Prognostic role of tumor subtype and germline BRCA mutation in advanced breast cancer patients treated with palbociclib plus endocrine therapy

Palbociclib is a cyclin-dependent kinase 4 and 6 inhibitor which shows promising effect in hormone receptor-positive breast cancer. The purpose of this study is to evaluate the real-world efficacy and toxicity of palbociclib plus endocrine therapy. This is a retrospective study performed in two tertiary referral hospitals in Korea. Advanced breast cancer patients who were treated with 1st-line palbociclib plus endocrine therapy were enrolled. A total of 216 patients were included between August 2016 and May 2019. Median age was 56 (29–89) years old and 75 patients (34.7%) were premenopausal. Median progression-free survival (PFS) was 33.0 months [95% confidence interval (CI) 24.7 to 41.3] and objective response rate was 59.3%. Luminal B patients had shorter PFS (33.0 months vs. Not reached, p = 0.019) and tendency of lower ORR (58.3 vs. 62.0%, p = 0.19) compared to luminal A patients. Multivariate analysis revealed luminal B (adjusted hazard ratio 1.90, p = 0.038) and germline BRCA mutation (adjusted hazard ratio 5.57, p = 0.002) as an independent poor prognostic factor for PFS. The most common grade 3 or 4 adverse event was neutropenia (86.7%). The efficacy and toxicity of palbociclib in the real world were comparable to those of clinical trials. In addition, palbociclib with endocrine therapy was an effective treatment option for young patients. Luminal B and germline BRCA mutation were associated with inferior outcome.


Background
Breast cancer is the most commonly diagnosed cancer and is the leading cause of cancer death in women [1,2]. The incidence rate of breast cancer is increasing annually in developing countries, which is probably due to nation-wide cancer screening and westernization of lifestyle [3,4]. Up to 75% of breast cancers express hormone receptor (HR) and endocrine therapy is the main treatment option for these patients [5].
Recently, cyclin-dependent kinase (CDK) 4 and 6 inhibitors have emerged as a new standard treatment option for patients with advanced HR-positive, human epidermal growth factor receptor 2 (HER2)-negative breast cancer [6]. Three selective CDK 4/6 inhibitors (palbociclib, ribociclib, abemaciclib) have significantly improved progression-free survival (PFS) and overall survival (OS) in patients with HR-positive advanced breast cancer and have been approved by the United States Food and Drug Administration (FDA) [7,8,9,10]. In the phase III PALOMA-2 trial, palbociclib plus letrozole showed significant longer PFS compared to endocrine alone in HR-positive, HER2-negative advanced breast cancer [10]. In another phase III PALOMA-3 trial, patients who received palbociclib plus fulvestrant had longer median PFS and OS compared to those who received placebo plus fulvestrant [11,12].
Asian have younger age of breast cancer onset (45 to 50 years old) compared to Western population (55 to 60 years) [13,14]. In Korea, the median age at the time of breast cancer diagnosis is increasing, but still the highest incidence is seen at the age of 40-49 [15]. While palbociclib has been approved by the US FDA since 2015, there are paucity of data on the real-world efficacy of palbociclib, especially in young Asian patients [16]. The purpose of this study is to evaluate the real-world efficacy and toxicity of palbociclib plus endocrine therapy in Korean patients with advanced breast cancer. This study included relatively younger patients compared to previous performed clinical trials.

