“Vertical” Inhibition of HER2 Yields Horizontal Gains in the Clinic

In this issue of Clinical Cancer Research (CCR), Ring and colleagues (1) present clinical data for the combination of trastuzumab with afatininib, a combination that demonstrates both the challenges and promises of combination vertical blockade of HER2.

Following the initial proof-of-concept studies with trastuzumab, investigators confronted the reality that although HER2-targeted monotherapy significantly improved outcome, it did not eliminate HER2-driven breast cancer. Indeed, an increasing understanding of HER2 biology has led to multiple new avenues of attack beyond trastuzumab. Given the evidence of trastuzumab's manifest efficacy in the adjuvant and metastatic settings (2), these new approaches regularly involved “combinations” with trastuzumab.

These included drugs targeting the tyrosine kinase moiety of the HER2 molecule (e.g., lapatinib and neratinib), drugs preventing HER2 dimerization (e.g., pertuzumab), antibody–drug conjugates (e.g., T-DM1), and drugs targeting downstream pathway molecules (e.g., PI3K/AKT/mTOR; see Fig. 1). Afatanib now joins the list of agents combined with trastuzumab, as described in the phase I trial presented by in this issue of CCR.

Afatinib is an irreversible, oral, 4-anilinoquinazoline small molecule that binds to the kinase domains of EGFR (ErbB1), HER2 (ErbB2), and HER4 (ErbB4), as well as selected EGFR exon 19 deletion mutations, or exon 21 L858R mutations, to inhibit their tyrosine kinase activity. It was FDA approved in 2013 as first-line monotherapy for patients with metastatic non–small cell lung cancer whose tumors have detectable EGFR exon 19 deletions or exon 21 (L858R) substitution mutations. In that setting, common toxicities (≥20%; 40 mg, once daily dose) include diarrhea, rash, stomatitis, paronychia, dry skin, decreased appetite, and pruritus (3).

Ring and colleagues combined afatinib with trastuzumab in a phase I trial using a standard 3+3 dose escalation. Trastuzumab was used in its standard weekly dosing of 4 mg/kg for 1 week, then 2 mg/kg weekly thereafter. Afatinib ultimately was tested at just two dose levels, 20 mg and 30 mg once daily, with dose-limiting toxicity (DLT), namely grade 3 diarrhea, being common at both dose levels (4/13 and 2/2, respectively). In contrast, afatinib monotherapy is administered at a higher dose of 40 mg daily.

This experience mimics that seen previously with lapatinib, where full doses of the oral kinase inhibitor could not be administered with trastuzumab, due to DLT events defined by grade 3 fatigue and diarrhea, as well as nausea and vomiting (4). Whether this necessity for dose reduction to half the monotherapy dose will also decrease combinatorial efficacy is uncertain. At a dose of 20 mg daily, the reported afatinib C_max and trough concentrations at steady state were 36.8 nmol/L and ∼20.5 nmol/L, respectively, close to the published EC₅₀ of afatinib for HER2 kinase (14 nmol/L) in enzymatic assays (5). However, the toxicities experienced by patients (in the form of diarrhea, rash, and fatigue) were certainly significant. As there was no obvious pharmacokinetic interaction between these two agents, the increased toxicity must be occurring at a pharmacodynamic level.

It is certainly premature, based on the results of a phase I trial, to predict the ultimate fate of the trastuzumab plus afattanib combination. How well have the other combinations fared? The results have been mixed.

Trastuzumab plus lapatinib was the first combination anti-HER2 therapy to be extensively tested. The rationale for “vertical” inhibition of the same receptor kinase target, in this case HER2, was based upon (i) demonstration of synergistic antiproliferative effects in vitro (6; (ii) laptinib inhibition of truncated HER2 receptors that retain a highly functional HER2 kinase domain but lack the extracellular domain and results in intrinsic trastuzumab resistance (7); and (iii) accumulation of HER2 at the cell surface by lapatinib enhanced immune-mediated trastuzumab-dependent cytotoxicity (8).

