To the best of our knowledge, the present study was the first to find that the positivity level of ER expression had a significant prognostic role, even in HR+/HER2– IBC. Furthermore, the GE exploratory analyses indicated that MYC was the key gene in understanding the biological behavior of HR+/HER2– IBC.
In contrast to the predictive value, we identified the prognostic role of ER in HR+/HER2– IBC. Basically, the HR positive breast cancer shows low percentage of pCR because of the tumor dormancy, and as more tumor stage advanced, it could be more difficult to obtain pCR [8, 7]. Indeed, present study showed only thirteen (7%) in pCR for IBC and thirty-three (5%) for non-IBC, which was very small compared with early breast cancer. Therefore, the low rate of the event possibly made it difficult to determine the effect of HR positivity on pCR. Notably, patients with HR+/HER2– IBC had a wider range of heterogeneity in survival outcome according to ER positivity, and those with high ER expression had a better prognosis, which was similar to that of non-IBC patients. The results indicated that ER positivity also had an important prognostic role even in patients with HR+/HER2– IBC.
The present study also detected the optimal cutoff points for survival in HR+/HER2– IBC at 91.5% for ER and 9% for PR. Furthermore, these cutoff points were IBC-specific since they could not be applied to corresponding non-IBC. Unfortunately, however, the external validation study failed to show the universality of newly detected cutoff points on prognosis. In fact, the distribution of HR expression was significantly different between MD Anderson’s cohort and the validation cohort, showing 72.2% and 80.4% in mean ER and 40.4% and 53.8% in PR for MD Anderson’s cohort and the validation cohort, respectively. Accordingly, OS was generally better in the validation cohort than in the MD Anderson cohort. The difference in survival was probably due to the fact that most of the patients with IBC at MD Anderson were referred from community clinics and our dataset included more complexed or advanced cases with comorbidities. In addition, the difference in diagnostic criteria for IBC could affect the outcome. Further investigation is needed to establish the globally applicable cutoff point.
In the GE analysis, MYC was found to be upregulated in 3 of 4 pathways overlapping the ER pathway, and the gene had a significant impact on survival outcome in IBC. MYC is a regulator gene coding for transcriptional factors involved in cell cycle and cell growth. Generally, MYC amplification was observed in more aggressive subtypes such as HER2 and triple-negative types [14], as well as in advanced clinical status [15], leading to poor survival outcome [16, 17]. For IBC, MYC has been investigated mainly in the triple-negative type [18, 19]; however, the present study found that MYC was also upregulated in HR+/HER2– IBC, leading to a significant association with poor survival outcome.
Generally, MYC expression was associated with cell cycle activity with increased cyclin B1 and Ki-67 expression [17] and can be a predictive marker for endocrine therapy resistance [20]. Indeed, we observed MYC upregulation in ERK/MAPK and PDGF pathways, which have a significant role in endocrine therapy resistance [21, 22]. The activation of ERK/mitogen-activated protein kinase induces tamoxifen resistance by altering the level of estrogen-related receptor γ (ERRγ), which is an orphan member of the nuclear receptor superfamily. Furthermore, ERRγ-driven transcriptional activity is impaired by the mutation of ERK target sites, leading to the tamoxifen resistance [21]. For PDGF pathway, the clinical study on 45 breast cancer patients treated with aromatase inhibitor showed that the protein expression of PDGF receptor α and β in tumor was significantly increased at the point of relapse and the higher expression was correlated with shorter time to treatment failure [22]. Although the detailed mechanism for endocrine therapy resistance by MYC for HR+/HER2– IBC needs to be further investigated, the results in the present study suggest that MYC possibly contributed to poor prognosis due to either intrinsic characteristics or endocrine treatment resistance.
Notably, MYC upregulation contributed to survival outcome only in RFS and DDFS but not in OS for HR+/HER2– IBC. Previous studies had suggested that IBC has a unique metastasis process characterized by higher lymphatic invasion, tumor embolization, activated inflammatory pathways, and increased growth factors [23]. The MYC gene codes transcription factors and regulates every stage of the metastasis process including cell proliferation, angiogenesis, and epithelial-to-mesenchymal transition [24]. However, it is unclear whether MYC has any specific effect on the metastasis process, especially for IBC. We reported that the metastasis for IBC was associated with a risk allele at 8q24 where MYC located [25]. Moreover, we determined that the MYC activation in IBC was caused by the dysfunctional antagonization of MYC by the activation of SMAD3, which was located downstream of the TGF-beta signaling pathway [26]. Since MYC can be activated by upstream signaling pathways and codes many transcriptional factors, more comprehensive gene analysis will be needed to elucidate how MYC affects the metastasis process in HR+/HER2– IBC.
The limitation of the present study is that we excluded a certain number of patients during the selection process because they did not have a detailed pathological report; most of these patients had been evaluated outside of MD Anderson. Although we cannot estimate the result of excluding these patients, it is possible that HR distribution and the cutoff point may have been different if all cases had been included in the analysis. Moreover, the antibody used for IHC and the definition of HER2 positivity was not consistent over the study period, which possibly affected the overall results.