Study design and population
This study is a retrospective study performed in two tertiary referral hospitals in Korea [Seoul National University Hospital (SNUH, Seoul, Korea) and Seoul National University Bundang Hospital (SNUBH, Gyeonggi-do, Korea)]. Two hundred and sixteen breast cancer patients who received 1st-line palbociclib plus endocrine therapy between August 2016 and May 2019 were included. Patients were excluded if they received previous systemic chemotherapy for advanced/ metastatic disease. Neoadjuvant and adjuvant chemotherapy and adjuvant endocrine therapy were allowed. Patients with histologically confirmed HR-positive and HER2-negative breast cancer were included.
All patients received 125 mg of palbociclib per day administered orally in 4-week cycles (3 weeks of treatment followed by 1-week off) combined with endocrine therapy. Dose reductions of palbociclib were allowed (100 mg or 75 mg) on the discretion of treating physician. Palbociclib was combined with one of three endocrine therapies (letrozole, exemestane, or fulvestrant). Letrozole was administered orally at a dose of 2.5 mg per day, exemestane was administered orally at a dose of 25 mg per day, and fulvestrant was administered by intramuscular injection at a dose of 500 mg at day 1 and 15, 29, and then once monthly thereafter. Patients were assessed every 2-4 months during palbociclib plus endocrine treatment. All imaging studies from patients who had measurable disease were reviewed by authors (S.Y.P. and K.J.S) to ensure accuracy by RECIST v.1.1.
Eligible patients were identified from the electronic database and medical charts were identified from the electronic database medical record system of SNUH and SNUBH. The study protocol was reviewed and approved by the institutional review board of SNUH [H-1904-

Analysis of tumor subtype
Immunohistochemical (IHC) staining of estrogen receptor (ER), progesterone receptor (PR), and HER2 was performed with tumor specimen. HER2 fluorescence in situ hybridization was performed in patients with HER2 IHC 2+. Nuclear expression of tumor cells was interpreted as positive for ER and PR, while membrane staining of tumor cells was considered positive for HER2. ER and PR expression were categorized as positive when ≥ 1% of the tumor cells were stained [17]. HR was defined positive when either ER or PR was expressed. The intrinsic subtypes were defined by the 2013 St. Gallen Consensus Recommendations [18]. HRpositive, HER2-negative and Ki-67 < 14% were defined as luminal A. HR-positive, HER2-negative and Ki-67 ≥ 14% were considered luminal B [19]. Germline BRCA 1/2 mutation was tested based on the polymerase chain reaction and gene sequencing [20].
Primary endocrine resistance was defined as a relapse which occurred during the first 2 years of adjuvant endocrine therapy. Secondary or acquired endocrine resistance was defined when there was a relapse while on adjuvant endocrine therapy but after the first 2 years or within 12 months of completing adjuvant endocrine therapy [4].

Statistical analysis
The primary endpoint of this study was to investigate the real-world efficacy of palbociclib in terms of PFS. PFS was defined as the time from the start of palbociclib to radiologically or clinically disease progression or death from any cause, whichever occurred first. Secondary outcomes were objective response rates (ORR), according to RECIST criteria version 1.1 and hematologic adverse events.
Categorical variables were compared using the χ 2 test and continuous variables were compared using the independent samples T test. Missing data were not imputed. Kaplan-Meier method was used to obtain estimates of median PFS, and comparisons were made using the logrank tests. Two-sided p values of less than 0.05 were considered statistically significant. Univariate and multivariate proportional hazards regression models were used to identify independent risk factors for survival by means of the Cox proportional hazards regression models. Factors associated with prognosis were included in the multivariate analysis and forward stepwise methods were used to eliminate nonsignificant variables. Statistical analyses were performed using IBM SPSS statistics version 25 (SPSS, Chicago, IL, USA).

Patient characteristics
A total of 216 breast cancer patients who were treated with palbociclib plus endocrine therapy were included. Two hundred and five (94.9%) were treated palbociclib plus letrozole, 7 (3.2%) received palbociclib plus exemestane, and 4 (1.9%) received palbociclib plus fulvestrant. Baseline characteristics are summarized in Table 1. All patients were female with a median age of 56 (range 29-89) years. Among 75 patients (34.7%) who were premenopausal at the time of diagnosis of advanced breast cancer, 71 received bilateral oophorectomy prior to palbociclib treatment and 4 received gonadotropin-releasing hormone agonist (GnRHa) with palbociclib treatment. All patients had histologically confirmed breast cancer with HR-positive and HER2-negative. According

