Characteristics, Treatment, and Outcomes of Real-World Talazoparib-Treated Patients With Germline BRCA-Mutated Advanced HER2-Negative Breast Cancer

Abstract Background Talazoparib is a poly (adenosine diphosphate-ribose) polymerase inhibitor approved for the treatment of adult patients with deleterious or suspected deleterious germline BRCA-mutated (gBRCAm), HER2-negative, locally advanced or metastatic breast cancer (LA/mBC), with approval based on the EMBRACA trial. To date, there are no published data on talazoparib use in the real-world United States (USA) setting. Patients and Methods Characteristics, treatment patterns, and clinical outcomes of real-world US patients with gBRCAm HER2-negative LA/mBC treated with talazoparib monotherapy were collected via retrospective chart review and summarized using descriptive statistics. Results Among 84 eligible patients, 35.7% had hormone receptor-positive tumors and 64.3% had triple-negative LA/mBC (TNBC). At talazoparib initiation, 29.8% had ECOG PS of ≥2 and 19.0% had brain metastasis. Mutations in gBRCA1 or 2 were detected among 64.3% and 35.7% of patients, respectively. Talazoparib was given as 1st-line therapy in 14.3% of patients, 2nd-line in 40.5%, and 3rd- or 4th-line in 45.2%. Median time to talazoparib treatment failure was 8.5 months (95% CI, 8.0-9.7), median progression-free survival was 8.7 months (95% CI, 8.0-9.9), the median time from initiation to chemotherapy was 12.2 months (95% CI, 10.5-20.1), and the overall response rate was 63.1%. No differences in clinical outcomes were observed between patients with HR-positive/HER2-negative LA/mBC and patients with TNBC by using unadjusted statistical comparisons. Brain metastasis and ECOG PS ≥2 at talazoparib initiation were associated with treatment failure and progression or mortality. Conclusion Overall, talazoparib clinical outcomes in this real-world population are consistent with findings from EMBRACA.


Introduction
Breast cancer continues to be the leading cancer diagnosis and a significant contributor to cancer-related mortality with 290 560 new cases and 43 780 deaths estimated in the United States (USA) in 2022. The prognosis for locally advanced or metastatic breast cancer (LA/mBC) is poor, having an estimated 5-year survival rate of approximately 29.0% with distant metastasis. 1,2 Treatment decision-making is typically informed by molecular characteristics including hormone receptor (HR) status, human epidermal growth factor receptor 2 (HER2) expression, and breast cancer susceptibility gene (BRCA) mutation status. Inherited, also known as germline, BRCA (gBRCA) mutations are associated with a higher lifetime risk of developing LA/mBC and experiencing poorer outcomes. 3 However, these mutations are relatively rare, and are detected in less than 5% of unselected patients with metastatic disease. 4-7 gBRCA1/2 mutations are associated with worse breast cancer-specific survival compared with sporadic or negative BRCA status. 3 Inhibition of poly (adenosine diphosphate-ribose) polymerase (PARP) activity has been demonstrated to effectively target tumors with deleterious mutations in BRCA by blocking the ability of cancer cells to effectively repair DNA damage. 7 Since 2018, 2 PARP inhibitors (PARPis) have been approved by the US Food and Drug Administration (FDA) for the treatment of adult patients with deleterious or suspected deleterious gBRCA mutated (gBRCAm) HER2negative advanced breast cancer: olaparib and talazoparib. 8,9 Differences between these agents include PARP trapping potency and safety profiles.
