Weekly paclitaxel/carboplatin/trastuzumab therapy improves pathologic complete remission in aggressive HER2-positive breast cancers, especially in luminal-B subtype, compared with a once-every-3-weeks schedule

Abstract

Background: The efficacy and tolerability of two different schedules of paclitaxel, carboplatin, and trastuzumab (PCarH) for HER2-positive, locally aggressive (stage IIB-IIIC) breast cancers were evaluated in this phase II trial.

Methods: Patients were randomly assigned to receive either weekly (12 doses over 16 weeks) or once-every-3-weeks (4 doses over 12 weeks) treatment. The primary endpoint was pathologic complete remission (pCR) in the breast and axilla. To detect an assumed 35% pCR absolute difference between the two schedules, a minimum of 26 assessable patients in each group was required (two-sided α = 0.05, β = 0.2).

Results: A total of 56 patients were enrolled (weekly group, n = 29; every-3-weeks group, n = 27). In the intent-to-treat analysis, pCR in the breast/axilla were found in 31 patients (55%; 95% confidence interval [CI]: 41%-69%). Compared with the every-3-weeks schedule, the weekly administration achieved higher pCR (41% vs. 69%; p = .03). After adjustment for clinical and pathological factors, the weekly administration was more effective than the every-3-weeks schedule, with hazard ratio of 0.3 (95% CI: 0.1-0.9; p = .03). Interestingly, weekly administration resulted in high pCR rates in both luminal-B (HER2-positive) and ERBB2+ tumors (67% vs. 71%; p = .78), whereas luminal-B (HER2-positive) tumors benefited less from the every-3-weeks schedule compared with the ERBB2+ tumors (21% vs. 62%, p = .03). These results remain after multivariate adjustment, showing weekly administration was more effective in the luminal-B (HER2-positive) subgroup (p = .02) but not in the ERBB2+ subgroup (p = .50).

Conclusion: A more frequent administration might improve the possibility of eradicating invasive cancer in the breast and axilla, especially in the luminal-B (HER2-positive) subtype. Further studies to validate our findings are warranted.

Figure 1.

Treatment plan. Abbreviations: AUC, area under the concentration-time curve; BC, breast cancer; Car, carboplatin; FISH, fluorescence in situ hybridization; H, trastuzumab; IHC, immunohistochemistry; P, paclitaxel.

Figure 2.

Consolidated Standards of Reporting Trials diagram. Abbreviations: Car, carboplatin; H, trastuzumab; ITT, intention-to-treat; P, paclitaxel.

Figure 3.

Pathological response of the two schedules in overall subjects and subgroups. (A): pCR in the breast and in the breast and axilla according to treatment groups. *p = .03; **p = .03. (B): Forest plot of univariate analysis. The odds ratios (ORs) with 95% confidence intervals (CIs) for the patients assigned to the weekly schedule, as compared with the every-3-weeks schedule, were assessed using unadjusted logistic regression by different clinical and pathologic parameters. The dashed vertical line denotes an OR of 1.0. The diamond/square denotes the OR of each subgroup. An OR <1.0 means achieving higher pCR rate from the weekly schedule; an OR >1.0 means achieving higher pCR rate from the every-3-weeks schedule. (C): Each bar represents one patient, with the green bar denoting the patient achieving pCR and the yellow denoting non-pCR. The p values of the pCR difference between luminal-B (HER2-positive) tumors and ERBB2+ tumors were .78 and .03 for the weekly schedule and every-3-weeks schedule, respectively. The p values of the pCR difference between the weekly schedule and every-3-weeks schedule were .01 and .59 for luminal-B (HER2-positive) tumors and ERBB2+ tumors, respectively. Abbreviations: CI, confidence interval; ER, estrogen receptor; LN, lymph node; OR, odds ratio; PcarH, paclitaxel, carboplatin, and trastuzumab; pCR, pathologic complete remission; Post., postmenopausal; PR, progesterone receptor; Pre., premenopausal.