Real-world efficacy of palbociclib plus endocrine therapy
After a median follow-up duration of 16.0 months, 75 progression events and 11 deaths have occurred. The median interval from palbociclib initiation to the first treatment response evaluation was 9.8 weeks [95% confidence interval (CI) 9.4-10.1 weeks], and the median interval of response evaluation during palbociclib plus endocrine therapy was 12.2 weeks (95% CI 12.0-12.8 weeks). The median PFS of palbociclib plus endocrine therapy was 33.0 months (95% CI 24.7 to 41.3). Among 123 (56.9%) patients who had measurable disease, there was no complete response, 73 (59.3%) had partial response, 31 (25.2%) had stable disease, and 19 (15.4%) had progressive disease. The ORR was 59.3% (Table 2).
We next evaluated whether baseline characteristics may affect palbociclib efficacy. Patient age or menopausal status did not affect treatment response. There was no difference in PFS (not reached vs. 33.0 months, p = 0.56) and ORR (72.7% vs. 67.1%, p = 0.11) ( vs. not reached, p = 0.031) and patients who were not tested for germline BRCA status (9.0 months vs. 33.0 months, p = 0.001) (Fig. 2). PFS was similar between germline BRCA WT patients and those who did not have germline BRCA test (Not reached vs. 33.0 months, p = 0.54) (Fig. 2). There were 5 patients with germline BRCA variants of uncertain significance (VUS) and these patients showed similar outcome compared to germline BRCA WT (p = 0.591).
To adjust for baseline characteristics, we performed multivariable analysis with a Cox proportional hazard model. Due to limited number of patients who had germline BRCA status and low incidence of germline BRCA 1/2 mutation, patients who did not test germline BRCA status were combined with germline BRCA WT in the multivariate analysis. Multivariate analysis revealed nuclear grade, intrinsic subtype, and germline BRCA 1/2 mutation as an independent prognostic factor for PFS (Table 3). Luminal B subtype (adjusted hazard ratio 1.90, 95% CI 1.04-3.47, p = 0.038) and germline BRCA 1/2 mutation (adjusted hazard ratio 5.57, 95% CI 1.91-16.24, p = 0.002) were independent poor prognostic factors for PFS.
The mean time to objective response was 3.8 months. There was no significant difference in mean time to objective response between luminal A and luminal B (3.9 vs. 3.6 months, p = 0.32). The mean duration of objective response was 15.1 months. Tumor subtype did not impact mean duration of response (16.5 months in luminal A vs. 13.1 months in luminal B, p = 0.37).

Adverse events
The most common hematologic adverse events were neutropenia. Among 216 patients, 210 (97.2%) reported adverse events with any grade of neutropenia, and 182 (86.7%) reported events of grade 3 or higher-grade neutropenia. And 7 patients (3.2%) reported events of neutropenic fever. Other hematologic adverse events included anemia and thrombocytopenia. Sixty (34.5%) showed grade 3 or 4 anemia and 7 (8.8%) showed grade 3 or 4 thrombocytopenia. There was

