Prognostic impact of human epidermal growth factor-like receptor 2 and hormone receptor status in inflammatory breast cancer (IBC): analysis of 2,014 IBC patient cases from the California Cancer Registry

Inflammatory breast cancer (IBC) is an aggressive disease characterized biologically by dermal lymphatic invasion with tumor emboli and angiogenesis, and characterized clinically by breast tenderness and skin induration overlying the breast, typically without underlying palpable mass [1]. IBC is a rapidly progressive tumor, with propensity for metastatic tumor spread, and poor overall survival (OS) compared with non-IBC locally advanced breast cancer (LABC). Although relatively rare (that is, 2% of all breast cancer patient cases), time-trend data from the US National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) database suggest that the incidence of IBC may be increasing [2, 3].

Despite multidisciplinary treatment with chemotherapy, radiation and surgery, IBC still carries a poor prognosis, with a 5-year survival rate of about 30% [4], with no significant change in the prognosis over the past 30 years [5]. Advances in the treatment of IBC have been hampered by a general lack of prognostic or predictive parameters, in part because the rarity of this tumor has made it difficult to conduct large clinical trials [6]. Population-based analyses of SEER data reveal that IBC has distinct epidemiologic differences from LABC and non-T4 breast cancer [2]. For example, IBC patient cases with estrogen receptor (ER)-negative tumors were shown to have decreased breast cancer-specific survival (BC-SS) compared with those with ER-positive tumors (2.0 years vs. 4.0 years). In addition, several small studies have shown that a high proportion of IBC tumors are positive for Her2/Neu (human epidermal growth factor-like receptor 2 (HER2)) receptor compared with historic data for non-T4 patient cases [7‐9]. HER2 is a proto-oncogene located on chromosome 17. It is overexpressed in about 25% to 30% of breast cancer and in general is associated with a more aggressive breast cancer phenotype [10]. There has been interest in investigating the role of HER2 receptor status in IBC survival, but recording of HER2 receptor status is not available in the SEER database. Previously Caggiano and colleagues used data from the California Cancer Registry (CCR) to show that HER2 status did not differentiate survival among ER-negative/progesterone receptor (PR)-negative patient cases [11]. However, results for IBC were not reported. One report indicates that HER2 receptor status is not an independent risk factor for survival in ER-negative/PR-negative invasive breast cancer in general [12]. Another report shows that HER2 status does not significantly affect recurrence-free survival in both univariate and multivariate models among 179 IBC patients [13]. Thus far, however, the usefulness of HER2 receptor status as a prognostic factor for survival among IBC patient cases in a population-based setting is still unclear.

In the present study, we use data from the CCR to estimate differences in OS for IBC, LABC, metastatic breast cancer (MBC) and non-T4 breast cancer based on hormone and HER2 receptor status. In doing so, our goal is to evaluate these potential prognostic factors for survival after diagnosis of IBC. We have restricted our analyses to a period prior to 2005, when trastuzumab became widely used in the adjuvant setting for treatment of locoregional breast cancer [14, 15].

As expected, we observed that BC-SS is lower for IBC patient cases compared with non-T4 and LABC patient cases (Figure 1). Similar to what we observed for LABC, MBC, and non-T4 breast cancer patient cases, hormone receptor-positive status was associated with improved BC-SS and OS among IBC patient cases, even after adjustment for HER2 status and other clinically relevant factors (Table 3). These findings are generally in agreement with the prior study by Brown and colleagues showing that, in invasive breast cancer, hormone receptor-negative status rather than HER2 status is a major contributor to poor survival [12]. We observed that the proportion of HER2-positive tumors is greater among IBC patient cases (40% in stage IIIB and 43% in stage IV) than LABC, MBC, or non-T4 patient cases. These findings are consistent with previous studies worldwide estimating HER2 positivity at 36% to 50% of IBC patients [9, 25]. This conserved rate of HER2 positivity among IBC patient cases occurs even in areas of the world with historically higher incidence of the disease, such as Tunisia – where IBC accounts for approximately 6% of all breast cancer (contrasted with 2.5% IBC in our study) [26, 27].

Despite reported observations that HER2 positivity occurs more frequently in IBC, very few studies have specifically examined whether HER2 receptor status was prognostic for survival in IBC patients. Dawood and colleagues reported that, in the absence of treatment with the HER2 receptor antibody trastuzumab, there is no statistically significant difference in recurrence-free survival observed between IBC patients with HER2-positive breast cancer compared with HER2-negative breast cancer [13]. Interestingly, after adjusting for patient and tumor characteristics, patients with HER2-positive disease had increased OS, presumably because most of them received trastuzumab for their recurrent disease. Trastuzumab may therefore play an important role in treatment of metastatic or recurrent IBC. Sawaki and colleagues also did not find HER2 positivity to be a significant prognostic factor in IBC; however, their study was probably underpowered to detect such a difference due to the small sample size (that is, an analysis of 46 patients within a single Japanese hospital) [9]. We obtained similar results in the present study, involving a large population-based database and sufficient statistical power to detect small differences. We observed no statistically different BC-SS characteristics for IBC patient cases based on HER2 receptor status at the time of diagnosis on unadjusted analysis, and a borderline significantly improved survival for patient cases with HER2⁺ tumors in the BC-SS adjusted analysis. Additionally, in the present analysis of data from the largest geographically contiguous cancer registry in the world, we found that HER2 is not prognostic for overall survival among LABC or MBC patient cases, and that HER2-positive status is associated with poor overall survival and BC-SS only among non-T4 breast cancer patient cases.

