Sunitinib in combination with trastuzumab for the treatment of advanced breast cancer: activity and safety results from a phase II study

Trastuzumab is approved in combination with taxanes for first-line treatment and as monotherapy for second-line treatment of HER2-positive metastatic breast cancer (MBC). Objective response rates (ORRs) for patients with MBC receiving trastuzumab monotherapy were 26% in the first-line setting [1] and approximately 15% in the second-line setting [2, 3]. However, resistance to trastuzumab alone or in combination with chemotherapy generally develops within 1 year of initiating treatment [1, 2, 4‐6].

Sunitinib malate (SUTENT®; Pfizer Inc., New York, NY, USA), an oral, multitargeted tyrosine kinase inhibitor (TKI) of vascular endothelial growth factor receptors (VEGFRs), platelet-derived growth factor receptors (PDGFRs), and other receptor tyrosine kinases [7‐12], is approved multinationally for the treatment of advanced renal cell carcinoma (RCC), imatinib-resistant/-intolerant gastrointestinal stromal tumor (GIST), and progressing metastatic pancreatic neuroendocrine tumor. In a phase II study of heavily pretreated patients with MBC (N = 64), single-agent sunitinib at 50 mg/day on Schedule 4/2 (4 weeks on treatment followed by 2 weeks off treatment) demonstrated antitumor activity (ORR, 11%; clinical benefit rate [CBR; percent objective responses plus stable disease (SD) ≥ 24 weeks], 16%) [13]. In some patients in this trial, tumor regrowth was observed during the period off treatment following initial shrinkage on treatment. Sunitinib administration at 37.5 mg on a continuous daily dosing (CDD) schedule has also been found to be active and feasible in RCC, GIST, and pancreatic neuroendocrine tumor [14‐17], and was used subsequently in sunitinib trials in MBC.

Preclinical and clinical data support the rationale that concurrent inhibition of VEGF and HER2 signaling may be more efficacious than inhibition of either target alone [18‐21]. VEGF may in part mediate the aggressive breast cancer (BC) phenotype associated with HER2 overexpression [18, 19], and constitutively active HER2 has been shown to increase VEGF protein synthesis levels in human BC cells [20].

In a mouse model of HER2-amplified BC, sunitinib plus trastuzumab elicited a 75–80% greater decrease in tumor volume than either agent alone (Pfizer Inc., data on file). A phase II study of the anti-VEGF monoclonal antibody bevacizumab and trastuzumab as first-line therapy for HER2-positive MBC yielded an ORR of 46% [21], which compared favorably with results obtained with trastuzumab alone in other studies [1, 22]. Preliminary results of a phase III study also suggested that addition of bevacizumab to the combination of trastuzumab and docetaxel in a similar setting leads to modest improvements in clinical outcomes [23]. Since preclinical studies had demonstrated that dual inhibition of the VEGF and PDGF signaling pathways provide greater antitumor activity than inhibition of either pathway alone [24, 25], it was hypothesized that the addition of the multitargeted TKI sunitinib to a trastuzumab-based regimen would be especially efficacious.

The current phase II study evaluated the efficacy and safety/tolerability of sunitinib on a CDD schedule in combination with trastuzumab, weekly or every 3 weeks, in patients with HER2-positive advanced (metastatic or locally recurrent) BC (ABC). The primary objective of the study was to determine the antitumor activity of the combination. Sunitinib is not approved for ABC, and Pfizer did not submit a request for regulatory review of sunitinib in ABC by the FDA or other regulatory bodies. This decision was made following the findings that sunitinib did not meet primary study endpoints in phase III trials in this setting [26‐29].

Sunitinib 37.5 mg on a CDD schedule in combination with trastuzumab (weekly or 3-weekly) demonstrated substantial antitumor activity in patients with ABC, with an ORR of 37% and a CBR of 56%. The null hypothesis that the ORR of the sunitinib - trastuzumab combination was no different than the historical trastuzumab ORR was therefore rejected. A 1-year survival rate of 91% was achieved, and median OS was not reached (survival was only followed for 2 years post-dose).

