Phase I clinical trial of HER2-specific immunotherapy with concomitant HER2 kinase inhibtion

The human epidermal growth factor receptor 2 (HER2), overexpressed in 20-30% of breast cancers, is associated with more aggressive tumor behavior [1]. Treatment with combinations of the anti-HER2 antibody trastuzumab and chemotherapy lengthens survival in patients with metastatic HER2-overexpressing breast cancer [2]. However, progressive disease typically occurs within one year. Lapatinib, a potent reversible inhibitor of HER2 and epidermal growth factor receptor (EGFR) tyrosine kinases [3], in conjunction with chemotherapy, increases time to progression in these patients [4]. Unfortunately, responses to lapatinib are generally short-lived, and progression remains a significant clinical problem.

Intriguingly, the overexpression of HER2 persists in trastuzumab and lapatinib-refractory tumors [5, 6], and thus, targeting HER2 with cancer immunotherapy is a potentially effective strategy. A variety of vaccines targeting HER2, based on proteins, peptides, modified tumor cells, viral vectors, pDNA and dendritic cells (DC) have been developed. Results from phase I and II studies of HER2-targeting cancer vaccines [7] have demonstrated that HER2 is immunogenic, and that immune responses against HER2 may be associated with an improved clinical outcome [8‐13].

One protein-based vaccine, dHER2 Antigen-Specific Cancer Immunotherapeutic (ASCI) a recombinant HER2 protein, including a truncated intracellular domain (ICD) and the complete extracellular domain (ECD), combined with the immunological adjuvant AS15, containing MPL, QS21, CpG and liposome, was evaluated in two early phase clinical studies of patients with HER2-overexpressing breast cancer (NCT 00058526 and NCT 00140738) [14]. In both studies, the data showed that dHER2 immunizations were well tolerated, consistently immunogenic at the 500 μg dose and that clinical activity (including prolonged stable disease) was associated with antibody and T cell responses.

One important observation from the prior dHER2 ASCI studies was that the polyclonal antibody-containing serum from immunized patients had functional activity against signaling pathways mediated by HER2. Specifically, incubation of breast cancer cell lines with serum from two immunized patients demonstrated an impact on molecular pathways resembling that of trastuzumab [14]. Because clinical trials have demonstrated that combinations of lapatinib and trastuzumab lead to enhanced clinical activity and combined effects on signaling pathways [15], there has been interest in combining the polyclonal anti-HER2 serum with trastuzumab and indeed, increased apoptosis of human HER2-overexpressing breast cancer cells was observed when lapatinib was combined with HER2-specific polyclonal antisera generated from rabbits immunized with dHER2 ASCI [16]. We therefore hypothesized that the lapatinib would enhance the anti-signaling activity of the polyclonal Abs induced by the dHER2 vaccine in humans. First, it was necessary to establish that the induction of anti-HER2 antibodies by the dHER2 vaccine was not affected by lapatinib and this was the primary purpose of this study.

The purpose of this study was to determine the clinical and immunologic effects of immunization against HER2 concurrent administration of the HER2 tyrosine kinase inhibitor lapatinib. Previously, we reported that a HER2-targeting vaccine could induce antibody and T cell responses when administered concomitantly with lapatinib in a murine model [18]. We extended this work in the present study confirming for the first time in humans that concomitant administration of lapatinib did not appear to affect the immune response induced by the HER2 immunotherapy. Although we are not aware of other data regarding the effect of lapatinib on the immune response in humans, other tyrosine kinase inhibitors have demonstrated negative effects on the immune response, such as sorafenib (which targets RAF protein in the EGFR pathway as well as other targets) while others, such as sunitinib, have had no detrimental effects or potentially improve immune responses [19, 20].

We observed potent antibody responses to HER2 ECD and ICD, suggesting that we could still break tolerance to a self-antigen in the context of lapatinib. In the prior phase I/II studies of dHER2 ASCI, antibody responses were consistently observed, and CD4+ and CD8+ T cell responses were also reported. The preferential induction of antibodies using HER2 protein-based vaccines has been already reported. In a series of patients immunized with the truncated 146HER2 protein complexed with cholesteryl pullulan alone or with various adjuvants [21‐23], antibody responses against HER2 were detected in 14 of 15 patients, but only 5 out of 9 developed detectable HER2-specific CD8+ and/or CD4+ T-cell immune responses. Because the development of antibodies generally requires Th2 cells, but our method for measuring IFN-gamma ELISPOT identifies Tc1 and Th1, and not Th2 responses, it is possible that a Th2 response occurred, but was not detectable. It is also possible, as has been reported by others using protein-based vaccines, that regulatory T cells (Treg), typically increased in advanced cancer patients, caused decreases in T cell responses [24]. Future strategies to induce greater T cell responses could be warranted as some authors have suggested the T cell responses correlate with clinical outcome [25]. We have developed a series of recombinant viral vectors expressing HER2 that have been evaluated preclinically and were shown to induce HER2-specific CD8+ T cells. These vectors will be entering clinical trials shortly [26]. We have also reported strategies for depleting Treg [27].

