ABSTRACT

Background: Optimal surgical margin width for patients with phyllodes tumors (PTs) of the breast remains debated. The aim of this study was to assess the influence of margin width on long-term local recurrence risk.

Patients and Methods: This was a single-institution retrospective review of patients with confirmed PT treated from 2008–2015. Margins were defined as positive (ink on tumor), narrow (no tumor at inked margin but < 10mm), or widely free (>/= 10mm). LR rates were estimated by the Kaplan–Meier method.

Results: Among 117 female patients, histology included 55 (47%) benign, 29 (25%) borderline, and 33 (28%) malignant PT. Final margins were positive in 16 (14%), narrow in 32 (27%), widely free in 64 (55%), and unknown in 5 (4%) patients. Compared with margins > 10 mm, patients with positive and narrow margins had a higher LR risk [HR 10.57 (95% CI 2.48–45.02) and HR 5.66 (95% CI 1.19–26.99), respectively]. Among benign PTs, the 10-year LR-free rates were 100%, 94%, and 66% for widely negative, narrow, and positive margins, respectively (p = 0.056). For borderline/malignant PT, the 10-year LR-free rates were 93% and 57% for widely negative and narrow margins, respectively (p =0.02), with no difference in LR between narrow and positive margin groups (p = 1.00).

Conclusions: For benign PTs, a margin of no ink on tumor appears sufficient to optimize local control. In patients with borderline or malignant PTs, achieving a wide surgical margin may remain important as narrower margins were associated with LR rates comparable to those with positive margins.

Keywords: Phyllodes tumor · Breast sarcoma · Surgical margin · Local recurrence · Long-term outcomes