Optimal duration of prior endocrine therapy predicts the efficacy of Fulvestrant in a real‐world study for patients with hormone receptor‐positive and HER2‐negative advanced breast cancer

Abstract Fulvestrant 500 mg is standard of care for endocrine therapy‐naive or pretreated women with hormone receptor‐positive (HR+) metastatic breast cancer (MBC). This study was conducted to explore the potential factors and duration of last endocrine therapy as predictors for the efficacy of fulvestrant 500 mg on Chinese patients in real‐world practice. Two hundred and fifty‐two MBC patients who were treated with fulvestrant 500 mg consecutively between January 2011 and December 2015 in our institute were included in this study. Efficacy outcomes included progression‐free survival (PFS), overall survival (OS), and clinical benefit rate (CBR). The optimal cut‐off value for duration of last endocrine therapy was determined by survival ROC analysis. Adverse events were graded according to NCI‐CTC AE 4.0. Fulvestrant 500 mg demonstrated a median PFS of 5.8 months (95%CI 4.6‐6.9), and a median OS of 35.9 months (95%CI 30.2‐41.4). CBR was 41.3% (95%CI 35‐47). Liver metastasis, bone alone metastasis, lines of endocrine therapy for MBC, and sensitivity to last endocrine therapy were statistically significant in the Cox multivariate analysis (P values of 0.022, 0.02, 0.03, and 0.038, respectively). The optimal cut‐off values for duration of last endocrine therapy to predict the efficacy of fulvestrant 500 mg were 25.08 months for adjuvant endocrine therapy and 5.17 months for first‐line endocrine therapy, which showed no difference in prediction power with ABC clinical definition. Patients with prior adjuvant endocrine therapy ≥25.08 months or first‐line therapy≥5.17 months reached a longer PFS of fulvestrant (p = 0.04). Six patients discontinued the treatment due to intolerable adverse events. Patients with the duration of prior endocrine therapy longer than optimal cut‐off points indicate better PFS of fulvestrant. Liver metastasis, bone alone metastasis, line of fulvestrant, and sensitivity to last endocrine therapy were also predictors for response of fulvestrant. ClinicalTrials.gov Identifier: NCT03708432.


| INTRODUCTION
Hormone receptor-positive, HER2 (ERBB2)-negative (HR+HER2−) breast cancer accounts for 70% of all breast cancers. 1 Endocrine therapy is the mainstay of treatment for HR+HER2− breast cancer. 2 Though adjuvant endocrine therapy reduces the relative risk of recurrence by approximately 40%, one-third of the patients finally relapse. 3 In metastatic setting, endocrine therapy is often successful initially, but inevitably fails due to resistance. Hence, more effective treatment strategies are needed to overcome the resistance to endocrine therapy for breast cancer.
Fulvestrant is an endocrine agent used for the treatment of patients with HR+metastatic breast cancer (MBC). It has a novel mode of action since it binds, blocks, and accelerates degradation of estrogen receptor protein, leading to estrogen signaling inhibition. 4,5 Fulvestrant demonstrated efficacy in postmenopausal women with endocrine therapy-naive or pretreated breast cancer. 6,7 A phase 3 study (FALCON) 6 compared fulvestrant 500 mg with anastrozole in endocrine therapy-naive patients. Fulvestrant 500 mg showed a progression-free survival (PFS) of 16.6 months when compared to the anastrozole group of 13.8 months, and was associated with significant improvement in the PFS (hazard ratio [HR] 0·797, 95% CI 0·637-0·999, p = 0·0486). Hence, fulvestrant 500 mg proved to be an effective endocrine agent and remained a standard for the first-line treatment of HR+patients. CONFIRM trial 7 compared fulvestrant 500 mg with fulvestrant 250 mg in postmenopausal women with advanced breast cancer who experienced progression after prior endocrine therapy. Fulvestrant 500 mg significantly prolonged PFS when compared with fulvestrant 250 mg (mPFS 6.5 vs. 5.5 months). Fulvestrant 500 mg also showed a favorable clinical efficacy in postmenopausal MBC patients who progressed on prior endocrine therapy.