Olaparib was approved in January, 2018 by the FDA for treatment of adult patients with gBRCAm HER2-negative metastatic breast cancer who received prior chemotherapy in the neoadjuvant, adjuvant, or metastatic setting, and, if the disease is HR-positive, had either received prior endocrine therapy or been considered inappropriate for it. 8 Approval in this indication was based on phase III randomized trial, OlympiAD (NCT02000622), which demonstrated that olaparib was associated with superior progression-free survival (PFS) over physician's choice of single-agent chemotherapy (capecitabine, vinorelbine, or eribulin) (7.0 vs. 4.2 months, respectively, hazard ratio 0.58; 95% CI, 0.43-0.80; P < .001). 6,7 In addition, treatment with olaparib was associated with a higher objective response rate than treatment with chemotherapy (59.9% [95% CI, 52.0-67.4] vs. 28.8% [95% CI, 18.3-41.3], respectively). 6 Talazoparib was approved by the FDA for use in adult patients with gBRCAm HER2-negative LA/mBC in October, 2018 based on the phase III EMBRACA trial (NCT01945775), 10 which demonstrated that treatment with talazoparib in comparison to protocol-specific, nonplatinum-based single-agent chemotherapy of the physician's choice (ie, capecitabine, eribulin, gemcitabine, or vinorelbine) showed statistically significant improvement in PFS (median 8.6 vs. 5.6 months; hazard ratio 0.54; 95% CI, 0.41-0.71; P < .001]) and a significantly higher objective response rate (62.6% vs. 27.2%; odds ratio [OR] 5.0; 95% CI, 2.9-8.8; P < .001). 11,12 The median treatment duration for EMBRACA was 6.9 months with talazoparib and 3.9 months with single-agent chemotherapy. 13 Several studies have reported patient characteristics, treatment patterns, and outcomes for talazoparib-treated patients with advanced breast cancer in real-world settings in France, 14 Turkey, 15 and Russia. 16 However, to the authors' knowledge, there are no published studies to date on talazoparib utilization and outcomes in the real-world US setting. The objective of this retrospective chart review study (NCT04987931) was to describe the characteristics, treatment patterns, and clinical outcomes among adult patients with gBRCAm HER2negative LA/mBC treated with talazoparib in real-world practice settings in the USA.

Data Source and Study Design
This study consisted of a retrospective, observational, physicianabstracted, and multi-site medical chart review. Medical oncologists from the Cardinal Health Oncology Provider Extended Network (OPEN) abstracted data from medical records of US patients selected based on prespecified eligibility criteria. OPEN is a geographically diverse, electronic medical record/group purchasing organization-agnostic, US community of >7000 oncologists, hematologists, and urologists in community practices ranging in size from private solo practitioners to hospital systems.
Study inclusion criteria required patients to have a diagnosis of gBRCAm HER2-negative LA/mBC to have initiated talazoparib monotherapy on or after October 16, 2018 (the date of FDA approval in this setting), to be at least 18 years of age at initiation, and to have at least 6 months of follow-up after initiation unless deceased within this time frame. Exclusion criteria included participation in any breast cancer clinical trial after initiation of talazoparib, treatment with a PARPi as neoadjuvant/adjuvant therapy, lack of known gBRCA1/2 mutation, unknown HER2 status, and diagnosis with any other malignancy except carcinoma in situ or nonmelanoma skin cancer within the 5 years prior to data collection. Participating oncologists were instructed to select up to 20 eligible patients chronologically, starting with the earliest patient meeting the criteria, and selecting patients consecutively thereafter, to minimize selection bias. Physicians provided information about physician characteristics and abstracted data from medical records related to patient demographics, clinical characteristics, treatment patterns, and clinical outcomes (eg, response and progression) in an electronic case report form. Data were collected between August 20, 2021 and October 11, 2021.
The study received approval and exemption for obtaining patient informed consent by a central Institutional Review Board (IRB) (WCG IRB). The study followed acceptedstandard research guidelines.
Clinical outcome measures for this study included timeto-treatment failure (TTF) for talazoparib, defined as the time from initiation of talazoparib to discontinuation for any reason, including disease progression, treatment toxicity, and death; real-world PFS (rwPFS) for talazoparib, defined as the time from initiation of talazoparib to charted disease progression or death from any cause, whichever occurred first; time from initiation of talazoparib to chemotherapy; real-world overall response rate (rwORR) for talazoparib, calculated as the sum of the number of physician-reported complete responses (CRs) and partial responses (PRs) divided by the total number of patients with reported disease response assessment; talazoparib real-world duration of response (rwDOR); and overall survival (OS) as the time from initiation of talazoparib to death from any cause.
For the evaluation of treatment patterns, a line of therapy for LA/mBC included chemotherapy-, hormonal therapyand/or targeted therapy-based regimens. A line of therapy ended when the patient discontinued treatment with a regimen, added a new treatment to the regimen, or switched to a new regimen. The addition of a new treatment to an ongoing regimen was considered a new line of therapy. If a treatment used as part of combination therapy was held or discontinued, this was not considered a new line of therapy.

