Early Change in Chemotherapy Based on Elevated Circulating Tumor Cells Does Not Improve Outcome in Metastatic Breast Cancer

Get Permission

Circulating Tumor Cells as a Marker in Breast Cancer Treatment

Daniel F. Hayes, MD

Jeffrey B. Smerage, MD, PhD

For this population, there is a need for more effective treatment than standard chemotherapy.

—Jeffrey B. Smerage, MD, PhD, and colleagues
At a recent European meeting, I heard someone say regarding S0500, ‘The technology succeeded, it was the treatment that failed.’ I think that sums up our study nicely.

—Daniel F. Hayes, MD

Elevated circulating tumor cells are associated with poor prognosis in metastatic breast cancer. In a phase III trial (Southwest Oncology Group [SWOG] S0500) reported in Journal of Clinical Oncology, Jeffrey B. Smerage, MD, PhD, Clinical Associate Professor in Medical Oncology at the University of Michigan Comprehensive Cancer Center, Ann Arbor, and colleagues assessed whether changing chemotherapy after one cycle of first-line treatment in patients with persistent circulating tumor cell elevation could improve overall survival.1 They found that change in cytotoxic chemotherapy did not prolong overall survival and their findings confirmed the strong prognostic effect of circulating tumor cell level in this setting.

Daniel F. Hayes, MD, Professor of Medical Oncology at the University of Michigan Comprehensive Cancer Center, is the corresponding author for the Journal of Clinical Oncology article.

Study Details

The study enrolled women with histologically confirmed breast cancer and clinical or radiographic evidence of metastatic disease. Patients could have measurable or nonmeasurable but evaluable bone-only disease with an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2. Patients were eligible if they had stable brain metastases for ≥ 90 days after completing brain radiotherapy; those with leptomeningeal disease were ineligible.

Patients could not have received any prior chemotherapy for metastatic disease; prior and concurrent use of hormone, bisphosphonate, trastuzumab (Herceptin), or bevacizumab (Avastin) therapy was permitted. Adjuvant chemotherapy must have been completed ≥ 12 months prior to the study. Choice of chemotherapy was at the discretion of treating physicians; however, only patients who were to be treated with single-agent chemotherapy were eligible.

Patients were assigned to treatment in four groups. Patients without elevated circulating tumor cells (< 5 circulating tumor cells/7.5 mL whole blood) at baseline remained on initial therapy until progression (group A). Patients with initially elevated circulating tumor cells (≥ 5 circulating tumor cells/7.5 mL) that declined after 21 days of therapy remained on initial therapy (group B). Patients with persistently elevated circulating tumor cells after 21 days of therapy were randomly assigned to continue initial therapy (group C1) or change to an alternative single-agent chemotherapy selected by physicians (group C2). The primary endpoint was overall survival.

Patient Assignment and Characteristics

Of 595 evaluable patients, 276 (46%) did not have elevated circulating tumor cells at baseline (group A). Of those with initially elevated circulating tumor cells, 31 (10%) were not retested due to death, progression, or dropout, 165 were assigned to group B, and 123 were randomized to group C1 (n = 64) or C2 (n = 59).

There were no significant differences among the three groups (A, B, C) or between the two randomized groups (C1 vs C2) in age (59%–62%, 59% vs 66% ≥ 55 years), race (14%–18%, 23% vs 12% black), or measurable disease (79%–83%, 73% vs 85%). Some differences were evident in hormone receptor–positive and HER2-negative disease (54%–72%, 78% vs 66%), triple-negative disease (20%–27%, 17% vs 27%), and HER2-positive disease (6%–22%, 5% vs 7%) status. The likelihood of elevated circulating tumor cells at baseline among these three biologic subtypes was 55%, 45%, and 45% (P = .06). Of patients with elevated circulating tumor cells, 51%, 55%, and 84% converted to low circulating tumor cells after 22 days of first-line treatment
(P = .001).

Outcome in Randomized Comparison

For group C2 (persistently elevated circulating tumor cells, changed treatment) vs C1 (persistently elevated circulating tumor cells, maintained treatment), there were no differences in either median overall survival (12.5 vs 10.7 months, hazard ratio [HR] = 1.00, P = .98) or median progression-free survival (4.6 vs 3.5 months, HR = 0.92, P = .64).