One emerging aspect of cancer immunotherapy is that long periods (between 4 to 6 immunizations) are needed to reach an optimal humoral or cellular response. We noted that some patients rapidly progressed during immunization, but prior to development of detectable HER2 specific antibody responses, or until the antibody responses are of sufficient quantity or quality to impact disease. In most cases these were patients who had been maintained on trastuzumab prior to participation. The decline in anti-ECD Ab responses in the early weeks of the study likely represents trastuzumab levels declining. This suggests that for future vaccine studies, patients may need to receive concomitant trastuzumab until they have an adequate induction of antibody responses. Disis et al. showed that patients could be vaccinated in the setting of trastuzumab without additional toxicity. The expected immune responses including epitope and antigen spreading were observed [28]. Indeed, in our study, two patients with detectable pre-vaccination anti-ECD antibodies demonstrated an increase in HER2-ECD antibody titer following vaccination (Figure 3), suggesting we could also induce antibody responses against the dHER2 despite the presence of trastuzumab.

Recent clinical trials have established longer survival for patients receiving trastuzumab plus lapatinib, suggesting clinical benefit for antibody plus tyrosine kinase inhibition of HER2. Because this was not a randomized trial, we are not able to determine the clinical benefit of combined dHER2 ASCI plus lapatinib; however, our observed median PFS and overall survival in a heavily pretreated population with multiple sites of metastasis who had progressive disease on trastuzumab and prior lapatinib compares favorably with other clinical data. For example, although there is insufficient data in the literature regarding the outcome of patients refractory to both drugs, among patients who have progressive disease while receiving trastuzumab therapy, overall survival has ranged from 10 - 19 months [29‐31]. In the EGF104900 study, which randomized patients who had progressed on trastuzumab to either lapatinib alone or lapatinib in combination with trastuzumab, PFS of 8 weeks and OS of 39 weeks (or 273 days) was reported for the lapatinib only arm [31]. It has also been observed in some immunotherapy studies, that OS is improved despite a lack of effect on PFS suggesting that delayed effects of the vaccine might have led us to prematurely discontinue immunization. Future studies should permit a greater extent of progression (such as 50% growth) in asymptomatic progression as has been suggested in the literature [32].

Another important aspect of the strategy of combining a HER2 immunotherapy with lapatinib is the potential combined inhibition of HER2 signaling. A detailed analysis of effects of anti-HER2 serum by our group is reported elsewhere (Rong et al., manuscript submitted). For the current manuscript, we performed a preliminary analysis in which antibodies, purified from post-vaccination serum from selected patients (P002, P004, and P008), were tested for their ability to reduce phosphorylation of the downstream molecule AKT in the HER2 overexpressing cell line SKBR3. In the patient with the highest titer of ECD among the three (P004), pAKT was decreased by antibodies purified from the post-vaccination serum, providing a suggestion that there are anti-signaling effects of the dHER2-induced antibodies (Additional file 2: Figure S1). Serum from patients in the previous phase I/II studies of dHER2 alone bound HER2-overexpressing breast cancer cell lines and inhibited growth of these cell lines with an effect on molecular pathways resembling that of trastuzumab [14]. We previously reported that the polyclonal sera induced in mice by a HER2 vaccine were superior to monoclonal antibodies in mediating receptor internalization and degradation, resulting in ablation of HER2 signaling over time. Furthermore, there was increased anti-tumor activity when this vaccine was administered concomitantly with lapatinib to mice [18]. These data suggest a potential advantage to our approach. First, the vaccine may provide durable periods of exposure to biologic levels of anti-HER2 antibody, whereas trastuzumab must continue to be administered. Second, the vaccine induces polyclonal antibody responses which could have immunologic functions (such as antibody dependent cellular cytotoxicity) as well as direct functions on HER2 signaling. Recently, it was reported that lapatinib induced HER2 surface expression in HER2-positive breast cancer cell lines, leading to the enhancement of trastuzumab-mediated ADCC [33]. These potential advantages strongly support the further assessment of HER2-targeting immunotherapies with concomitant lapatinib and/or trastuzumab in human clinical trials.

One potential disadvantage of the vaccine strategy is that the antibody titers are lower than those achieved during trastuzumab administration, suggesting future vaccine strategies may need to enhance the titer of antibody induced. Based on the tolerability and safety established in the current study, we will be embarking on a study to evaluate the use of viral vector-based vaccine encoding HER2 along with lapatinib. The potential benefits of such immunotherapeutic strategies over a monoclonal antibody approach (e.g., induction of both T cell and polyclonal antibody responses with multiple mechanisms of action) warrant this testing. More broadly, we believe that targeting receptor molecules using immunotherapy as a mean to perturb signaling offers potential new opportunities to target cancer beyond the conventional lytic killing of tumors by the immune system.

In summary, the data presented is consistent with prior preclinical and clinical observations and supports the expansion of combination studies. However, the limited numbers of patients, the inability to assess clinical responses compared with controls, and the potential effects of waning trastuzumab levels are challenges that would be best addressed in a randomized trial.