Purpose: The objective of this study was twofold: first, to identify patients with locally advanced breast cancer (LABC) who will achieve a pathological response to a preoperative regimen of concurrent paclitaxel and radiation; and second, to explore associations between molecular markers from the original tumors and pathological response.
Methods and materials: Patients with previously untreated LABC were eligible to receive a regimen of preoperative concurrent paclitaxel, 30 mg/m(2) twice a week for a total of 8 weeks, and radiation delivered Weeks 2--6, 45 Gy at 1.8 Gy per fraction to the breast, ipsilateral axilla, and supraclavicular nodes. At mastectomy, pathologic findings were classified as pathological complete response (pCR) = no residual invasive cells in the breast and axillary contents; pathological partial response (pPR) = presence of < or = 10 microscopic foci of invasive cells; no pathological response (pNR) = pathological persistence of tumor. For each patient, pretreatment breast cancer biopsies were prospectively analyzed by immunohistochemistry (IHC) for estrogen and progesterone (ER/PR) hormonal receptors, HER2/neu and p53 overexpression. Estrogen receptor (ER), HER2/neu, metablastin, beta-tubulin III and IV, microtubule-associated protein-4 (MAP-4), bcl-2, bax, and cyclooxygenase-2 (COX-2) gene expression were measured using real-time quantitative polymerase chain reaction (PCR).
Results: A total of 36 patients had pretreatment biopsies and were evaluable for the analysis of the association of molecular markers with pathological response. Pathological response in the mastectomy specimen was achieved in 12 of these 36 patients (33%). Only HER2/neu and ER gene expression were found to be significantly associated with the extent of pathological response to the regimen, i.e., tumors with low HER2/neu gene expression and negative estrogen receptors were more likely to respond to the tested regimen (p = 0.009 and p = 0.006, respectively). Conversely, p53 protein expression measured by IHC did not appear to be associated with pathological response (p = 0.67).
Conclusion: Further studies in LABC should assess whether patient selection for treatment based on the original tumor molecular characteristics could affect their chance to achieve a pathological response.