There are several possible explanations for why HER2 is not prognostic of decreased survival in IBC. IBC's aggressive disease characteristics may be unrelated to the presence of HER2 receptor or the HER2 receptor tyrosine kinase activity. Various theories have been proposed. For example, IBC tumors are known to be highly angiogenic and angioinvasive, with expression of proangiogenic factors that may contribute to early metastatic disease [28, 29], which therefore could be related to poor survival. Increased expression of chemokine receptors such as CCR7 [30] and genes that are related to higher metabolic rates such as Ki-67 [31] may also contribute adversely to survival. Numerous other molecules may contribute to IBC's aggressive behavior, including E-cadherin [32], Rho proteins, and WISP3 [33, 34]. Some of these molecules may be responsible for micrometastatic disease early on and thus increase IBC's tendency for recurrence [35]. Future studies may determine whether expression of these molecules is prognostic of poor survival in IBC. Finally, it has been proposed that an anthracycline-based chemotherapy regimen may be more effective in HER2-positive tumors compared with HER2-negative ones because HER2-positive tumors often have higher expression of topoisomerase II, which is the target for anthracyclines [13]. Selective killing of HER2 positive tumor cells with anthracycline-based chemotherapy (with or without anti-HER2 trastuzumab treatment) could therefore possibly eliminate HER2 positivity's impact on recurrence-free survival and OS.

Since HER2 receptor status is not prognostic for decreased survival in IBC patients, it calls into question whether antibodies directed to the HER2 receptor, such as Herceptin, may improve survival in these patients. We were not able to perform this analysis in our study because data on specific chemotherapeutic agents utilized are not available in the CCR. As described previously, in one relatively small study, trastuzumab appeared to have improved survival in recurrent or metastatic IBC, and HER2 positivity was associated with a trend towards higher overall survival [13]. A strength of the study design is that we restricted our analyses to the period prior to 2005, when trastuzumab became widely used in the adjuvant setting for treatment of locoregional breast cancer [14, 15]; our analyses should therefore not be confounded by this landmark event.

Currently there are no specific treatment guidelines for IBC. These IBC patients are often treated the same way as LABC patient cases, with intensive adjuvant or neoadjuvant chemotherapy in addition to surgery and radiation therapy. Further trials involving larger numbers of subjects with adequate power are warranted to assess the efficacy of trastuzumab in HER2-positive IBC. It is interesting that HER2 receptor status is not an adverse prognostic factor for survival after diagnosis of IBC, and yet trastuzumab may still benefit IBC patients, a finding that warrants further research. It is noteworthy that the BC-SS estimates for IBC noted in the CCR during the period 1999 to 2003 are higher than previous estimates in the SEER during the period 1988 to 2000. Many factors could explain this difference, as management strategies have evolved over time. Optimal administration schedules of anthracyclines and taxanes have been adopted in routine practice. For example, dose-dense anthracycline-based chemotherapy [36, 37], dose intensification therapy [38], and metronomic (weekly) taxane regimens [37, 39, 40] are associated with improved outcomes in breast cancer. More recently, use of trastuzumab at time of diagnosis has shown increased pathologic complete response in HER2-positive IBC, and promising progression-free survival has been reported with early follow-up [41, 42]. In as much as pathologic complete response is a surrogate of prolonged survival in IBC, future population-based data after 2005 may reveal further HER2-positive survival improvements among individuals with IBC.

There are several limitations to the present study. First, a large proportion of the data were missing for ER (16%), PR (20%), and HER2 (31%) tumor receptor status. All multivariate models were therefore analyzed with inclusion of patient cases lacking tumor receptor status. The level of HER2 amplification was not recorded in the cancer registry database. Additionally, there are no available data on the specific chemotherapy agents that patients received. It is unclear how many people received trastuzumab, for example, or received other therapy that might have affected survival. Only 81% of IBC patient cases received chemotherapy, only 75% received surgery, and 49% received radiotherapy. While these rates are low compared with what one might expect from clinical trial data, population-based data typically have lower proportions of cases receiving treatment, as all patients are included in the analysis (that is, even those with severe comorbid conditions and poor performance status). Consistent with other population-based research, there is also no information in the CCR reporting performance status or comorbid conditions that might influence survival. Finally, these population-based data came from a wide variety of sources, and thus variation of reference values for tests of hormone receptor status probably exists.

Our observations help characterize the relationship between IBC, HER2 receptor status and the effects of HER2 status effects on survival. Based on these observations, we conclude that IBC's poor survival is probably unrelated to HER2 status. Rather, the aggressive biological nature of IBC or other clinical factors may explain the poor OS for IBC. Optimization of chemotherapy schedules may have preferentially improved the outcome of HER2-positive IBC in our study. Additional improvements in HER2-positive survival could be attributed to use of trastuzumab in the relapsed or metastatic setting. Further survival improvements in HER2-positive IBC could be expected with use of trastuzumab at diagnosis, widely implemented in 2005. Validation of these findings from other population-based studies is needed, in order to help characterize the true prognostic and predictive factors for survival in IBC.