Discussion
In recent years, CDK 4/6 inhibitors have changed the management of advanced HR-positive, HER2-negative breast cancer. There are ethnic differences in the prevalence of breast cancer that Asian countries have relatively younger age of onset compared to Western. There are yet paucity of data on the real-world efficacy of CDK 4/6 inhibitor in young Asian population. This study was conducted to investigate the real-world efficacy of palbociclib plus endocrine therapy in Asian women. We revealed that the efficacy of palbociclib plus endocrine therapy in young Asian women is comparable to that of previous conducted phase III clinical trials. The phase III PALOMA-2 trial evaluated the efficacy of 1st-line palbociclib plus letrozole in postmenopausal women with HR-positive, HER2-negative advanced breast cancer [10]. Palbociclib-letrozole group had higher ORR (55.3 vs. 44.4%, p = 0.03) and improved PFS (27.6 vs. 14.5 months, p < 0.0001) compared to placebo-letrozole group [10,21]. In PALOMA-2 trial, PFS in patients over 65 was similar to that of patients under 65 (hazard ratio for disease progression or death, 0.57) [10]. In the phase III PALOMA-3 study, women with HR-positive, HER2-negative advanced breast cancer who had progressed on previous endocrine therapy were enrolled [11]. Patients were randomly assigned to palbociclib plus fulvestrant group or placebo plus fulvestrant group. Patients who received palbociclib plus fulvestrant had improved ORR (10.4 vs. 6.3%, p = 0.16), PFS (9.2 vs. 3.8 months, p < 0.001), and OS (34.9 vs. 28.0 months, p = 0.09) compared to those who receive placebo plus fulvestrant. Twenty percentage of patients enrolled in the PALOMA-3 trial was pre-/perimenopausal and menopausal status did not affect PFS or OS [11,12].
In this study, we have evaluated the real-world efficacy of 1 st -line palbociclib plus endocrine therapy in advanced or metastatic breast cancer patients. Palbociclib plus endocrine therapy showed a median PFS of 33.0 months (95% CI 24.7 to 41.3) with an ORR of 59.3%, which is similar to that of PALOMA-2 trial. All patients included in the present study were Asian who have younger age of breast cancer onset. The median age of our study was 53 years old which is  Another important result of our study is that we showed potential negative prognostic role of germline BRCA 1/2 mutation. Patients with germline BRCA 1/2 mutation had a worse PFS compared to patients with germline BRCA wild type and those who did not test germline BRCA status. Multivariate analysis revealed germline BRCA 1/2 mutation as an independent negative prognostic factor for PFS compared to patients with germline BRCA wild type or those who did not have germline BRCA test (adjusted hazard ratio 5.57, 95% CI 1.91-16.24, p = 0.002).
Although there are limited number of patients with germline BRCA 1/2 mutation, recent studies support our data that patients with germline BRCA mutation may have worse outcome with CDK4/6 inhibitors [24]. Recent study by Safonov et al. reported that patients with germline BRCA2 mutation had worse PFS on CDK 4/6 inhibitors. It is proposed that loss of retinoblastoma protein (Rb), which is one of the main target of CDK 4/6, is associated with resistance to CDK 4/6 inhibitors [25]. RB1 and BRCA2 are both located on chromosome 13q, and germline BRCA2 loss of heterozygosity (LOH) and RB1 mono-allelic loss are highly correlated [24]. In addition, majority of patients with RB1 LOH developed RB1 loss of function mutation after CDK4/6 inhibitor treatment [24]. These results suggested that germline BRCA2 mutation could be associated with poor outcome with CDK 4/6 inhibitors. In our study, all patients who had pathogenic germline BRCA mutation had BRCA2 mutation. Our results are consistent with recent studies showing poor outcome of germline BRCA2-mutated patients treated with CDK4/6 inhibitor. We believe future study comparing frontline poly-adenosine diphosphate-ribose polymerase (PARP) inhibitor with CDK 4/6 inhibitors in HR-positive breast cancer with germline BRCA 2 mutation will provide valuable information.
There are some limitations in the present study. The tumor specimens we used to identify tumor subtype were acquired in various time points. It would be ideal to obtain tissue prior to CDK 4/6 inhibitor treatment, but unfortunately, most samples were acquired at initial breast cancer diagnosis. Another limitation of the present study is that we did not have data on other genetic characteristics to draw answer on the prognostic role of molecular findings. However, the strength of our study is that our study was relatively homogeneous in terms of ethnic and that all patients received top-of-the-line medical care in a high-volume tertiary referral hospital.

Conclusion
In conclusion, the efficacy and toxicity of palbociclib in the real world were comparable to those of previous conducted phase III clinical trials. In addition, palbociclib with endocrine therapy was an effective treatment option for young Asian breast cancer patients. However, as the ORR and PFS is inferior in luminal B subtype, CDK 4/6 inhibitor plus endocrine therapy should be used in caution in luminal B patients with visceral disease.