Parallel upregulation of Bcl2 and estrogen receptor (ER) expression in HER2+ breast cancer patients treated with neoadjuvant lapatinib.

Background: We previously showed in HER2+ models of breast cancer (BC) that potent inhibition of the HER receptor layer can lead to re-expression of estrogen receptor (ER) or activation of the ER pathway. Consequently, the anti-apoptotic ER gene product Bcl2 is upregulated, resulting in enhanced tumor cell survival and treatment resistance. In this study, we investigated whether Bcl2 and ER expression levels are simultaneously increased by neoadjuvant treatment with the dual HER1/2 tyrosine kinase inhibitor lapatinib in HER2+ BC patients. Methods: In a neoadjuvant phase II clinical trial 49 HER2+ BC patients were treated with lapatinib as a single agent for 6 weeks, followed by trastuzumab/docetaxel for 12 weeks before surgery. Tumor specimens were prospectively collected at different time-points during lapatinib treatment (baseline, and weeks 2 and 4). Bcl2, ER, progesterone receptor (PR), total (t) and phosphorylated (p)-HER2, and Ki67 were assessed by immunohistochemistry. Spearman correlation was used to evaluate the association among the biomarkers at baseline, and the correlation of their changes over time. Fisher's Exact test and non-parametric Wilcoxon rank sum test were used respectively to determine if the frequency and the magnitude of Bcl2 expression changes were associated with baseline ER status. Results: 35/49 HER2+ tumor specimens (71%) were available for baseline evaluation of Bcl2 and ER. Of those, 12 (34%) were ER-positive (Allred score 3) and 23 (66%) ER-negative. Baseline Bcl2 expression correlated positively with ER (r = .75; p < .0001) and PR (r = .53; p = .0015), and inversely with t-HER2 (r = -.43; p = .0097). ER baseline expression correlated positively with PR (r = .57; p = .0004), and inversely with t and p-HER2 (r = -.55; p = .0005, and r = -.37; p = .0282, respectively) and Ki67 (r = -.39; p = .0271). Bcl2 changes at week 2 (w2) positively correlated with changes in both ER and PR levels (r = .70; p = .0002 and r = .57; p = .0076, respectively). Additionally, the increase in Bcl2 expression, observed in 9 of the 23 (39%) tumors with tissue available at w2, was significantly more frequent (p = .0147) and of greater magnitude (p = .0001) in ER-pos vs. ER-neg tumors -- 8/9 ER-pos tumors at w2 (including 3 converted from ER-neg by lapatinib) had increased Bcl2, while only 1 of the 14 (7%) ER-neg tumors (at baseline and w2) had increased Bcl2. The expression of ER itself at w2 also increased in 3 out of the 6 (50%) tumors which were originally ER-pos and had tissue available at w2, and in all of them Bcl2 increased in parallel. Of note, the single baseline ER-pos tumor that showed a reduction in ER at w2 had a parallel decrease in Bcl2. Similar observations or trends were found at week 4. Conclusion: Our study suggests that Bcl2 is upregulated as a result of enhanced/restored ER activity upon anti-HER2 therapy with lapatinib in HER2+ tumors. This further supports the use of endocrine along with anti-HER-2 therapy to block this escape pathway which could otherwise cause treatment resistance. In addition, the ER re-expression with lapatinib treatment observed in this study emphasizes the need to re-biopsy HER2+/ER--patients receiving anti-HER2 therapy and to add endocrine therapy if the tumor becomes ER-positive. [ABSTRACT FROM AUTHOR]