This study evaluated the predictive factors of pathological and clinical characteristics that can inform the prognosis of MBC patients treated with CDK4/6i. Our results demonstrated that CDK4/6i therapies had longer PFS in patients with ER levels greater than 80% and Ki-67 expression less than 30% (p < 0.05). In univariate analysis, PFS was significantly longer in patients with luminal A subtype and PR levels greater than 50%. However, multivariate analysis showed that PR levels and luminal subtypes are not associated with prolonged PFS. Similarly, the presence of liver metastases was significantly prognostic in the univariate analysis but not in the multivariate analysis. No significant OS difference was observed between luminal A and B subtypes (p:0.438).
CDK4/6is are now the standard first-line treatment for HR-positive and HER2-negative MBC. However, no strong predictor exists which tumors will respond better to CDK4/6i. It is known that the luminal subtype is not just sufficient to decide on treatment choice and predict CDK4/6i response (15). ER and PR positivity and HER2 status are widely accepted in breast cancer prognosis. It is known that higher levels of ER and PR are associated with longer PFS and OS in MBC (16). However, it is not clear to what extent ER and PR levels predict the response to CDK4/6i. ER positivity was the best predictor of CDK4/6i and hormone therapy in a large pooled US Food and Drug Administration (FDA) analysis. However, the relationship between ER level and response rate has not been investigated, and PR status was not considered as a predictive marker in this study (6). In our study, the patients with ER levels > 80% had significantly longer PFS than those with ER levels < 80% (33.7 vs 17.1 months, p: 0.002). The SONIA trial compared the patients who received first-line nonsteroidal AI (NSAI) plus a CDK4/6i and then received single-agent fulvestrant upon disease progression or the patients who received first-line NSAI and then received fulvestrant plus a CDK4/6i upon disease progression. No significant differences in PFS2 or OS emerged between treatment arms. In this case, it was discussed in which patients using CDK4/6is in the first-line setting should be preferred in terms of cost-effectiveness (17).
In an analysis of the MONARCH-2 and MONARCH-3 studies, shorter PFS was observed in patients receiving AI alone with either PR-negative or high-grade tumors. However, there was no significant relation between PFS and these factors in the group that received abemaciclib in combination with endocrine therapy (18). Cristofanilli et al. investigated the factors that predicted the long-term benefit of palbociclib and fulvestrant in their analysis of the PALOMA-3 study. They showed that a high level of PR expression was associated with a longer duration of benefit (19).
The choice of therapy in MBC is based on prognostic factors such as histological tumor type, tumor size, nodal status, grade, age, HR status, and the Ki-67 expression. Ki-67 is a nuclear protein identified in 1983 and expressed during all cell cycle phases except G0. Many studies have investigated the potential use of Ki-67 as a prognostic marker for breast cancer (16). Clinical and pathological responses to neoadjuvant chemotherapy in breast cancer patients were better in those with higher Ki-67 levels (20). The MonarchE study showed that the patients with high Ki-67 achieved a statistically significant improvement in invasive disease-free survival (iDFS) with abemaciclib in combination with adjuvant endocrine therapy (21). Although studies have shown that high Ki-67 levels are associated with better chemotherapy response in MBC, it is known that Ki-67 is a poor prognostic marker (15). Metaanalysis by Azambuja et al. confirmed that high Ki-67 levels confer a worse DFS and OS in breast cancer (22). Wang et al. identified Ki-67 as a significant predictor of PFS in patients with MBC receiving palbociclib, which is consistent with the findings of Lee et al. (23, 24). Our study showed that PFS was significantly shorter in the patients with Ki-67 expression ≥ 30%. mPFS was not reached in the patients with Ki-67 expression < 30%, while mPFS was 19.2 months in the patients with Ki-67 expression ≥ 30% (p:0.004). In the study that investigated treatment with CDK4/6i, similar to our research, PFS was negatively affected by high Ki-67 expression but not associated with PR levels (12).
The mPFS of all patients was 27.3 months, and 19.2 months in the patients with the luminal-B subtype. In the MONALEESA-7 study, mPFS reached 23.8 months in premenopausal patients with ribociclib, while in the MONALEESA-3 study, mPFS in postmenopausal patients was 20.5 months. In the MONALEESA-7 study, some of the patients had received chemotherapy before receiving CDK4/6i in the advanced stage, while in the MONALEESA-3 study, they had received hormone therapy (25, 26). The longer PFS in our study may be because of the first treatment of the patients in the study was CDK4/6i in the advanced stages. While liver metastasis is associated with poor prognosis in breast cancer patients, better PFS may be the result of small number of patients with liver metastasis in our study. Our data showed that PFS was shorter in patients with liver metastasis in univariate analysis. However, this was not statistically significant on multivariate analysis, possibly because of the small number of patients with liver metastasis.
In our study, PFS was significantly shorter in the patients using fulvestrant with CDK4/6i. The patients using fulvestrant consist who progressed while taking adjuvant AI therapies, and these patients are considered endocrine resistant. As shown in clinical studies, the contribution of CDK4/6i to PFS is less in endocrine-resistant patients than in endocrine-sensitive (23).
In the PALOMA-2 and MONALEESA-3 studies, dose was reduced in 36% and 37.9% of the patients, respectively, while this rate was 50% in our study (26, 27). Similar to the PALOMA-2 study, no significant difference in PFS was observed in patients with dose reduction in CDK4/6i treatment. The rate of all-grade adverse events in the PALOMA-2 study was 98.9%, whereas 99% in the MONALEESA-3 study. In our study, any grade adverse event occurred in 93.3% of the patients (26, 27).
This study has some limitations. Our research is a retrospective study with a limited sample size, and endocrine-resistant patients were also included. However, the present study has multiple strengths. The patients who were admitted to three different oncology outpatient clinics were included in the study, and our findings reflect real-world patient populations. The patients did not use any other treatment, such as chemotherapy or hormone therapy, before CDK4/6i treatment in the advanced stage.