No differences in overall survival were observed among patients with hormone receptor–positive/HER2-negative (17.4 vs 12.4, HR = 0.80, P = .35), triple-negative (7.9 vs 9.1 months, HR = 1.19, P = .67), or HER2-positive disease (7.3 vs 13.3 months, HR = 4.18, P = .16). No differences in progression-free survival were observed among patients with hormone receptor–positive/HER2-negative (5.7 vs 4.2, HR = 0.86, P= .51), triple-negative (2.2 vs 1.9 months, HR = 0.70, P = .38), or HER2-positive disease (7.3 vs 4.7 months, HR = 1.44, P = .69). There was no significant interaction among subtypes for overall survival (P = .21) or progression-free survival (P = .88).

Outcomes Across Groups

Median overall survival was 34.8 months in group A (no circulating tumor cell elevation at baseline), 22.9 months in group B (elevated circulating tumor cells at baseline but not after 21 days), and 13.1 months in the total group C (P < .0001 for trend). The hazard ratio for group C vs group B was 2.13 (95% confidence interval [CI ] = 1.63–2.79).

Post hoc pairwise comparisons showed significant differences between groups after adjustment for multiple comparisons (P < .001). Differences among groups remained significant after adjustment for hormone receptor and HER2 status (P < .0001). Five-year overall survival was approximately 25% in group A and < 10% in group C. Median progression-free survival was 11.1, 8.9, and 4.9 months (P < .0001 for trend), and the HR for group C vs group B was 1.94 (95% CI = 1.56–2.47).

According to biologic subtype, median overall survival was 37.3, 28.1, and 15.0 months (P < .0001; HR =1.96 for C vs B, 95% CI = 1.40–2.75) among hormone receptor–positive/HER2-negative patients, 22.1, 12.5, and 9.5 months (P < .0001; HR = 1.73 for C vs B, 95% CI = 1.02–2.95) among triple-negative patients, and not estimable, 33.0, and 14.1 months (P < .0001; HR = 4.62 for C vs B, 95% CI = 1.68–12.76) among HER2-positive patients. There was no significant interaction by subtype (P = .30).

Within each group, patients with HER2-negative disease who were hormone receptor–positive had better overall survival than those who were hormone receptor–negative. Patients in group A with triple-negative disease had the worst overall survival in group A, which was better than overall survival in the hormone receptor–positive/HER2-negative patients in group C.

Patients with HER2-positive vs HER2-negative disease had better overall survival in both groups A and B. Group C had too few HER2 positive patients for analysis. Overall survival was particularly poor in group C patients with triple-negative disease, with 75% of these patients dying within 15 months. Nearly 25% of women with triple-negative disease in group A survived 3 years, whereas median survival was only 12.5 months for those in group B, a finding that emphasizes the prognostic significance of circulating tumor cell level.

Median progression-free survival was 12.4, 10.6, and 5.5 months (P < .0001) among hormone receptor–positive/HER2-negative patients, 6.5, 6.5, and 2.8 months (P < .0001) among triple-negative patients and 17.9, 8.8, and 5.5 months (P = .008) among HER2-positive patients.

The investigators concluded, “This study confirms the prognostic significance of [circulating tumor cells] in [metastatic breast cancer] patients receiving first-line chemotherapy. For patients with persistently elevated [circulating tumor cells] after 21 days of first-line chemotherapy, early switching to an alternate cytotoxic therapy was not effective in prolonging [overall survival]. For this population, there is a need for more effective treatment than standard chemotherapy.”

Dr. Hayes told The ASCO Post, “At a recent European meeting, I heard someone say regarding S0500, ‘The technology succeeded; it was the treatment that failed.’ I think that sums up our study nicely.” ■

Disclosure: The study was supported by National Cancer Institute grants and by Immunicon and Veridex. Circulating tumor cell assays were performed by Janssen Diagnostics LLC at no charge to SWOG. Dr. Smearage has received honoraria from Veridex and research funding from Immunicon. Dr. Hayes has received research funding from Veridex/Janssen Diagnostics and has related patents, royalties, or licenses. For full disclosures of all study authors, visit


1. Smerage JB, Barlow WE, Hortobagyi GN, et al: Circulating tumor cells and response to chemotherapy in metastatic breast cancer: SWOG S0500. J Clin Oncol. June 2, 2014 (early release online).


Related Articles

Circulating Tumor Cells in Metastatic Breast Cancer: Are We Afraid of the Truth?

All truths are easy to understand once they are discovered; the point is to
discover them.”

—Galileo Galilei


There are several “truths” in breast oncology that have been discovered over the years, become widely understood, and changed the way we practice. Prospective randomized studies have...