nab-Paclitaxel for the treatment of breast cancer: an update across treatment settings

Purpose The purpose of this systematic review is to discuss recent studies and ongoing trials of nab-paclitaxel in breast cancer and to examine the potential role of nab-paclitaxel as a backbone for immuno-oncology therapies. Methods PubMed and selected congress proceedings were searched for studies of nab-paclitaxel in breast cancer published between 2013 and 2015. All phase II and III clinical trials, retrospective analyses, and institutional studies were included. Active, ongoing, phase II or III trials on nab-paclitaxel that were listed on ClinicalTrials.gov were also included. Results Sixty-three studies, including 23 in early-stage and 30 in metastatic breast cancer (some studies not classifiable by setting), were included in this analysis. Trials of neoadjuvant nab-paclitaxel–containing regimens have reported pathological complete response rates ranging from 5.7 to 53%. Median overall survival in metastatic breast cancer studies ranged from 10.8 to 23.5 months, depending on dose and regimen. Adverse event profiles of nab-paclitaxel were generally similar to those reported from previous studies. Several ongoing trials are evaluating nab-paclitaxel in the early-stage and metastatic settings, including in combination with immuno-oncology agents. Conclusions nab-Paclitaxel continues to demonstrate promising efficacy in breast cancer. Recent studies demonstrate high pathological complete response rates in early-stage breast cancer, particularly in triple-negative breast cancer, an area of high unmet need, and encouraging overall survival in metastatic breast cancer across doses and schedules. Ongoing trials will provide further insights into the role of nab-paclitaxel in breast cancer including use as a potential backbone chemotherapy agent for immuno-oncology therapies such as checkpoint inhibitors.

Treatments for localized breast cancer consist of surgical resection with or without radiation therapy [1]. Neoadjuvant chemotherapy is recommended for large tumors (stage IIA-B or T3N1M0). The primary approach Brufsky Exp Hematol Oncol (2017) 6:7 for managing metastatic breast cancer (MBC) is systemic therapy, consisting of cytotoxic chemotherapy, endocrine therapy for hormone receptor-positive disease, and HER2-targeted agents for HER2-positive cancers. Many of the neoadjuvant, adjuvant, and metastatic chemotherapy regimens preferred by the National Comprehensive Cancer Network include paclitaxel [1]. One disadvantage of paclitaxel is the development of hypersensitivity reactions to the solvent, Kolliphor EL (formerly called Cremophor EL) [2]. Nanoparticle albumin-bound paclitaxel (nab ® -paclitaxel, Celgene Corporation, Summit, NJ) is solvent-free, minimizing hypersensitivity reactions and potentially other solvent-related toxicities, such as neutropenia [3][4][5]. Due to minimal risk of hypersensitivity, premedication with prophylactic steroids is not required [3,4]. Another advantage of nab-paclitaxel vs standard paclitaxel is the increased rate of transport across endothelial cell layers, greater and faster tissue penetration, and slower elimination of paclitaxel [6][7][8]. nab-Paclitaxel also demonstrates increased intratumoral delivery and retention, resulting in 33% higher intratumoral drug concentrations [6]. Compared with paclitaxel, nab-paclitaxel yields a 10-fold higher mean maximal concentration of free paclitaxel [8].
nab-Paclitaxel is currently approved for locally advanced or metastatic non-small cell lung cancer, metastatic pancreatic cancer, and MBC that has progressed on combination chemotherapy or relapsed within 6 months of adjuvant chemotherapy [2]. The approval in MBC was based on a randomized phase III trial of nab-paclitaxel 260 mg/m 2 vs paclitaxel 175 mg/m 2 every 3 weeks (q3w). nab-Paclitaxel demonstrated a significantly higher overall response rate (ORR; 33 vs 19%; P = 0.001) and longer time to tumor progression (23 vs 17 weeks; hazard ratio [HR], 0.75; P = 0.006) vs paclitaxel in the intention-totreat (ITT) population [5]. Overall survival (OS) was not significantly different between the 2 treatment groups in the ITT population; however, in the second-or laterline setting, OS was significantly longer for nab-paclitaxel vs paclitaxel (median, 56 vs 47 weeks; HR, 0.73; P = 0.024). Significantly less grade 4 neutropenia (9 vs 22%; P < 0.001) and more grade 3 sensory neuropathy (10 vs 2%; P < 0.001) were reported with nab-paclitaxel [5]. Grade 3 sensory neuropathy in patients who received nab-paclitaxel improved to a lower grade after a median of 22 days of treatment interruption.
Recent studies have examined nab-paclitaxel in early-stage breast cancer, primarily as a neoadjuvant agent. Studies also continue to evaluate the efficacy of nab-paclitaxel in MBC in combination and across doses and schedules. This review summarizes data from recent studies of nab-paclitaxel across breast cancer settings, discusses ongoing trials, and provides perspectives on the future role of nab-paclitaxel in breast cancer.

Methods
PRISMA guidelines were followed in this systematic review. PubMed was searched for articles published between January 1, 2013 and February 28, 2016. Abstracts from the American Society for Clinical Oncology (ASCO) annual meeting and the ASCO Breast Cancer Symposium 2013-2015 were included. The entry terms for the search were "nab-paclitaxel" and "breast. " Abstracts from the 2014 European Breast Cancer Conference and the 2013-2015 San Antonio Breast Cancer Symposium proceedings were searched using the term "nab-paclitaxel. " Phase II and III clinical trials, retrospective analyses, and institutional studies were included. Duplicate studies, topic reviews, case studies, nonhuman or preclinical studies, and non-English articles were excluded. One article in PubMed was embargoed and inaccessible.