Besides, fulvestrant significantly improved OS of patients with HR+HER2− metastatic breast cancer, [8][9][10] and a subset of patients when given endocrine therapy alone acquired long durable disease control. Further research is warranted to study if there are any patterns for patients to distinguish who will benefit from fulvestrant, or who will be resistant and receive more intensive regimens, such as chemotherapy or combined targeted therapy. And second, duration of disease control for patients who had prior endocrine therapy is an important predictor of endocrine therapy sensitivity. According to the consensus, we always define two years in adjuvant setting and 6 months in metastatic setting as the cut-off points to separate primary and secondary resistance. 2 To date, no study has described on optimal cut-off points of prior endocrine therapy duration, especially the last-line therapy, to predict the efficacy of fulvestrant. Given that the real world study focuses on the "real effectiveness" in clinical practise, and helps us better understand the potential determinants of treatment outcomes. Therefore, this study aimed to provide additional clinical data of HR+MBC patients treated with fulvestrant 500 mg and explore the potential factors affecting its efficacy in Chinese patients.

| Patients and treatment
Our analysis comprised HR+HER2− advanced breast cancer patients who were treated with fulvestrant 500 mg from January 2011 to December 2015 in Fudan University Shanghai Cancer Center. All data were retrospectively collected from the medical records. Advanced breast cancer is defined as unresectable, locally advanced breast cancer, de novo stage IV breast cancer, and recurrent BC. HR+ is defined as ER-positive and/or PR-positive. Fulvestrant was administered by intramuscular injection in a 500 mg regimen that incorporates a day 14 loading element (500 mg on days 0, 14, and 28, and every 28 days thereafter). All the premenopausal patients received concurrent luteinizing hormonereleasing hormone analogs (LHRHa).The patients received treatment until disease progression, intolerable toxicity, or voluntary refusal. All patients provided written informed consent before collection and the process of study was approved by the relevant independent ethics committees (No. 1812195-6).
Assessment variables included PFS, OS (overall survival), ORR (objective response rate), and CBR (clinical benefit rate). PFS is defined as time from first fulvestrant administration to disease progression or death due to various causes, whichever is earlier (Disease progression is according to RECIST 1.1 criteria for tumor assessment). OS is defined as time from first fulvestrant administration to death due to various causes or the last visit of follow-ups. ORR is defined as the percentage of evaluable patients at baseline who had a best objective tumor response of either complete response (CR) or partial response (PR). CBR is defined as the percentage of evaluable patients at baseline who had a best objective tumor response of complete response (CR), partial response (PR) or stable disease (SD) ≥24 weeks.

| Statistical analysis
Descriptive statistics were used to summarize patient characteristics. Kaplan-Meier plots revealed median PFS and median OS with corresponding 95% confidence intervals and P values for all patients. A multivariate Cox proportional hazard model was developed using stepwise regression (forward selection) to explore independent predictors of PFS. Effects of variables were expressed as hazard ratios with corresponding 95% confidence intervals and P values. Significant variables in the univariate analysis entered into the model. The enter limit and remove limit were p = 0.10 and p = 0.15, respectively. Potential variables of prognostic significance included: age (<65 years or ≥65 years), menstrual state (postmenopausal vs. premenopausal), liver metastasis (no vs. yes), bone alone metastasis (no vs. yes), sensitivity to last endocrine therapy (no vs. yes), lung metastasis (no vs. yes), lines of endocrine therapy for MBC (=1 vs. ≥2), previous endocrine therapy (TAM/TOR vs. AI/Both), sensitivity to last endocrine therapy prior to fulvestrant (yes vs. no) and previous lines of chemotherapy for MBC (=0 vs. ≥1). Sensitive to endocrine therapy is defined as a relapse ≥12 months of completing adjuvant endocrine therapy, or PD ≥6 months after initiating endocrine therapy for MBC. The overall response rate (ORR) and clinical benefit rate (CBR) were calculated with its 95% CI. SPSS software (SPSS version 21.0, SPSS Inc., Chicago, IL) was used for statistical evaluations.
