line therapies were at the discretion of the treating physician, and neither were informed in any way by CTC results. Thus, the study tested the hypothesis that switching among chemotherapy regimens based on early CTC measures of nonresponse might improve outcome, compared with waiting for more traditional radiological evidence of disease progression. The two arms of the randomized study were well balanced with respect to baseline demographic and tumor characteristics and powered to detect a 70% increase in OS (8 to 13.6 months). The study confirmed the prognostic significance of CTC measurements: women who had elevated CTCs at baseline and did not have a decline in CTC numbers following first-line chemotherapy had an OS of only 13 months, compared with 23 months for women whose baseline CTCs declined following first-line chemotherapy and 35 months for women without detectable CTCs at baseline. However, for women in the worst prognostic group (ie, those who continued to have elevated CTCs following a cycle of first-line chemotherapy), switching to an alternative chemotherapyregimendidnotchangeeitherOS(10.7v12.5months;P .98) or PFS (3.5 v 4.6 months; P .64). The trial included patients with diverse breast cancer subtypes, including those who had received endocrineorbiologic therapiesbefore thefirst-linechemotherapy, thusconstituting a heterogeneous population with metastatic breast cancer. In addition, CTC change was only assessed as a binary categorical variable (patients were either positive or negative for CTCs, based on the detectability threshold of 5 CTCs per 7.5 mL of blood). Thus, major changes in CTC numbers in either direction were not scored unless they crossed the threshold of detectability. Whether any of these issues could have modified the magnitude or significance of the results is unclear, but they are important considerations as we move into the next generation of CTCdriven clinical trials. The SWOG S0500 CTC trial is an important study for several reasons. First, it was a well-designed biomarker trial addressing an important clinical question related to clinical utility of CTCs in metastatic breast cancer. Extending beyond past observational studies, this trial involved random assignment based on CTC measurements and sets up a good model of conducting a biomarker trial using standardized technology across multiple institutions. Second, the trial results strongly validate the JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 32 NUMBER 31 NOVEMBER 1 2014