ASO Author Reflections: Malignant/Borderline Phyllodes Tumors Without Uniformly Poor Histologic Features Have an Excellent Prognosis

Past:

Malignant/borderline phyllodes tumors of the breast are rare neoplasms with poorly understood prognosis following surgical excision. Rates of reported local and distant recurrence are widely variable in the literature due to changes in the grading of phyllodes tumors over time as well as subjectivity in grading attributed to overlapping histologic features and tumor heterogeneity.¹ In addition, the current literature has failed to identify reproducible clinical and pathologic factors associated with both local and distant recurrence, limiting our ability to identify a “high-risk” cohort of patients who may be at higher risk for recurrence. We reviewed our single-institution experience with malignant/borderline phyllodes tumors to assess rates of local and distant recurrence, and to identify factors associated with recurrence.

Present:

Of 125 malignant/borderline phyllodes tumors surgically treated at Memorial Sloan Kettering Cancer Center, the 10-year locoregional recurrence rate was 12%, with no difference between malignant vs borderline tumors (p=0.87) or lumpectomy vs mastectomy (p=0.11). Patients with close (<1mm) or positive margins after surgery had a 33% 10-year locoregional recurrence rate compared to a 9% rate in those with negative margins, highlighting the importance of achieving at least a 1mm margin after lumpectomy or mastectomy to minimize the risk of local recurrence. Uniformly poor pathologic features, including stromal cellularity, stromal overgrowth, infiltrative borders, and ≥10 mitoses per 10 hpf were identified in 20% of malignant phyllodes tumors and predicted for a poor prognosis, with all distant recurrences occurring in this group and with a 63% 10-year disease-specific survival. In contrast, the 10-year disease-specific survival for malignant/borderline phyllodes patients without uniformly poor pathologic features was excellent (100%),² underscoring the current limitations of our grading system.

Future:

Factors associated with local recurrence after surgical excision of malignant/borderline phyllodes tumors were distinctly different than those associated with distant recurrence. Local recurrence was not influenced by histologic features of the tumor and was minimized by achieving at least a 1mm margin (10-year locoregional recurrence rate: <10% at 10 years). As a result, it is difficult to justify routine radiotherapy use following surgical excision for malignant/borderline phyllodes tumors, particularly since no studies have demonstrated a survival benefit to adjuvant radiotherapy in these patients.^3–5 Most patients with malignant/borderline phyllodes tumors did not experience a distant failure; only those with uniformly poor pathologic features were at risk for distant failure and death, suggesting that our current grading system is inadequate for prognostication and should be revised. Unfortunately, there is no systemic therapy known to improve outcomes in patients with malignant phyllodes tumors. Therefore, identification of effective novel therapies that can prevent distant disease development among high-risk patients is urgently needed.