Following early combinatorial trials in the metastatic HER2 setting, investigators rapidly moved this combination into the neoadjuvant and adjuvant settings. The preoperative trials were promising, with improved (sometimes numerically, NSABP B-41; sometimes statistically, neoALTTO) pCR rates for the combination over monotherapy. The results of lapatinib/trastuzumab combination in the adjuvant disease setting, first presented at the 2014 ASCO annual meeting, were therefore a shock to some and a disappointment to all: a small, non–statistically improved disease-free survival (compared with trastuzumab) accompanied by substantial toxicity (9). However, it is important to note that just 555 disease-free events had occurred at 4.5 years, which fell far short of the 850 that were needed to achieve the desired power in the statistical plan. Moreover, a substantial fraction of subjects enrolled in the ALTTO trial were lymph node–negative (40%), or low T stage (40% ≤ 2 cm), many of whom may never contribute events in the final analysis. Accordingly, further subset analysis of higher risk populations within the ALTTO cohort (i.e., stage II subjects) could still prove to be enlightening.

In contrast, the combination of trastuzumab with pertuzumab has offered impressive outcomes in both the neoadjuvant and metastatic settings. In the neoadjuvant setting, pCR rates were significantly improved with the combination (45.8% vs. 29%; P = 0.0141; ref. 10), and in the metastatic setting the combination of trastuzumab and pertuzumab resulted in an unprecedented 56.5-month median overall survival as first-line therapy in the CLEOPATRA trial, the longest ever recorded for a first-line HER2-positive patient population (11). Taken together, the findings from CLEOPATRA and NeoSphere suggest that targeting HER2-positive tumors with two anti-HER2 monoclonal antibodies that have complementary mechanisms of action results in a more comprehensive blockade of HER2 and underscores the clinical importance of preventing the ligand-dependent formation of HER2-containing dimers (particularly, ligand-activated heterodimers) in order to best silence HER2-driven downstream signaling events. A study of pertuzumab plus trastuzumab adjuvant therapy in patients with newly diagnosed HER2-positive early breast cancer has been initiated, and is currently under way [Adjuvant Pertuzumab and Herceptin in Initial Therapy in Breast Cancer (APHINITY); NCT01358877].

We currently lack published data in a phase III setting with other HER2 combinations, although a recent press release has suggested that the addition of T-DM1 to pertuzumab (the Marianne trial) failed to improve outcome over monotherapy with these agents. This is particularly disappointing given the significant activity seen for T-DM1 in the metastatic setting when compared with the combination of capecitabine and lapatinib (the EMILIA trial; ref. 12). Nonetheless, a two-arm, randomized, open-label, multicenter study is ongoing to evaluate the efficacy and safety of T-DM1 in combination with pertuzumab versus trastuzumab in combination with pertuzumab and a taxane as adjuvant therapy in patients with HER2-positive primary invasive breast cancer, following surgery and anthracycline-based chemotherapy (NCT01966471).

Where will the combination of afatinib and trastuzumab ultimately land? The 11% response rate reported in this trial is unimpressive, but phase I trials are particularly poor places to judge a combination's ultimate activity; the patients entered in such trials tend, as occurred in this trial, to be a highly selected and heavily pretreated population.

Predicting benefit for HER2-targeting combinations has been particularly difficult, and so far we have been reduced to the empiricism of large, expensive trials. This is a particular problem in the adjuvant setting, where the early hope that neoadjuvant trials would represent a simple step to predicting outcome has been called into question by the ALTTO trial results. Thus, we still need predictors of clinical benefit for HER2 combinations, and indeed may still need novel combinations of HER2-targeting agents, with agents affecting either other growth factor receptor pathways or the immune system.

G.W. Sledge reports receiving a commercial research grant from Roche/Genetech and is on the Board of Directors for Syndax. M.D. Pegram is a consultant/advisory board member for Genentech. No other potential conflicts of interest were disclosed.

Conception and design: G.W. Sledge

Writing, review, and/or revision of the manuscript: G.W. Sledge, M.D. Pegram