Statistical Analysis
Data related to patient demographics and clinical characteristics, treatment patterns, and clinical outcomes were summarized using descriptive statistics. Analyses were conducted for the full final analysis cohort and 2 patient subgroups based on HR status (ie, patients with HR-positive disease and patients with triple-negative LA/mBC (TNBC). The Kaplan-Meier (KM) method was used to describe time-to-event outcomes for talazoparib including TTF, rwPFS, time from initiation of talazoparib to chemotherapy, rwDOR, and OS. Patients who were still alive or had not developed the progressive disease were censored on the date of talazoparib discontinuation or the last in-person or telemedicine visit at the physician's practice, whichever occurred first. Statistical comparisons of clinical characteristics and outcomes between the HR-positive and TNBC subgroups were conducted using Chi-square tests for categorical variables and t-tests or Wilcoxon tests for continuous variables. Log-rank tests were used for subgroup comparisons of time-to-event outcomes. Cox proportional hazards (PH) regression was used to investigate the association of baseline demographic and clinical variables with TTF and rwPFS. All analyses were conducted in SAS v9.4.

Patient Characteristics
In total, 84 patients treated by 9 community oncologists met eligibility criteria and were included in this analysis. Patient characteristics are shown in Table 1 At the initiation of talazoparib, all patients had stage IV disease, 96.4% had visceral metastases, and 19.0% had brain metastases. Approximately 30.0% had an Eastern Cooperative Oncology Group performance status (ECOG PS) ≥2 at the initiation. A gBRCA1 mutation was detected in 64.3% of patients, whereas a gBRCA2 mutation was detected in 35.7% of patients. gBRCA1/2 mutations were identified from testing using multigene panels among 79.8% of patients, single gene testing among 4.8%, and unknown panel types among 15.5%. Testing most frequently occurred prior to 1st-line (1L) therapy for LA/mBC (69.0% of patients). Testing for gBRCA1/2 was conducted during 1L treatment for metastatic disease in 25.0% of patients, between 1L and second-line (2L) in 2.4%, and during third-line (3L) in 3.6% of patients. Median (25th percentile to 75th percentile [p. 25-75]) time to talazoparib initiation after gBRCA mutation testing results was 15.8 months (8.6-37.2 months). Among 44 patients with triple-negative LA/mBC (TNBC) and programmed death ligand 1 (PD-L1) testing results available, PD-L1 positivity was noted in 47.7% of patients. In addition, among patients with available testing results, PIK3CA mutations were found in 2 of 29 patients (6.9%) and ESR1 mutations were found in 1 of 8 patients (12.5%).
HR-positivity was reported for 35.7% (n = 30) of patients and TNBC classification was reported for 64.3% (n = 54). Although similar distributions across sex, race, and ethnicity were observed between patients with HR-positive disease and those with TNBC, the former had a higher median age at initiation of talazoparib (68.6 vs. 59.1 years, P < .01). Differences in geographic distribution were observed between these subgroups, with a higher proportion of patients with TNBC   (Table 2).
In unadjusted comparisons, no statistically significant differences were detected between patients with HR-positive disease and patients with TNBC in talazoparib TTF, rwPFS, time from talazoparib initiation to subsequent chemotherapy initiation, rwORR, or rwDOR (Table 3; Fig. 1).
Patient characteristics associated with TTF and rwPFS were identified using univariate and multivariate analyses, the latter of which adjusted for age at initiation of talazoparib, HR status, race, brain metastasis at the initiation of talazoparib, ECOG PS at the initiation of talazoparib, and line of therapy for LA/mBC in which talazoparib was received. In univariate and multivariate Cox PH model analyses, the presence of brain metastases and ECOG PS of ≥2 was associated with a significantly higher risk of treatment failure (respective TTF hazard ratios, 2.3; 95% CI, 1.1-4.8; P = .02 and 2.7; 95% CI, 1.3-5.8; P = .01) and significantly higher risk of progression or death (respective rwPFS hazard ratios, 2.4; 95% CI, 1.2-5.0; P = .02 and 2.7; 95% CI, 1.3-5.8; P = .01) ( Table 4).

Discussion
This study aimed to describe talazoparib utilization and clinical outcomes in the initial 3 years post-FDA approval among US adult patients with gBRCAm HER2-negative LA/mBC. This study was the first real-world evaluation conducted in the USA. However, similar studies have been conducted in France, 14 Turkey, 15 and Russia 16 and have been published.