One of the purposes is to evaluate whether the duration of last endocrine therapy prior to fulvestrant can be a predictor of PFS for fulvestrant. We investigated the optimal cut-off values and the predictive accuracy of last endocrine therapy duration using survival receiver operating characteristic (ROC) analysis. 11,12 We used the area under the curve at 6 months to measure predictive accuracy. We used R software version 3.1.3 and the "survival ROC" package to perform survival ROC curve analysis. The data cut-off time was May 18, 2017. At the cut-off time, fulvestrant was still ongoing in 17 patients. Treatment was discontinued due to disease progression (71.0%), intolerable toxicity (2.4%), unwillingness to follow the treatment plan (2.0%), economic reasons (9.5%), and loss of follow-up (8.3%).
Patients with no liver metastasis had better PFS compared to those with liver metastasis (Figure 2A). Median PFS was 3.7 months versus 7.1 months in patients with or without liver metastasis (p < 0.001). Patients who received prior endocrine therapy for MBC demonstrated a worse PFS compared to those who did not ( Figure 2B). Median PFS was 11.9 months in patients with first-line endocrine therapy, while 4.7 months in patients with subsequent endocrine therapy (p < 0.001). Patients sensitive to last endocrine therapy reached a median PFS of months, significantly longer than patients who were not sensitive (median PFS, 7.6 months vs. 3.6 months, p = 0.002) ( Figure 2C). Median PFS was decreased significantly with successive chemotherapy regimen ( Figure 2D). For patients receiving no, first-, and second-line chemotherapy or beyond, median PFS was 9.5, 5.8, and 3.8 months, respectively (p = 0.001). Bone alone metastasis indicated a longer PFS and median PFS was 9.6 months for patients with it and 5.1 months for those with other metastasis sites (p = 0.01) ( Figure 2E).
Univariate analysis (Table 3) revealed that no liver metastasis, bone alone metastasis, first-line fulvestrant administration, previous TAM/TOR use, sensitive to last endocrine therapy and no palliative chemotherapy prior to fulvestrant was associated with significantly longer PFS. Yet, menstrual state and lung metastasis did not influence PFS. Multivariate analysis (Table 3) demonstrated that liver metastasis, bone alone metastasis, lines of endocrine therapy for MBC, sensitivity to last endocrine therapy for MBC were independent predictive factors for PFS.
In order to investigate the correlation between the duration of last endocrine therapy and PFS, we selected two subgroups: 56 patients who relapsed on adjuvant endocrine therapy and 103 patients who progressed on first-line endocrine therapy. All these patients had detailed records of prior endocrine therapy. The optimal cut-off values by survival ROC analysis were 25.08 months for adjuvant endocrine therapy and 5.17 months for first-line endocrine therapy. We further compared the predictive accuracy of optimal cut-off values in our study and conventional cut-off values for primary/secondary resistance from ABC3 guidelines 2 ( Figure  3). Results revealed no significant difference between the two classifications. Univariate analysis indicated patients who had longer duration of prior endocrine therapy (≥25.08 months in adjuvant setting/≥5.17 months in first-line) reached a longer PFS when receiving fulvestrant 500 mg (Figure 4, p = 0.038). Multivariate analysis demonstrated that duration of prior endocrine therapy was also an independent predictive factor for PFS of fulvestrant (Table 4).

| Safety
Six patients discontinued the treatment due to adverse events (AEs). One patient experienced grade 3 anorexia and discontinued the treatment. One patient stopped the treatment due to grade 3 dizziness, grade 2 nausea, grade 1 fever, and grade 1 arthralgia. These symptoms were discontinued after treatment and were considered as AEs but not symptomatic progression. One patient stopped the treatment due to grade 3 fatigue. One patient discontinued the treatment due to grade 3 blood bilirubin increase and grade 1 AST increase. Two patients stopped the drug due to grade ≥3 anemia. All patients were recovered after discontinuation of the drug and had palliative therapy for these symptoms.

