Large-Core Needle Biopsy Reduces Need for Surgical Biopsies

October 1, 2001

SEATTLE-A fully implemented large-core needle biopsy program can dramatically improve the positive predictive value of surgical biopsy of breast lesions, based on the experience of the Brigham and Women’s Hospital, Boston. Jessica Leung, MD, instructor in radiology, Harvard Medical School, presented the study at the 101st Annual Meeting of the American Roentgen Ray Society.

SEATTLE—A fully implemented large-core needle biopsy program can dramatically improve the positive predictive value of surgical biopsy of breast lesions, based on the experience of the Brigham and Women’s Hospital, Boston. Jessica Leung, MD, instructor in radiology, Harvard Medical School, presented the study at the 101st Annual Meeting of the American Roentgen Ray Society.

"By obviating the need for surgery of benign lesions, core biopsy should increase the cancer yield at surgical biopsy after wire localization," Dr. Leung said.

Other studies have addressed this topic, but the study presented by Dr. Leung had some key differences. "Our design is unique in that our study period is 7 years after the implementation of core biopsy—in other words, sufficient time has elapsed for our program to be fully implemented," she said. Furthermore, the study had large enough numbers to achieve high statistical significance.

The researchers also investigated the manner and degree in which core biopsy affects surgical biopsy yield. They compared the cancer yield of surgical biopsy during a 2-year study period after full implementation of the large-core needle biopsy program (July 1, 1998, through June 30, 2000) and a 2-year control period before core biopsy was widely used (January 1, 1987, through December 31, 1988). Surgical biopsy was performed for 1,163 lesions in the study group and 1,261 lesions in the control group.

The overall positive predictive value of surgical biopsy was 0.19 before core biopsy was implemented and 0.42 after core biopsy was implemented. When the lesions were stratified into masses and calcifications, the positive predictive value for masses increased from 0.21 to 0.47, and for calcifications, from 0.17 to 0.37. The results were statistically significant (P < .0001) for all three comparisons.

The histologic subtypes of the cancers diagnosed by surgical biopsy differed little for the two periods. In both the study and control groups, invasive ductal carcinoma was the most common diagnosis (53.8% and 59.5%, respectively) followed by ductal carcinoma in situ (36.2% and 35.8%, respectively).

"The total number of surgical biopsies remained constant despite the implementation of core biopsy. This reflects an increase in our practice over the years in the overall volume of imaging-guided breast biopsy," Dr. Leung said. The ratio of in situ vs invasive carcinomas remained constant, she added, "supporting the contention that our historical cohort and our study population are comparable."

She noted that initial surgical biopsy may still be preferable in some settings. "A core biopsy is not always technically possible, as when the lesion is not well seen by core biopsy equipment or when the lesion is too close to the skin or chest wall. At other times, either the patient or the referring physician may simply prefer a surgical biopsy," she said.