The ORR of the combination was greater than the 11% ORR reported in two previous studies of sunitinib monotherapy (administered at 50 mg/day on Schedule 4/2 or 37.5 mg/day on a CDD schedule) in previously treated patients with ABC [13, 32]. In the current study, most responses (71%) were noted in treatment-naïve patients and in patients who had received only prior adjuvant therapy. These patients achieved an ORR of 44%, similar to that observed in an earlier trial of first-line bevacizumab plus trastuzumab (46%) [21]. The high ORR obtained with the sunitinib–trastuzumab combination in patients with visceral disease (44%) was also encouraging. These observations provide additional support for synergy between tumor-specific HER2-targeted and antiangiogenic therapies for aggressive disease, as predicted in preclinical studies (Pfizer Inc., data on file). In particular, the action of sunitinib on both vascular endothelial cells and pericytes [7, 8, 33] as a result of dual targeting of VEGFR and PDGFR may complement HER2-targeting by trastuzumab, although trastuzumab resistance did not appear to be overcome in patients receiving the combination as second-line therapy.

In general, the sunitinib - trastuzumab combination appeared to have an acceptable safety profile that was broadly consistent with the profiles of both drugs administered as monotherapy [13, 32, 34], with the majority of AEs being of mild to moderate severity. Dosing modifications were frequently used to manage AEs, with 80% of patients having temporary treatment discontinuations and/or dose reductions of one or both study drugs due to AEs.

Cardiac dysfunction is a known side effect of trastuzumab, with reported incidences of 3–7% when given as monotherapy and 27% when administered with anthracycline-containing chemotherapy [35]. Cardiac dysfunction has also been associated with sunitinib treatment, with reported incidences of 11–19% in patients with RCC or GIST [36‐38]. Given that the drugs were used in combination in this study, LVEF was monitored frequently. Forty percent of patients (24/60) experienced AEs related to measured LVEF declines, although in the majority of these patients (75%) the events were asymptomatic (CTCAE grade 1/2), with measured changes in LVEF during the events being between -4% and -37% relative to baseline. Pre-exposure to anthracyclines appeared to be a major factor contributing to cardiac dysfunction: 19 of the 24 patients with LVEF-related AEs overall (79%) and all six with symptomatic LVEF-related events had received prior anthracycline treatment, either with or without prior trastuzumab therapy. Of the 23 patients in the study who had not received prior anthracyclines, only five experienced LVEF-related AEs (22%). This rate of LVEF decline following prior anthracycline treatment was consistent with that reported in other studies using trastuzumab [2, 22]. Median baseline LVEF was similar between the 37 patients who had received prior anthracyclines and the 23 who had not (61% vs. 64%).

In total, five patients (8%) discontinued the present study for reasons related to cardiac dysfunction: three discontinued due to measured LVEF declines (ranging from -18% to -30% relative to baseline), one due to acute heart failure (with an LVEF decline of 27%), and one due to asthenia, cardiac insufficiency, and dyspnea (with an LVEF decline of 37%). Additionally, as noted above, one patient died due to cardiogenic shock (with a change in LVEF from 54% at screening to 23% 3 days prior to death). In the majority of the 24 patients with LVEF-related AEs (67%), the events resolved either spontaneously or following temporary or permanent discontinuation of trastuzumab (and, in one case, sunitinib as well), in contrast to cardiac dysfunction associated with anthracycline treatment, which is usually irreversible [39]. Nevertheless, the results of the present study, including the high rate of grade 3/4 cardiotoxicity in anthracycline-exposed patients, indicate that great caution along with proactive cardiac monitoring and management of cardiotoxicity by treatment interruption/discontinuation are critical when using a drug combination such as that tested in this study.

Steady-state concentrations of sunitinib, the active metabolite SU12662, and total drug were consistent with those obtained with single-agent sunitinib administered at 50 mg/day on Schedule 4/2 to patients with MBC [13]. Pharmacokinetic evaluations suggested that no clinically relevant drug - drug interactions had occurred. In addition, antitumor response appeared to correlate with plasma drug exposure: ORR and PFS appeared to be greater in patients with higher plasma drug exposures. Correlations were also observed between plasma drug exposures and several key safety parameters.

In conclusion, sunitinib on a CDD schedule in combination with trastuzumab (weekly or 3-weekly) demonstrated antitumor activity in patients with HER2-positive ABC, particularly in those who were treatment-naïve or had only received prior adjuvant treatment. Sunitinib plus trastuzumab had acceptable safety and tolerability in patients with ABC who had not received prior anthracycline therapy. The regimen is not being developed further, however, based on disappointing results obtained in four phase III studies of sunitinib in patients with ABC [26‐29]. Nevertheless, the results obtained in this study contribute to the field of antiangiogenesis by adding to the evidence supporting a beneficial effect of targeting both the VEGF and HER2 pathways and by providing a platform for further exploration with other agents that may lead to benefit in specific patient populations.