Patient demographic and clinical characteristics differed between this real-world study and the EMBRACA trial. Patients in this real-world study were older (median age at talazoparib initiation of 62 vs. 45 years in EMBRACA), had worse performance status (ECOG PS at talazoparib initiation of 2+ among 29.8% vs. 2.1% in EMBRACA), and were treated with talazoparib in later lines of therapy (talazoparib was received as 3L or 4L therapy among 45.2% of patients in this study, whereas in EMBRACA, 24.0% of patients initiated talazoparib after 2 or 3 prior cytotoxic regimens for advanced breast cancer). 11,13 Despite these differences, median rwPFS and rwORR for talazoparib observed for this real-world US cohort were consistent with those reported for patients in the talazoparib arm of the EMBRACA trial. The median rwPFS for talazoparib among real-world US patients was 8.7 months (median follow-up 8.2 months), and the rwORR was 63.1%; for patients in EMBRACA, the median PFS for talazoparib was 8.6 months (median follow-up 11.2 months) and the objective response rate was 62.6%. [10][11][12][13] The median OS of 11.6 months from talazoparib initiation observed in this real-world study The Oncologist, 2023, Vol. 28, No. 5 419 was shorter than the median OS of 19.3 months observed in the EMBRACA trial. 13 However, in this real-world study, the median duration of follow-up was only 8.2 months, the rate of censoring was >50%, and patients were older, with worse performance status, and more heavily pretreated than in EMBRACA. Notable differences between this real-world study and EMBRACA were observed in transfusions received during talazoparib treatment and in talazoparib dosage reductions or interruptions. Among talazoparib-treated patients in the EMBRACA trial, 39.2% received at least 1 red blood cell

All patients (N = 84) HR+ (n = 30) TNBC (n = 54)
Line of therapy that talazoparib was administered for the treatment of LA/mBC, n (%) Talazoparib dosage modification (reduction or temporary interruption), n (%) a 13 ( transfusion and 3.5% received platelet transfusion; the rates in this study were 8.3% and 0.0%, respectively. 13 Talazoparib dosage reductions and interruptions due to adverse events in the EMBRACA trial were reported in 53.1% and 62.6% of patients, respectively. In this real-world study, 14.3% of patients had talazoparib dosage reduction for any reason and 6.0% had temporary dosage interruption for any reason. These differences may have resulted from protocol requirements for supportive care and dosage modification according to hemoglobin level in the EMBRACA trial, which may not reflect routine clinical practice in the real-world setting. 13 Findings from the real-world US study were also similar to those reported in the phase IV ViTAL study in France, which included 86 patients with gBRCAm HER2-LA/mBC, as the median TTF of talazoparib for US patients was 8.5 months and the median time to treatment discontinuation, defined as the time between the date of the first dose of talazoparib and the date of last dose or death, for patients in the French study was 8.6 months. 14 No difference was observed according to HR status for either study. However, the median rwPFS and rwORR for talazoparib were more favorable in this realworld US study versus those observed in small real-world studies conducted in Turkey (n = 47) and Russia (n = 24). It should also be noted that patients in the Turkish and Russian studies were heavily pretreated and initiated talazoparib in later lines of therapy (3 or more lines of therapy were received prior to talazoparib among 10.7%, 51.5%, and 33.5% in the real-world studies set in the USA, Turkey, and  a Statistical comparisons between HR-positive/HER2-and TNBC subgroups. Comparisons between subgroups did not adjust for differences in patient characteristics. b Time from talazoparib initiation to discontinuation for any reason. Patients still on therapy at last encounter were censored at last encounter date. c Time from initiation of talazoparib to charted disease progression based on radiographic imaging or death from any cause, whichever occurred first. Patients who discontinued talazoparib for a reason other than progression or death were censored at talazoparib discontinuation date. Patients still receiving talazoparib at last encounter were censored on date of last encounter. d Sum of complete and partial responses reported as best response divided by all patients with reported disease response assessment. e Time from initial documentation of disease response to talazoparib (complete or partial) to disease progression or death. Patients who did not progress and were reported to be alive or lost to follow-up were censored at last encounter date. f Time from initiation of talazoparib to initiation of subsequent chemotherapy, received as any line of therapy after talazoparib. Patients who had not received chemotherapy at last encounter were censored at last encounter date or date of death, whichever occurred first. g Time from initiation of talazoparib to death. Patients who were known to be alive or lost to follow-up were censored on the date of last encounter. Abbreviations: CI, confidence interval; CR, complete response; gBRCAm, germline breast cancer susceptibility gene mutated; HER2-, human epidermal growth factor receptor 2 negative; HR+, hormone receptor-positive; KM, Kaplan-Meier; LA/mBC, locally advanced or metastatic breast cancer; max., maximum; min., minimum; OS, overall survival; PR, partial response; rwDOR, real-world duration of response; rwORR, real-world overall response rate; rwPFS, real-world progression-free survival; TNBC, triple-negative breast cancer; TTF, time-to-treatment failure.
