Pharmacokinetic and exposure–response analyses of pertuzumab in combination with trastuzumab and docetaxel during neoadjuvant treatment of HER2+ early breast cancer

Pertuzumab (PERJETA^®, F. Hoffmann-La Roche, Basel, Switzerland) is a recombinant, humanized, immunoglobulin (Ig)G1κ monoclonal antibody, which targets human epidermal growth factor receptor 2 (HER2). Pertuzumab is the first in a new class of targeted cancer treatments called HER2 dimerization inhibitors. Non-clinical data indicate that pertuzumab and trastuzumab (Herceptin^®, F. Hoffmann-La Roche, Basel, Switzerland) bind to distinct epitopes on the HER2 without competing with each other and have distinct mechanisms for disrupting HER2 signaling [1, 2]. These mechanisms are complementary and result in augmented anti-proliferative activity in vitro and in vivo when pertuzumab and trastuzumab are given in combination [3‐5]. By binding to the subdomain II of the extracellular domain of HER2, pertuzumab prevents heterodimerization of HER2 with other members of the HER family (HER1, HER3, and HER4). As a result, ligand-activated downstream signaling is blocked by pertuzumab. Pertuzumab is also capable of activating antibody-dependent cell-mediated cytotoxicity (ADCC) similar to trastuzumab [6]. Pertuzumab in combination with trastuzumab and docetaxel was shown to significantly improve progression-free survival (PFS) and overall survival (OS) in patients with first-line metastatic HER2-positive breast cancer, which led to its approval [7] in the USA in 2012 and in the European Union in 2013 with intravenous dosing at a fixed (non-weight-based dose) loading dose of 840 mg, followed by 420 mg on a every three-week (q3w) schedule [7‐11]. In the pivotal trial, CLEOPATRA, no DDI between pertuzumab and trastuzumab and between pertuzumab and docetaxel was detected in a limited number of patients evaluated [12].

NeoSphere, a Phase II, multicenter study spread across 16 countries for HER2-positive breast cancer patients, was conducted to assess the activity of pertuzumab (PERJETA^®) by comparing the therapeutic effects of the conventional combination of trastuzumab (Herceptin^®) plus docetaxel with the combination of pertuzumab with either docetaxel or trastuzumab, or both, in a neoadjuvant setting. This clinical trial was a four-arm study evaluating the efficacy and safety of neoadjuvant treatment regimens in female patients with locally advanced, inflammatory or early-stage HER2-positive breast cancer. Before surgery, patients were randomized to receive four cycles of one of the following four treatment arms: (A) trastuzumab + docetaxel, (B) trastuzumab + docetaxel + pertuzumab, (C) trastuzumab + pertuzumab, and (D) pertuzumab + docetaxel. Post-surgery patients in arm A, B, and D received three cycles of 5-fluorouracil, epirubicin, and cyclophosphamide (FEC) and trastuzumab to complete 1-year treatment (17 cycles in total). Patients in arm C received four cycles of docetaxel followed by three cycles of FEC and trastuzumab to complete 1-year treatment (21 cycles in total). The primary end point was pathological complete response (pCR) evaluated after Cycle 4. Pertuzumab increased the pCR response rate in patients when used in combination with trastuzumab and docetaxel (Table 1) [6]. Overall, in the NeoSphere study, a significantly higher proportion of women given neoadjuvant pertuzumab and trastuzumab plus docetaxel achieved pCR in the breast than did those given trastuzumab and docetaxel alone, leading to its approval in the USA and in the European Union in 2013 and 2015, respectively. Although pertuzumab plus docetaxel was efficacious, the combination of chemotherapy with both antibodies was more active than chemotherapy with either antibody alone [9]. A 5-year analysis showed that patients achieving a total pCR with all groups combined had a longer PFS compared with patients that did not achieve total pCR, thus suggesting that pCR could be an early indicator of long-term outcome in early-stage HER2-positive breast cancer [13].

The objectives of this analysis were to: (1) compare pertuzumab PK between the neoadjuvant population (early breast cancer [EBC]) in NeoSphere to a population of patients with tumor types including the first-line metastatic breast cancer (MBC) population, (2) to further explore the potential impact of trastuzumab and docetaxel on pertuzumab PK, and (3) perform an exposure–response (E–R) analysis to explore whether an E–R trend existed at the administered pertuzumab dose to further support selection of the clinical dosing regimen in the target patient population.

In total, there were 180 patients with a BSR blood sample collected; 139 in the pertuzumab-containing arms (Arms B, C, and D) and 41 in the trastuzumab plus docetaxel arm (Arm A). Of the 180 patients, 173 had Cycle 4 pCR assessments available.

The ER population consisted of 173 patients that had both PK and pCR assessments available. Eighty-eight (88) patients from treatment groups A and B with week 4 pCR assessments were used in the ER analysis of pertuzumab. In each treatment group, the pCR rates of patients with PK results were similar to those of the overall treated patients for each arm of the study [6].

The pCR rate versus the model-predicted pertuzumab C _trough serum concentrations at Cycle 2 are illustrated in Fig. 4. The patients included in the plot comprised of two groups: Arm A treated with trastuzumab + docetaxel and Arm B treated with trastuzumab + docetaxel + pertuzumab combined together for the analysis. The model-predicted pertuzumab C _trough serum concentrations ranged from 3.4 to 103.2 μg/mL. Forty-six of 49 (94%) patients treated with pertuzumab (Arm B) had a predicted C _trough pertuzumab serum concentration of >20 μg/mL, the target efficacious exposure based on non-clinical efficacy models.