| DISCUSSION
While emerging evidences confirm fulvestrant as a novel endocrine therapy due to its favorable efficacy, recent studies are focusing on its "real efficacy" in clinical practice with many other "outside trial" factors. Our study included 252 patients using fulvestrant 500 mg in real-world medical practice, and indicated a median PFS of 5.8 months and median OS of 35.9 months. In addition, CBR was 41.3%. We also explored the potential determinants of PFS by multivariable analysis, and showed that no liver metastasis, bone alone metastasis, first-line fulvestrant administration, and sensitive to prior endocrine therapy were significantly associated with longer PFS. The optimally predictive cut-off values of last-line endocrine therapy duration by survival ROC analysis were 25.08 months for adjuvant endocrine therapy and 5.17 months for first-line endocrine therapy. This study provided insights into the practical administration of fulvestrant 500 mg and provided detailed data in Chinese patients. Compared with previous non-interventional studies, our study showed consistent efficacy based on large population size (Table S1). Ishida et al. retrospectively analyzed 117 patients who were treated with fulvestrant 500 mg. 13 Results revealed that 29.1% patients had liver metastasis and 75% patients received ≥1 lines palliative chemotherapy, which was 28.9% and 56.0% in our study. It indicated that CBR was 41.9% and median time to progression (TTP) was 6.1 months, which were similar to our study (CBR 41.3%, PFS 5.8 months). Results of multivariate analysis showed that duration of first-line endocrine therapy was a determinant of TTP. In another Italian observational study, 163 patients were enrolled who received ≤2 lines of endocrine therapy for MBC. 14 The study demonstrated CBR of 61%, median PFS, and OS of 7 and 35 months, respectively. Visceral involvement, endocrine sensitivity, and previous endocrine therapy were considered as prognostic factors for PFS by multivariable analysis. Our study enrolled 31.0% patients who received ≥3 lines of endocrine therapy prior to fulvestrant and showed lower median PFS and CBR. Similarly, our study indicated that PFS was longer in patients who received fulvestrant 500 mg in early lines of   15 This study showed a median time to treatment failure (TTF) of 5.4 months. Earlier fulvestrant 500 mg use, longer period from MBC diagnosis to fulvestrant 500 mg, and no prior palliative chemotherapy for MBC were associated with significantly longer TTF. Overall, these studies, including our study, suggest that liver metastasis, bone alone metastasis, lines of endocrine therapy for MBC, lines of palliative chemotherapy, previous endocrine therapy, and endocrine sensitivity for MBC were determinants for PFS/TTP. Discrepancy in these factors may result from differences in sample size, patient characteristics and variables included in the multivariable analysis.
Fulvestrant 500 mg showed a favorable efficacy in patients with no liver metastasis. FALCON study 6 demonstrated a significantly longer PFS in patients without visceral disease compared with those with visceral disease (HR 0·59, 95% CI 0·42-0·84, mPFS for fulvestrant group 22·3 months and for anastrozole group 13·8 months). The influence of liver F I G U R E 2 Kaplan-Meier curves for PFS. For patients stratified by potential factors related to PFS. A, Liver metastasis, B, Lines of endocrine therapy, C, Sensitivity to last endocrine therapy, D, Lines of prior chemotherapy, E, Bone alone metastasis. Abbreviations: HR, hazard ratio; CI, confidence interval; PFS, progression-free survival metastasis on PFS was also identified in CONFIRM study. 7 Patients with no visceral involvement benefited more from fulvestrant 500 mg compared with fulvestrant 250 mg. These data suggest that visceral metastasis can be a predictor of PFS in fulvestrant 500 mg therapy. Our study pointed out that liver metastasis could be a good determinant of PFS in patients treated with fulvestrant 500 mg. In this study, patients with no liver metastasis showed a significantly longer PFS than those with liver metastasis (mPFS 7.1 vs. 3.7 m, p = 0.000), and was further confirmed by multivariable analysis (Table  3). Liver metastasis showed highest frequency of visceral metastasis and approximately 50% patients may develop liver metastasis. 16 Patients who develop visceral metastases to the liver generally have poorer outcomes than patients with metastases to bone or even lung, with a median survival <6 months. [17][18][19] Patients with liver metastasis represent a high-risk group of MBC patients with poor prognosis and may not respond well to conventional endocrine therapy. 6,7,20 Thus, liver metastasis is not only a prognostic factor, but also a favorable factor for predicting the efficacy of fulvestrant 500 mg in MBC patients.
