Blood Test Predicts Breast Cancer Recurrence

April 18, 2014
Anna Azvolinsky
Anna Azvolinsky

A new pilot study demonstrates the ability of a blood-based test to detect metastatic breast cancer recurrence with high sensitivity and specificity.

As researchers continue to show that cell-free circulating tumor DNA (ctDNA) can be consistently detected in the blood of cancer patients, assays that can provide reliable information about a patient’s treatment progress or potential recurrence by a simple blood draw are becoming increasingly appealing.

A new pilot study demonstrates the ability of a blood-based test to detect metastatic breast cancer recurrence, with high sensitivity and specificity. The ctDNA test “shows great potential for development as a clinical laboratory test for monitoring therapy and disease progression and/or recurrence,” concluded the study authors.

The study was published in Cancer Research, a journal of the American Association for Cancer Research.

The test, called cMethDNA, detects the hypermethylation state of 10 genes through a quantitative, methylation-specific polymerase chain reaction (PCR) assay from ctDNA extracted from a patient’s blood sample. DNA methylation is the addition of methyl groups to either cytosine or adenine DNA nucleotides of a gene. Extensive methylation of a gene typically results in the silencing, or non-expression, of a gene. In breast tumors, for example, the tumor suppressor BRCA1 gene has been found to be either mutated or methylated, inactivating the gene or dampening its expression.

Using a prospective cohort of stage IV metastatic patients enrolled in clinical trials at Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, the cMethDNA test identified a sensitivity of 91.7% and a specificity of 96% for patients with recurrent metastatic breast cancer (P < .0001).

“I think we have achieved a very high sensitivity and specificity of detection for metastatic breast cancer [that is] higher than current tests. Our next step will involve testing the assay in a large prospective study through the large cooperative groups,” senior author Saraswati Sukumar, PhD, professor of oncology and pathology at Johns Hopkins University School of Medicine, told Cancer Network.

Currently, clinicians rely on self-reported patient data and imaging technologies to detect cancer recurrence. Clinical responses to treatment are typically evaluated within 8 to 12 weeks of initiation of therapy.

This study shows that “when patients are undergoing treatment for metastasis, methylation levels in their blood drop as early as 2 weeks after commencement of treatment, much earlier information than results of imaging scans, which are done at 2 to 3 monthly intervals,” said Sukumar. “One is not able to do scans more frequently, but a simple blood test can be done even at weekly intervals.”

Sukumar and colleagues selected the 10 genes based on their highly methylated status in breast tumor tissue and in serum extracted from metastatic breast cancer patients and their low methylation levels in the ctDNA from the serum of individuals who don’t have cancer. Prior reports from genome-wide and select gene panel methylation studies were comprehensively used for this initial analysis. Both estrogen receptor–positive and estrogen receptor–negative samples were used to identify the gene markers. The 10-gene panel, whose methylation status was well differentiated between breast cancer and normal samples, was then validated using The Cancer Genome Atlas database.

To confirm the utility of the test, Sukumar and colleagues then tested the assay in test and training sets of samples from stage IV metastatic breast cancer patients enrolled in prospective clinical trials. All patients had blood and tumor samples collected after treatment and disease recurrence. Samples were also collected from a control group of healthy individuals. Twenty-four and 33 cancer patients and 28 and 27 control individuals were included in the training and test cohorts, respectively.

The cMethDNA test showed that methylation levels reflected responses to treatment and that the methylation patterns of the genes detected with ctDNA were similar to the patterns in the tumor itself.

In a further validation, the authors collected samples from 29 metastatic breast cancer patients participating in two prospective clinical trials testing novel chemotherapy agents. Samples were collected prior to the start of therapy and 18 to 49 days after initiation of the chemotherapy regimen. A statistically significant decrease in median serum DNA methylation levels was seen in patients with stable disease or a partial response as measured by Response Evaluation Criteria in Solid Tumors (P = .01). In contrast, a decrease in methylation was not observed in patients with progressive disease (P = .66).

Of 10 patients analyzed, the core patterns of gene methylation were retained in serum and metastatic tumor samples for each given patient.

Seven of the 10 genes are novel markers in breast cancer: AKR1B1, COL6A2, GPX7, HIST1H3C, HOXB4, RASGRF2, and TM6SF1. Three were previously identified cancer markers-ARHGEF7, TMEFF2, and RASSF1-known to be involved in pathways important for many cancer types, including breast cancer. According to Sukumar, the biological properties of these genes in breast cancer have not yet been studied in detail.

Whether this test could also detect recurrence in breast cancer patients with earlier stages of disease is not clear.

Preliminary data from the study, using The Cancer Genome Atlas database, also showed that this methylation assay may be useful in detecting lung, colon, and rectal cancer recurrence, but that this result needs to be further validated.

There may be other possible applications of this test, said Sukumar, including monitoring survivors at regular annual intervals and personalization of the test to monitor patient-specific tumor gene alterations. According to Sukumar, a team of laboratory researchers and clinicians at Johns Hopkins are taking concrete steps to test some of these potential applications.