Conference Coverage

DCIS tool IDs axillary node biopsy candidates


Key clinical point: High risk patients who choose axillary biopsy could avoid a second procedure.

Major finding: The 4-item nomogram predicted upstaging with a C statistic of 0.71.

Data source: A retrospective sample (n = 827) and validation study (n = 579).

Disclosures: The source of funding was not disclosed. Dr. Jakub reported having no financial disclosures.


AT ASBS 2017

– A new tool identifies patients with ductal carcinoma in situ (DCIS) who are at high risk for being upstaged as a result of the pathology report. The screen could encourage patients to undergo axillary nodal staging during the core needle biopsy (CNB), thus avoiding a second procedure.

“The risk factors have been well described, but they haven’t helped give individual risk or individual percentages. Our goal was to try to individualize that risk so that we could counsel patients,” said lead researcher James Jakub, MD, chair of the division of breast endocrine and metabolic surgery at Mayo Clinic Rochester, at the annual meeting of the American Society of Breast Surgeons.

Dr. Brittany Murphy and Dr. James Jakub

Dr. Brittany Murphy and Dr. James Jakub

The researchers reviewed data on 827 patients with pure DCIS who had a total of 834 operations at their institutions between 2004 and 2014. Of those, 90% had been identified by screening. The researchers used tumor and patient characteristics with a multivariable model to create a nomogram, which they then validated on a patient population of 579 patients from two other large academic centers.

The researchers found that grade on CNB, mass lesion on imaging, multifocal/centric disease, and linear dimension combined to predict the likelihood of being upstaged to invasive disease (C statistic, 0.71; 95% confidence interval, 0.66-0.77).

They then combined those characteristics to create a nomogram and tested it against the validation set. In that group, 11% of patients were upstaged to invasive disease. The nomogram performed almost identically in the external validation set (C statistic, 0.71; 95% CI, 0.63-0.79). The model predicted 56 upstages, and 46 occurred.

This wasn’t the first attempt to create a nomogram to predict upstaging in DCIS patients, but others were not tested against external datasets. “That’s a weakness with other studies. Unless it’s validated externally, you don’t really know if you can apply it to your population. That was a very large plus: having colleagues who were willing to collaborate to validate it,” said Dr. Jakub.

The team is working on posting the nomogram online for widespread use.

It remains to be seen whether the availability of the nomogram will, in fact, change patients’ decision-making and result in more axillary nodal biopsies during CNB procedures in high risk patients.

“We haven’t looked into that. It will be interesting to see how it influences [sentinel lymph node] biopsy rate at our institution, and we’d love to hear from institutions who apply it. My sense is that patients who are at high risk are going to be interested in doing that. I think it will definitely change their management,” said Dr. Jakub.

Next Article:

   Comments ()