Earlier fulvestrant 500 mg use, especially first-line therapy, demonstrated an enhanced benefit and prolonged PFS by 7 months (11.9 vs. 4.7 months, p = 0.000). Though there were no comparisons in prospective clinical trials between different lines of fulvestrant 500 mg, median PFS was 6.5 months in second-line patients from CONFIRM study 7 and 16.6 months in first-line patients from FALCON study, 6 demonstrating a numerically longer PFS in early fulvestrant usage. In addition, several retrospective studies showed that previous endocrine therapy or earlier fulvestrant 500 mg use was significantly associated with PFS of fulvestrant Abbreviations: CI, confidence interval; HR, hazard ratio; MBC, metastatic breast cancer.
*p < 0.05 is considered significant. 500 mg. 14,15 Subsequent lines of endocrine therapy increased the acquired resistance. Therefore, first-line administration of fulvestrant 500 mg is suggested in the treatment MBC. In our study, median PFS was decreased significantly with successive chemotherapy regimen. The results were similar to the previous studies. FALCON study 6 showed that fulvestrant 500 mg achieved a favorable efficacy in patients with no previous chemotherapy for MBC compared with anastrozole. Furthermore, Kawaguchi et al. 15 and Araki et al. 20 confirmed a significant association between PFS of fulvestrant 500 mg and previous palliative chemotherapy. These results suggested that fulvestrant 500 mg given before palliative chemotherapy could reach a longer PFS for patients with low tumor burden and /or no need for a quick response. However, we noted that patients who received fulvestrant 500 mg after one line chemotherapy for MBC achieved a median PFS of 5.8 months, which reached the median PFS of overall population, indicating that fulvestrant 500 mg has substantial activity as maintenance therapy subsequent to chemotherapy. FANCY study, 21 a single-arm phase 2 trial, explored fulvestrant 500 mg as maintenance endocrine treatment after chemotherapy. Results revealed that 58 patients with disease control after first-line chemotherapy reached a median PFS of 16·1 months (since fulvestrant treatment) and a CBR of 76%. Though this study was limited to its sample size and power of phase 2 trial, it still proved that fulvestrant 500 mg was active as maintenance therapy in non-progressive patients with HR-positive MBC after first-line chemotherapy. Our study did not estimate the median PFS of fulvestrant 500 mg after first-line chemotherapy, but only analyzed the patients who received one previous chemotherapy regimens (no limitation to lines of previous endocrine therapy), which showed a median PFS of 5.8 months. Nevertheless, all these results suggested that fulvestrant 500 mg is recommended as maintenance endocrine therapy after first-line chemotherapy.
On one hand, multivariable analysis showed no significant association of menstrual state with PFS. These results demonstrated that postmenopausal and premenopausal HR+MBC patients showed similar duration of disease For patients stratified by adjuvant ET duration and first-line ET duration. Abbreviations: HR, hazard ratio; CI, confidence interval; PFS, progression-free survival.ET, endocrine therapy control when treated with fulvestrant 500 mg. Previous phase 2-3 randomized clinical trials enrolled only postmenopausal patients and there were only few studies on the efficacy of fulvestrant 500 mg in premenopausal patients. As the mean age for diagnosing breast cancer in China is 45-55 years, which is considerably younger than the western women, 22 the treatment of premenopausal breast cancer patients remains to be a tough task in China. Recently, PROOF study, is enrolling premenopausal advanced breast cancer patients and compares efficacy of goserelin plus fulvestrant 500 mg with goserelin plus anastrozole as first-line endocrine therapy (NCT02072512). This study will provide more data regarding fulvestrant 500 mg in the premenopausal advanced breast cancer patients.
