Blood Test Predicts Breast Cancer Relapse 8 Months Earlier

Medscape Medical News > Oncology
Blood Test Predicts Breast Cancer Relapse 8 Months Earlier

Alexander M. Castellino, PhD
September 01, 2015



A blood test was shown to predict breast cancer recurrence 8 months earlier than conventional methods, such as scans, say UK researchers reporting a study in which they measured circulating tumor DNA (ctDNA).

The study, published online August 26 in Science Translation Medicine, was conducted in 55 women with early-stage breast cancer who had all been treated with neoadjuvant chemotherapy and surgery.

Some of these women (19%) were found to have ctDNA in blood samples as early as 2 to 4 weeks post surgery, and the majority of these women (89%) went on to experience a relapse.

"The findings open a window of opportunity to treat patients earlier in order to prevent cancer metastasis and suggest that a noninvasive blood test may one day help identify cancer patients at high risk of recurrence, guiding targeted therapy tailored for those individuals," according to a press release from the Institute of Cancer Research UK, which conducted the study in conjunction with the Royal Marsden National Health Service Foundation Trust.

This study paves the way for an "exciting future, in which serial noninvasive molecular monitoring of early-stage breast cancer might one day be aligned with precise tailoring of adjuvant therapy," comment Tilak Sundaresan, MD, and Daniel A. Faber, MD, from the Massachusetts General Hospital (MGH) Cancer Center, Boston, in an accompanying focus article.

"The investigators have shown that ctDNA can predict whether a woman is likely to experience relapse significantly in advance of conventional methods," Dr Sundaresan told Medscape Medical News.

"However, the real work now remains — to see whether an earlier prediction of relapse can improve survival for these women," he added.

Study Details

The UK researchers first determined somatic mutations present in tumor biopsy specimens from these patients and next determined the earliest time at which ctDNA was detected in a blood sample. In women who on the basis of scanning were found to have experienced disease relapse, the scientists linked the detection of ctDNA from the blood sample to disease recurrence.

Tissue DNA from tumor biopsy specimens was sequenced using massively parallel sequencing to identify somatic mutations present in the tumor before treatment. Using a panel of 14 frequently mutated genes, scientists showed that 43 patients (78%) harbored at least one mutation, with tumors from 12 patients showing two or more mutations.

The scientists then determined whether these mutations could be tracked in the DNA isolated from blood samples of these patients.

Designing a mutation-specific digital polymerase chain reaction (dPCR) assay individualized for each patient, the scientists further showed that in 29 patients, ctDNA could be identified in blood samples before they underwent any treatment for their breast cancer. However, this baseline analysis of ctDNA was not able to predict risk for relapse or disease recurrence.

As the women proceeded to surgery for their breast cancer following neoadjuvant chemotherapy, scientists showed that in 7 of 37 women (19%), ctDNA was detected as early as 2 to 4 weeks after their surgery. Of these women, 86% eventually experienced relapse, compared with 20% of women who did not have evidence of any ctDNA.

The sensitivity of the test increased as the scientists did a "mutation tracking" with dPCR repeated every 6 months. In 50% (6 of 12) of the patients who relapsed, ctDNA was detected in a single postsurgery sample, and in 80% (12 of 15), ctDNA was detected through mutation tracking.

Indeed, women who were found to have ctDNA through mutation tracking were at 12 times higher risk of experiencing relapse compared with women in whom no ctDNA was detected.

Of patients who did not relapse, 96% were not found to have any circulating ctDNA through mutation tracking.
However, the test was not able to predict relapse at all sites; the researchers note that brain metastasis was poorly correlated with ctDNA analysis. Dr Sundaresan and Dr Haber indicate that this may have significance down the road "as more effective therapies eliminate systemic disease and the proportion of women with intracranial relapse grows."

In addition, the scientists showed that in patients who had relapsed, high-depth sequencing of plasma DNA targeting 273 genes mutated in recurring breast cancer revealed a genetic profile that diverged from the primary tumor.

"[The] impressive predictive values derived from this prospectively designed pilot study support the potential utility of ctDNA analyses in personalizing the treatment of localized breast cancer," Dr Sundaresan and Dr Haber write.

Limitations and Clinical Implications

Can this blood test be used for all women with early-stage breast cancer? Although the median follow-up period was 2 years following surgery, Dr Sundaresan told Medscape Medical News: "Some women treated for early-stage breast cancer experience late relapses — even more than 5 to 10 years beyond surgery and adjuvant treatment."

"This study does not indicate that the blood test can be used for an early prediction of late relapses. A much longer follow-up would be required to answer that question," he said.

He also emphasized what was highlighted in the focus article — that the women in the study were at relatively high risk and received neoadjuvant therapy before surgery for their breast cancer.

He indicated that it was not possible to determine whether this approach could be equally effective for patients with low-risk tumors.

Further studies are required to test whether ctDNA is also detected in patients who proceed to surgery without neoadjuvant therapy for clinically defined low-risk primary tumors who nonetheless have an increased chance of relapse and might therefore benefit from adjuvant chemotherapy, Dr Sundaresan and Dr Haber pointed out.

How relevant are these observations to clinical practice and can we see this blood test used in the clinic very soon, Medscape Medical News wanted to know.

With respect to the clinical relevance of these observations, Dr Sundaresan and Dr Haber explained that the detection of ctDNA at a single time point following surgery indicated that women treated with neoadjuvant chemotherapy and surgery still had minimal residual disease. Observing ctDNA on mutation tracking "in effect constitutes screening for early signs of recurrence (that is, surveillance)," they observed.

"The primary clinical implication behind this distinction is that detection of minimal residual disease has the ability to inform the selection of adjuvant therapy, with the ultimate goal of avoiding overtreatment in women who have already been cured of their disease by surgical resection while targeting such therapy to those whose residual disease indicates a high likelihood of relapse," Dr Sundaresan and Dr Haber write.

"The study hints at the possibility that serial blood testing may one day be used to tailor therapy to patients who are likely to relapse," Dr Sundaresan told Medscape Medical News.

However, this test is not ready for clinical use yet. The study lays the groundwork for future clinical trials, which can answer how therapy can be tailored for each patient on the basis of genetic lesions indicative of recurrence, with the end point being improved survival, Dr Sundaresan indicated.

"Given its superior predictive value, the application of serial ctDNA analyses to surveillance for early recurrence of breast cancer is perhaps most exciting in that it opens the door for innovative clinical studies to test 'salvage adjuvant therapies' that might offer a chance of cure before clinically evident relapse," Dr Sundaresan and Dr Haber state.

The authors and Dr Sundaresan have disclosed no relevant financial relationships.

Sci Transl Med. Published online August 26, 2015. Abstract, Perspective