The pCR rate was higher in patients treated with pertuzumab plus trastuzumab and docetaxel compared with patient treated with trastuzumab and docetaxel (p < 0.05); however, there was no significant impact (p = 0.996) on the probability of pCR response with an increase in pertuzumab serum concentration (C _trough) beyond 20 μg/mL (Fig. 4, panel A). An analysis using observed pertuzumab concentrations collected during Days 14–21 of Cycle 2 yielded very similar results (Online Resource 3).

The NeoSphere clinical trial was a four-arm study conducted to assess the activity of pertuzumab in neoadjuvant (EBC) setting with pCR as the primary efficacy end point (Table 1). In the NeoSphere study, a significantly higher proportion of women given neoadjuvant pertuzumab and trastuzumab plus docetaxel achieved pCR in the breast than did those given trastuzumab and docetaxel alone, with pCR rates of approximately 46% and 30%, respectively [6, 13]. The objectives of this analysis were to compare pertuzumab PK between the EBC population (neoadjuvant treatment) in NeoSphere and a population of patients with other tumor types including the first-line MBC population, to explore the potential impact of trastuzumab and docetaxel on pertuzumab PK, and to assess the relationship between exposure (C _trough) and response (pCR) of pertuzumab (in combination with trastuzumab and docetaxel) in neoadjuvant treatment of EBC. pCR was selected as the outcome variable in our E–R analysis as it was the primary end point in the trial, later analyses showed a correlation between pCR and DFS/EFS, further supporting the utility of pCR as an end point. In the NeoSphere trial, the safety and tolerability of the triple regimen of pertuzumab, trastuzumab, and docetaxel were similar to those of trastuzumab plus docetaxel [15] and no unique safety signals were identified that could be attributed to pertuzumab exposure. Therefore, an ER analysis with respect to safety was not conducted.

As expected due to differences in demographics (100% females versus 62% females) and health status (EBC versus advanced solid tumors), the patients enrolled in the NeoSphere trial displayed LBW and albumin levels which were lower and higher, respectively, compared to the population PK model population, resulting in CL values that were slightly lower compared to the reference model. The PK model predictions of pertuzumab serum concentrations matched the observed serum concentrations after correcting for these baselines covariate differences. Pertuzumab PK in the EBC population in NeoSphere appear to be in agreement with the PK in patients with other tumor types, including the first-line MBC population, when adjusted for these characteristics.

The potential impact of trastuzumab or docetaxel on the pharmacokinetics of the pertuzumab was examined by comparing the individual model-predicted and observed C _trough and model-predicted PK parameters of pertuzumab among different treatment groups. The analyses showed that there was no evidence of impact of trastuzumab or of docetaxel in the presence of trastuzumab on the PK of pertuzumab. These results were not surprising since pertuzumab and trastuzumab are known to recognize different epitopes on the HER2 extracellular domain and do not compete for the same binding site [2, 16, 17]. Moreover, monoclonal antibodies (mAbs) and small molecules such as docetaxel are largely eliminated by distinct routes. Docetaxel is mainly metabolized by hepatic cytochrome P450 isoenzymes [18, 19], whereas mAbs are primarily eliminated through large-capacity, non-specific, Fc receptor-mediated IgG clearance mechanisms and through specific, target-mediated drug disposition pathways [20]. Additionally, these results are consistent with the results of a previous study that showed no DDI between pertuzumab and docetaxel and between pertuzumab and trastuzumab in the first-line MBC setting [12].

The pertuzumab clinical dosing regimen of a 840 mg fixed loading dose followed by 420 mg every three weeks was selected based on PK and safety data from studies where pertuzumab was administered as a single agent to patients with advanced refractory solid tumors, including ovarian cancer, metastatic breast cancer (low HER2 expressing), and hormone-refractory prostate cancer. Results from these clinical studies, from population pharmacokinetic analyses [21], and from dose–response studies in non-clinical xenograft models were used to determine the dose of pertuzumab used in late-stage clinical studies. In single ascending dose studies utilizing pertuzumab doses of 0.5–25 mg/kg [22, 23] and also in Phase II studies where patients were treated with either 420 mg q3w (following a loading dose of 840 mg) or 1050 mg q3w (with no initial loading dose) as a single agent or in combination with other chemotherapeutic agents, the maximum tolerated dose (MTD) was not reached and no clear dose–safety or dose–efficacy relationship was observed. In oncology, especially for chemotherapeutic agents, the MTD is often carried forward into Phase III studies. However, given that pertuzumab is a targeted monoclonal antibody and the MTD was not reached, the dose for Phase III studies was selected based on achievement of the clinical target pertuzumab concentrations (steady-state C _trough concentration of ≥20 µg/mL in 90% of patients) that resulted in maximal suppression of tumor growth in non-clinical xenograft dose–response studies [24]. In the NeoSphere study, the majority of patients (>90%) achieved the target serum concentrations. Consistent with the non-clinical xenograft studies, the ER analysis suggested that there was no association between pCR rate and pertuzumab concentrations within the observed concentration range of approximately 20–100 μg/mL, supporting no dose adjustments needed for patient with lower exposure. This analysis further supports the appropriateness of the fixed, non-weight-based pertuzumab dose of 840 mg followed by 420 mg q3w in the neoadjuvant treatment of early breast cancer patients.