On the other hand, patients with different previous endocrine therapies, whether TAM/TOR, AI or both, showed no statistically significant difference in PFS. CONFIRM study 7 showed that patients who had prior TAM/TOR therapy favored the use of fulvestrant 500 mg compared with fulvestrant 250 mg, but this trend was not observed in patients who had prior AI. In China, CONFIRM study 23 demonstrated that patients who had previous endocrine therapy were consistent with the overall effect on PFS. A numerically longer PFS was observed in the TAM/TOR group than in the AI group in patients treated with fulvestrant 500 mg (8.1 vs. 5.8 months, respectively). We demonstrated a similar efficacy in the post-TAM/TOR and post-AI patients. Numerical differences may result from different ages and high relapse risks in patients who chose AI rather than TAM/TOR agents as adjuvant endocrine therapy.
Using survival ROC analysis, the optimal cut-off values were shown as 25.08 months for adjuvant endocrine therapy and 5.17 months for first-line endocrine therapy. The optimal cut-off values showed a numerically different AUC compared with the conventional cut-off values used in the ABC3 guideline and previous clinical trials. Survival ROC is a method for displaying sensitivity/specificity and optimal cut-off point of a continuous diagnostic marker for time-dependent disease outcomes. This demonstrated that the ABC clinical definition of primary and secondary endocrine resistance corresponded with the predicted optimal cut-off point by survival ROC. Primary and secondary endocrine resistance can separate the patients who will benefit from fulvestrant 500 mg or not, which is helpful in guiding selection of endocrine treatment only or with targeted agents. However, no validation set is available for further identification in this study.
Though the combination with CDK4/6 inhibitor extended the PFS of ER+MBC patients, 8,9 overall survival was not improved in entire trial patients except those who were sensitive to previous endocrine therapy, postmenopausal, ≥65 years old, and had a DFI > 24 months. 24 For patients who will not gain a survival benefit from CDK4/6 inhibitor, fulvestrant alone may be an alternative considering a big burden for patients to take up a long course of CDK4/6 inhibitor treatment. Previous study showed that adding palbociclib to letrozole was estimated to cost an additional CHF342,440 and gain 1.14 quality-adjusted life years, resulting in an incremental cost-effectiveness ratio (ICER) of CHF301,227/QALY gained. 25 As CDK4/6 inhibitor is not covered in the national medical insurance in China, the addition of CDK4/6 inhibitor may cause a discounted cost-effectiveness. Therefore, fulvestrant alone will still fit for some selected patients who are not available to or will not benefit from CDK4/6 inhibitor.
Our study has a few limitations. First, our study is a retrospective research and bias is inevitable due to its nature. Second, we enrolled patients who were treated with only fulvestrant 500 mg but did not include any control group. Third, our patients were from single-center, which would lead to selection bias and confounding factors.
In brief, this study was a pioneer work, and provided first-hand data regarding the efficacy and potential determinants of fulvestrant 500 mg in Chinese patients with HR+MBC. No liver metastasis, first-line fulvestrant administration, and sensitive status to last endocrine therapy prior to fulvestrant were significantly associated with longer PFS. Cost of fulvestrant 500 mg may be a major factor for poor compliance. The study showed better cut-off values for last endocrine therapy duration was 25.08 months for adjuvant endocrine therapy and 5.17 months for first-line endocrine therapy, but further studies are needed to confirm this. Yet, the results should be interpreted considering the nature of retrospective study. We expect more studies to confirm the predictive factors and biomarkers for better use of fulvestrant 500 mg.
In conclusion, fulvestrant 500 mg is effective and has excellent safety profile in HR+MBC patients. The duration of prior endocrine therapy is a predictive factor of fulvestrant and the optimal cut-off values better guide the selection of endocrine therapy. No liver metastasis, bone alone metastasis, and first-line of fulvestrant also indicate better efficacy of fulvestrant.

DECLARATIONS
The Ethics Committee and Institutional Review Board of Fudan University Shanghai Cancer Center approved this study. The need for written informed consent was waived as it is a retrospective study.