In a phase III trial (NSABP B-41) performed to assess the potential benefit of neoadjuvant dual HER2 blockade in HER2-positive breast cancer, André Robidoux, MD, of Centre Hospitalier de l’Université de Montréal, and colleagues in the National Surgical Adjuvant Breast and Bowel Project (NSABP) evaluated tumor response in patients receiving neoadjuvant lapatinib (Tykerb) or trastuzumab (Herceptin) alone and in combination after doxorubicin/cyclophosphamide treatment.1 The study, reported in The Lancet Oncology, showed that rates of pathologic complete response were similar with lapatinib and trastuzumab, with the combination producing a numerically but not statistically greater response rate.
In this open-label trial, 529 women aged 18 years or older with an Eastern Cooperative Oncology Group performance status of 0 or 1 with operable HER2-positive breast cancer received four cycles of doxorubicin 60 mg/m² and cyclophosphamide 600 mg/m² IV on day 1 every 3 weeks followed by four cycles of weekly paclitaxel (80 mg/m²) IV on days 1, 8, and 15 every 4 weeks. Patients were randomly assigned to receive weekly paclitaxel concurrently with either trastuzumab (4 mg/kg load, then 2 mg/kg IV) weekly until surgery (n = 181), lapatinib (1,250 mg orally) daily until surgery (n = 174), or weekly trastuzumab plus lapatinib (750 mg orally) daily until surgery (n = 174). After surgery, all patients received trastuzumab to complete 52 weeks of HER2-targeted therapy. Patients were stratified by clinical tumor size, clinical nodal status, hormone receptor status, and age. The primary endpoint was pathologic complete response in the breast.
The trastuzumab, lapatinib, and combination groups were generally well balanced for age (eg, ≤ 49 years in 56%, 53%, and 48%), clinical tumor size (eg, 2.1-4.0 cm in 56%, 47%, and 51%), nodal status (positive in 51%, 52%, and 49%), hormone receptor status (positive in 67%, 58%, and 62%), ethnic origin (eg, white in 85%, 84%, and 88%), histologic grade (eg, high in 48%, 56%, and 48%), and type of surgery (eg, lumpectomy in 55%, 46%, and 50%).
A total of 12 patients withdrew consent before response evaluation or did not have surgery; thus, a total of 519 patients, including 177 in the trastuzumab group, 171 in the lapatinib group, and 171 in the combination group had pathologic responses determined.
In total, neoadjuvant therapies were completed per protocol in 77% of trastuzumab patients, 65% of lapatinib patients, and 63% of combination group patients with complete neoadjuvant treatment information available. Targeted therapy was completed per protocol in 82%, 66%, and 66%, respectively. After discontinuing study therapy, 4% of the lapatinib group crossed over to trastuzumab.
Compared with the trastuzumab group, in the combination group the proportions of patients with pathologic complete response in breast (62.0% vs 52.5%, P = .095) and pathologic complete response in breast and nodes (60.2% vs 49.4%, P = .056) were numerically but not significantly greater. There was no difference between the trastuzumab group and the lapatinib group in pathologic complete response in breast (53.2%, P = .99 vs trastuzumab) or in breast and nodes (47.4%, P = .78 vs trastuzumab).
Odds ratios for complete pathologic response for the combination vs trastuzumab were 1.47 (95% confidence interval [CI] = 0.96–2.26) in breast and 1.55 (95% CI = 1.01–2.37) in breast and nodes, and those for lapatinib vs trastuzumab were 1.03 (95% CI = 0.67–1.57) in breast and 0.92 (95% CI = 0.60–1.40) in breast and nodes.
Outcome by Hormone Receptor Status and Subgroups
In hormone receptor–positive patients, pathologic complete response in the breast was observed in 55.6% of combination patients (P = .23 vs trastuzumab), 48.0% of lapatinib patients (P = .96 vs trastuzumab), and 46.7% of trastuzumab patients, with complete response in breast and nodes observed in 54.6% (P = .21), 42.0% (P = .70), and 45.5%, respectively. In hormone receptor–negative patients, complete response in breast was observed in 73.0% (P = .49), 60.6% (P = .71), and 65.5% of patients, respectively, with complete response in breast and nodes in 69.8% (P = .29), 54.9% (P = .85), and 58.2%, respectively. For all groups, compete response rates were greater among hormone receptor–negative patients than hormone receptor–positive patients.
Analyses by stratification subgroups indicated that the improvement in pathologic complete response rates with the addition of lapatinib occurred predominantly in clinically node-positive patients. Odds ratios for response in breast were 2.32 (95% CI = 1.25–4.31) in node-positive patients and 0.95 (95% CI = 0.53–1.74) in node-negative patients (P = .043 across nodal status). Odds ratios for response in breast and nodes were 2.58 (95% CI = 1.39–4.81) in node-positive patients and 0.96 (95% CI = 0.53–1.73) in node-negative patients (P = .023 across nodal status). Improvements were similar across subgroups according to tumor size, hormone receptor status, and age at study entry.
Patients in the two groups receiving lapatinib had more grade 3 or 4 adverse events (62% of lapatinib patients and 60% of combination patients) than did trastuzumab patients (50%). The most common grade 3 or 4 toxic effects were neutropenia, which occurred in 16% of trastuzumab patients (grade 4 in 3%), 16% of lapatinib patients (grade 4 in 5%), and 17% of combination patients (grade 4 in 5%), and grade 3 diarrhea, which occurred in 2% of trastuzumab patients, 20% of lapatinib patients, and 27% of combination patients (P < .0001 for both lapatinib-containing groups vs trastuzumab).
Symptomatic congestive heart failure defined as New York Heart Association class III or IV events occurred in 4% of patients in the trastuzumab group, 4% in the lapatinib group, and < 1% of the combination group (P = .185). Neoadjuvant treatment was discontinued due to adverse events in 15% of the trastuzumab group, 24% of the lapatinib group, and 26% of the combination group, including discontinuation of targeted therapy in 7%, 19%, and 24%.
The investigators concluded: “Substitution of lapatinib for trastuzumab in combination with chemotherapy resulted in similar high percentages of pathological complete response. Combined HER2-targeted therapy produced a numerically but insignificantly higher pathological complete response percentage than single-agent HER2-directed therapy; these findings are consistent with results from other studies. Trials are being undertaken to further assess these findings in the adjuvant setting.” ■
Disclosure: The study was funded by GlaxoSmithKline. For full disclosures of the study authors, visit www.thelancet.com.
1. Robidoux A, Tang G, Rastogi P, et al: Lapatinib as a component of neoadjuvant therapy for HER2-positive operable breast cancer (NSABP protocol B-41): An open-label, randomised phase 3 trial. Lancet Oncol 14:1183-1192, 2013.