Russia, respectively). 15,16 The median rwPFS for talazoparib was 8.7 months (median follow-up period of 8.2 months) for the US study population and 6.5 months (median follow-up time of 13.6 months) for the Turkish BRCA-mutated breast cancer patient study population. 15 In addition, rwORR for talazoparib was 63.1% for the US study population versus 31.9% for the Turkish study population. 15 In the Russian study, the median rwPFS and rwORR for talazoparib were 6.5 months and 29.0% for patients with gBRCA-mutated HER2-metastatic breast cancer. 16 It is likely that distinctions in patient characteristics and the small sample sizes of the Turkish and Russian studies may have impacted these differences, and rwPFS and rwORR were improved for subgroups treated with talazoparib in earlier lines. 15,16 Limitations This study may be limited by unobserved data and missing data bias (eg, undercounting of events that are unknown to the abstracting oncologists because of having occurred outside the office/clinical setting, loss to follow-up if patients transferred care to other providers or clinics). Source document verification was not conducted; however, all physicians had been required to submit to data validation checks, and failure to correctly validate data resulted in exclusion. Moreover, this study included a limited number of patients (84), with data abstracted by a limited number of oncologists (9). Patients were selected based on prespecified selection criteria and, hence, findings may not be representative of all patients who have received talazoparib. In addition, treatment and testing patterns may not reflect those of all oncologists managing patients with LA/mBC. Although the study sample size was limited, the potentially eligible patient population is considered to be rare, given the fact that gBRCA mutations are detected in less than 5% of unselected patients with mBC, 6 and gBRCA mutation testing is not universally performed. A US chart review study of 407 HER2-negative mBC patients reported that 47% were tested for gBRCA1/2 mutation with Patients still on therapy at last encounter were censored at last encounter date. b Time from initiation of talazoparib to charted disease progression based on radiographic imaging or death from any cause, whichever occurred first. Patients who discontinued talazoparib for a reason other than progression or death were censored at talazoparib discontinuation date. Patients still receiving talazoparib at last encounter were censored on date of last encounter. c Time from initiation of talazoparib to death. Patients who were known to be alive or lost to follow-up were censored on the date of last encounter. d Time from initiation of talazoparib to initiation of subsequent chemotherapy, received as any line of therapy after talazoparib. Patients who had not received chemotherapy at last encounter were censored at last encounter date or date of death, whichever occurred first. Abbreviations: CI, confidence interval; gBRCAm, germline breast cancer susceptibility gene mutated; HER2-, human epidermal growth factor receptor 2 negative; KM, Kaplan-Meier; LA/mBC, locally advanced or metastatic breast cancer; OS, overall survival; rwPFS, real-world progression-free survival; TTF, time-to-treatment failure.
or without somatic mutation. Testing also differed by patient age (more younger patients were tested), family history of BRCA1/2-related cancer (more patients with family history were tested), cancer subtype (more patients with TNBC versus HR-positive disease were tested), and care setting (patients were more likely to be tested in academic versus community practices). 17 In terms of data related to supportive care medication use during talazoparib treatment, data collection did not distinguish whether patients were prescribed medication as prophylactic therapy or as needed. Finally, the findings of this study may be impacted by a lack of uniform assessment criteria for certain variables such as disease response.

Conclusion
This study provides the first published findings from realworld gBRCAm HER2-LA/mBC patients treated with talazoparib outside the clinical trial setting. Importantly, the results are concordant with those of the EMBRACA trial, and randomized controlled trials remain the gold standard for the evaluation of safety and efficacy of pharmaceutical interventions. However, clinical outcomes in clinical trials and those observed in real-world populations may differ. 18 Moreover, the main limitation of clinical trial data remains the lack of representativeness and generalizability of the sample population to real-world patients for whom the drug approval was indicated. 19,20 In addition, barriers to clinical trial participation encompass sociocultural norms, health system impediments, and clinical attitudes and practices differentiating patient care by race and/or ethnicity. 20 Due to the voluntary nature of participation in clinical trials, most US patients (approximately 95%) are not represented and may therefore experience outcomes not replicated in the real-world setting (eg, community or academic practice). 20,21 This study included a racially/ethnically diverse cohort of patients similar in racial and ethnic distribution to that of the broader US population and was specific to real-world practice.
In conclusion, findings from this study, which represents the first such published evaluation conducted in the USA, show the clinical benefits of talazoparib treatment in gBRCAm HER2-LA/mBC in real-world practice. Clinical outcomes in this real-world population of patients treated with talazoparib were consistent with those reported in the EMBRACA randomized clinical trial.