Infrared Thermography Fails to Predict Breast Malignancy
04-May-12 Elsevier Global Medical News BY MIRIAM E. TUCKER
Infrared thermography did not accurately predict malignancy and produced an unacceptably high false-positive rate in women with radiologic abnormalities requiring breast biopsy in a 2-year prospective study.
The No-Touch Breast Scan (NTBS) is a noninvasive, non–radiation-based imaging tool that measures and compares thermal abnormalities in breasts using dual infrared cameras and computer analysis. It generates a score that reflects blood flow patterns based on the theory of tumor angiogenesis.
The technology is being explored as an alternative to radiation-based imaging in women at risk for breast cancer and as a way to reduce the number of benign biopsies, Dr. Andrea V. Barrio said during a press briefing at the annual meeting of the American Society of Breast Surgeons.
This study evaluated NTBS screening as a predictor of breast cancer in patients undergoing minimally invasive breast biopsy for suspicious mammogram, ultrasound, or MRI findings.
But the results demonstrated that NTBS “cannot be used as a successful adjunct to mammography, nor can it replace any of the screening modalities that are standard practice. Mammography remains the gold standard for breast cancer screening,” said Dr. Barrio, an attending breast surgeon at Bryn Mawr (Pennsylvania) Hospital.
“I think the utility of NTBS remains unclear. For the purposes of our study, NTBS could not discriminate between benign and malignant lesions in the low-specificity mode, and the high-sensitivity mode resulted in an unacceptable number of false-positive results,” she added in an interview.
A total of 181 women (median age at diagnosis 52.5 years) with 187 abnormal radiologic findings were evaluated from October 2009 to May 2011. Each patient had an NTBS prior to tissue biopsy, and final tissue pathologies were compared with the corresponding NTBS scan results. Each breast was interpreted as positive or negative based on computer analysis of thermal abnormalities. The contralateral breast was scanned in all patients.
Prior to Oct. 15, 2010, patients were initially scanned using a “high-specificity” NTBS mode termed NTBS1. Subsequently a “high-sensitivity mode (NTBS2)” was used to minimize false negative results. Following initial data analysis, all patients were retrospectively reevaluated in the NTBS2 mode.
Of the 181 patients initially evaluated, 3 were excluded due to a nonductal or lobular breast malignancy, leaving a total of 178 patients. Of those, 50 had 52 positive breast biopsies and 128 had 132 negative biopsies.
Of the 52 positive biopsies, only 26 had a positive NTBS, giving a sensitivity of just 50%. The sensitivity of NTBS was even lower in the 20 in situ cancers, compared with the 32 invasive cancers (35% vs. 59%, respectively), Dr. Barrio reported.
Of the 132 negative biopsies, 88 had negative NTBS scans, giving a 67% specificity. “The positive predictive value of NTBS was 37% and the negative predictive value was 77%,” the study results showed.
Of 173 normal contralateral breasts that were scanned, 42 (24%) had a positive NTBS scan.
Among the 178 patients retrospectively evaluated using NTBS2, 22 were excluded because of an uninterpretable scan. Of the remaining 156 patients, 44 had 46 positive breast biopsies and 112 had 116 negative biopsies. Forty of the 46 positive biopsies matched a positive NTBS (sensitivity 87%).
Sensitivity was not appreciably different between in situ and invasive cancers in the NTBS2 mode (88% vs. 86%, respectively), she said.
Of the 116 negative biopsies, 55 had a negative NTBS, giving a specificity of just 48%. “The positive predictive value of NTBS2 was 40% and the negative predictive value was 90%,” the study reported. Of the 151 normal contralateral breasts that were scanned, 72 (47%) had a positive reading.
In the interview, Dr. Barrio said that her group is not seeking a replacement for mammography, as it is a cost-efficient screening tool that has been proven to decrease mortality from breast cancer.
However, “we are looking for studies to supplement mammography, in order to address its limitations, i.e., dense breasts. I think molecular breast imaging in particular shows a lot of promise for the future in women with dense breasts.”
This study was funded by the Humler Oncology fund. Dr. Barrio had no other disclosures.