Genomic Profiling of DCIS Is Ready for Prime Time

There is a subset of low-risk patients with ductal carcinoma in situ for whom surgery alone is optimal and the Oncotype DX DCIS Score is a useful assay for identifying these individuals.

Patrick I. Borgen, MD

There is a subset of low-risk patients with ductal carcinoma in situ (DCIS) for whom surgery alone is optimal and the Oncotype DX DCIS Score is a useful assay for identifying these individuals, according to Patrick I. Borgen, MD.

“It’s really about understanding the disease in front of you and matching the treatment to that disease,” Borgen said at the 33rd Annual Miami Breast Cancer Conference®, at which he is serving as the program chair. “We use the DCIS Score routinely in our practice—it helps us, it helps our radiation oncologists, and I would argue that it is, in fact, ready for prime time.”

Borgen said the evidence that there is a “reservoir of DCIS” that is unlikely to threaten patients is compiled from screening mammography, autopsy series, and clinical data.

The growth in the use of mammography screening in the United States matches the growth in DCIS cases. “If you look at the slope of the percent change over time, DCIS incidence parallels screening mammography,” said Borgen. He interprets this to mean that with the massive increase in DCIS detection due to screening, it is likely that a subset of these cases would not be clinically relevant.

A number of autopsy series have assessed the incidence of DCIS in women who have died of other causes, revealing DCIS rates higher than those observed in screening studies.

Further, data can also be found in clinical studies supporting the existence of a low-risk subgroup for whom surgery alone is sufficient. One such trial by Van Zee et al was a retrospective database study of nearly 3000 women with DCIS undergoing breast-conserving surgery (Ann Surg. 2015;262[4]:623-631).

“Within this study there were a handful of women who either had DCIS transected at a margin or a 1 mm or less margin. In those patients who received no further surgery, no radiation, and no tamoxifen, 60% were free of breast cancer at 15 years’ follow-up,” said Borgen.

Based on the screening, autopsy, and clinical evidence, Borgen said, “It’s easy for me to put those pieces of information together and conclude that there must be a subset of DCIS that was really not going to harm my patients.”

Standard therapy in the United States for treating DCIS is lumpectomy and radiation, based on a number of large, prospective trials that randomized women with DCIS to lumpectomy alone versus lumpectomy plus radiation therapy.

The landmark study in that series was the NSABP B-17 trial which, according to Borgen, demonstrated a 50% reduction in locoregional recurrence in patients who received radiation therapy. “The locoregional recurrence fell from about 20% in this study down to 10%,” Borgen said.

It was initially thought that a subpopulation from this trial could not be identified who would do well with surgery alone; however, Borgen said the data now reveal that 80% of patients in the study who had lumpectomy alone are disease-free at 20 years.

The question now becomes, “How do we identify that population of patients who are going to be well with just surgery 15, 20, 25 years later?” said Borgen.

He believes that the DCIS Score can help do just that. The assay assesses expression of a series of cancer genes, including Ki-67, STK15, Survivin, Cyclin B1, MYBL2, PR, GSTM1, Beta-actin, GAPDH, RPLPO, GUS, and TFRC. Based on expression levels, the test provides a quantitative risk assessment (0-100). Patients are identified has having a low, intermediate, or high risk of local recurrence.

The DCIS Score was clinically validated through the ECOG E5194 analysis, which included patients with DCIS who had surgery without radiation, about a quarter of whom received tamoxifen (J Natl Cancer Inst. 2013;105[10]:701-710).

“The assay was able to divide the patients into low, intermediate, and high risk,” said Borgen. Sixty-three percent of the DCIS was low-risk. At 10 years, there was a 12% chance of anything happening and a 4% risk of invasive breast cancer, Borgen noted.

A confirmatory study of these results was conducted, known as The Ontario Provincial DCIS Cohort Analysis. The trial included 571 DCIS patients who received surgery without radiation; a subset of these patients received endocrine ablative therapy.

The data matched the initial study, with more than “60% of patients classified as low-risk, and at 10 years, about a 10% risk of any event, and a single-digit risk of an invasive cancer,” said Borgen.

“It’s incredible how closely similar these 2 data sets are in the breakdown of low-risk, intermediate, and high recurrence recurrence,” Borgen said.

In his concluding remarks, Borgen noted that using the DCIS Score does not mean eliminating radiation in all patients identified as low risk; it just means that an increased number of patients may, in consultation with their physician, choose to have surgery alone.

He also stressed that the test is not only useful for identifying patients whose DCIS is not likely to be harmful, but rather, is an important tool for identifying those high-risk individuals who may require additional treatment.

Going forward, Borgen said, “Further research is needed to optimally integrate the DCIS Score with clinicopathological risk factors to evaluate the outcomes and optimize treatment of individuals with clinicopathological risk factors and high risk DCIS score.”


View more from the 33rd Annual Miami Breast Cancer Conference