Better Prognosis for ‘Elsewhere' Local Breast Ca Recurrences

DENVER—When breast cancer recurs locally after breast-conserving therapy, prognosis is better for patients whose recurrence is "elsewhere" in the breast than for those whose recurrence is in the primary tumor bed, new research shows. However, regardless of the type, control of the local recurrence is the most significant predictor of subsequent distant metastasis and survival.

Breast-conserving therapy results in a local control rate of more than 90% in early-stage breast cancer, but local recurrences still occur, mainly in the original tumor bed, Evelyn Chen, MD, said at the 47th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (abstract 9). "Multiple studies have shown that local failure distant from the primary tumor bed may have a different temporal pattern and prognosis, compared with true tumor bed recurrences, and may actually represent new primaries," she said. "So in the era of increasing interest in partial breast radiation, it is important to understand the behavior of elsewhere tumor recurrence for comparison."

Dr. Chen and her colleagues at Massachusetts General Hospital retrospectively reviewed the records of 1,280 patients with stage I or II invasive breast carcinoma who were treated with breast-conserving therapy (lumpectomy or quadrantectomy, followed by whole-breast radiation therapy) between 1980 and 2000 at their institution. Their aim was to identify ipsilateral local recurrence detected by mammography, physical exam, or both, and confirmed by biopsy.

The review identified 64 patients with an isolated ipsilateral local recurrence. Their median duration of follow-up was 13 years from the time of diagnosis of the primary and 5 years from the time of diagnosis of the recurrence.

"We carefully reviewed the mammograms, operative notes, pathology reports, and clinical records, and classified the recurrent tumor either as a true tumor bed recurrence arising within the prior lumpectomy scar or cavity or in the same quadrant, or as an elsewhere recurrence arising outside of the prior quadrant and at least 3 cm from the primary," Dr. Chen said.

Overall, 81% of the ipsilateral recurrences were true recurrences. The temporal patterns of true and elsewhere recurrences differed significantly, she noted. The rate of true recurrence was higher than that of elsewhere recurrence at 5 years (0.95% vs 0.3%), at 10 years (2.2% vs 1.4%), and at 15 years (3.6% vs 2.3%). However, as time went on, elsewhere recurrences made up an increasingly larger percentage of all ipsilateral recurrences—23%, 38%, and 40% at 5, 10, and 15 years, respectively.

In terms of characteristics of the primary tumor—method of detection, T stage, size, margin status, estrogen-receptor status, age at diagnosis, and others—the true recurrence and elsewhere recurrence groups did not differ significantly. In terms of characteristics of the recurrence itself, a significantly larger proportion of elsewhere recurrences than of true ones were detected mammographically (asymptomatic)—83% vs 40%—Dr. Chen noted, as opposed to detected clinically (symptomatic).

In addition, true recurrent tumors tended to have more lymphovascular invasion and skin involvement than elsewhere recurrent tumors, Dr. Chen said, but size, grade, and histology (ductal carcinoma in situ vs invasive) did not differ between the groups.

Compared with their counterparts who had true recurrences, patients who had elsewhere recurrences had significantly better estimated 5-year rates of distant metastasis free survival (100% vs 68%) and overall survival (100% vs 66%). Other factors favorably influencing these outcomes included detection of the recurrence by mammography, a longer interval (3 years or more) between initial treatment and recurrence, and—most significantly—a controlled (as opposed to uncontrolled) local recurrence.

In a stepwise multivariate analysis, a shorter interval (less than 3 years) to recurrence and a clinically detected recurrence both independently predicted increased distant metastasis and poorer survival. However, with addition of control of the local recurrence, this factor became the most significant independent predictor of these outcomes.

"The results of our study corroborate other studies evaluating elsewhere recurrence in that there is a delayed rate of elsewhere recurrence, compared with true tumor bed recurrence, and, in some of the studies, there is a survival benefit to elsewhere recurrence," Dr. Chen said. "It remains to be seen whether elsewhere recurrences in the context of partial breast irradiation have similar natural history and favorable outcome."