Results
The publication selection process is depicted in Fig. 1. Twenty-three studies of nab-paclitaxel in early-stage breast cancer were retrieved, including 21 neoadjuvant and 2 adjuvant studies. Three post hoc analyses of previous neoadjuvant trials were included in the 21 retrieved neoadjuvant studies. Studies of neoadjuvant nab-paclitaxel in early-stage breast cancer are presented in Table 1. There were also 30 studies of nab-paclitaxel in MBC, including 3 health economic analyses. Studies of nab-paclitaxel in MBC that reported progression-free survival (PFS) or OS are presented in Table 2. Only studies with ≥50 patients are detailed in the text; however, all reports on early-stage HER2+ disease were included because of the small number of studies.

Studies of nab-paclitaxel in early-stage breast cancer Unselected (all subtypes)
Among 7 phase II and III studies of neoadjuvant nabpaclitaxel (majority administered weekly) that did not select for specific disease subtype, the pathological complete response (pCR) rate ranged from 22 to 40%; 4 phase II studies included <50 patients (Table 1).
A trend toward benefit for stage II disease was observed with nab-paclitaxel 125 mg/m 2 once weekly (qw) plus carboplatin vs paclitaxel 80 mg/m 2 qw plus carboplatin in a phase II trial with a pCR rate of 36.8 vs 15.8% (odds ratio [OR], 3.11; 95% CI 0.963-10.053; P = 0.051); however, no such trend was observed in the overall population [9]. More grade 4 neutropenia was observed with nab-paclitaxel than with paclitaxel (56.7 vs 21.1%; P < 0.001).
The large phase III GeparSepto trial compared neoadjuvant paclitaxel 80 mg/m 2 qw vs nab-paclitaxel 125 mg/m 2 (150 mg/m 2 before amendment) qw followed by epirubicin and cyclophosphamide (EC) for early-stage breast cancer, with trastuzumab plus pertuzumab added for HER2-positive cancers [11]. The original dose of nabpaclitaxel (150 mg/m 2 ) was amended to 125 mg/m 2 due to the frequency of treatment discontinuations and sensory neuropathy. Overall, patients achieved a significantly higher pCR rate with nab-paclitaxel vs paclitaxel (38.4 vs 29.0%; P < 0.001). The higher pCR rate for nab-paclitaxel vs paclitaxel was maintained in the set of patients who received treatment after the nab-paclitaxel dose amendment (41.4 vs 32.4%; P = 0.013). The largest difference between treatment arms was observed in the triple-negative breast cancer (TNBC) subgroup, with nab-paclitaxel achieving a pCR rate of 48.2 vs 26.3% with paclitaxel (P < 0.001).
The pCR rates in 4 phase II neoadjuvant studies of nabpaclitaxel for the treatment of early-stage breast cancer of unselected subtype ranged from 22 to 40%, with 71 to 77.5% of patients having breast-conserving surgery [9][10][11][12].

HER2-negative or TNBC
Nine studies of nab-paclitaxel in early-stage HER2-negative breast cancer or TNBC were retrieved. The overall pCR rate ranged from 5.7 to 53% ( Table 1). Most of these were combination studies.
The phase II Nabrax GEICAM study of neoadjuvant nab-paclitaxel 150 mg/m 2 the first 3 of 4 weeks (qw 3/4) in patients with ER-positive, HER2-negative breast cancer demonstrated an ORR of 76.5% [13]. A residual cancer burden (RCB) score of 0 + 1 was achieved by 24.7% of the treated population (n = 81), and the rate of conversion to breast-conserving surgery was 40%.
The phase II SWOG S0800 trial evaluated the backbone neoadjuvant regimen of nab-paclitaxel 100 mg/m 2 qw with dose-dense doxorubicin plus cyclophosphamide (AC) ± bevacizumab for the treatment of HER2-negative locally advanced or inflammatory breast cancer [14]. The overall pCR rate was 28%, with a significantly higher pCR rate reported in the bevacizumab vs non-bevacizumab arm (36 vs 21%; P = 0.021). In hormone receptor-positive disease, the difference in pCR rate between bevacizumab and no bevacizumab was not significant (25 vs 18%; P = 0.41). However, patients with hormone receptor-negative tumors demonstrated a significantly improved pCR rate with bevacizumab (59 vs 28%; P = 0.014). In addition, a significantly improved pCR rate was achieved with bevacizumab vs no bevacizumab in the locally advanced breast cancer group (37 vs 22%; P = 0.035).
The phase II ADAPT trial was designed as an umbrella trial in which patients with early-stage breast cancer   had 2 sequential core biopsies during neoadjuvant therapy (baseline and 3 weeks after treatment initiation) to assess early biomarker changes and guide adjuvant therapy selection [16]. Patients were assigned to 1 of 4 subtrials based on molecular subtyping. One subtrial was the ADAPT-TN trial, which evaluated a backbone of neoadjuvant nab-paclitaxel 125 mg/m 2 given the first 2 of 3 weeks (qw 2/3) with either carboplatin or gemcitabine in patients with TNBC [17,18]. The pCR rate differed significantly between arms (carboplatin, 47.4% vs gemcitabine, 29.7%; P = 0.0045). Patients who received nab-paclitaxel plus gemcitabine vs nab-paclitaxel plus carboplatin experienced a significantly higher frequency of dose reductions (20.6 vs 11.9%; P = 0.03), treatment-related severe adverse events (13 vs 5%; P = 0.02), grade 3-4 infections (6.1 vs 1.3%; P = 0.04), and alanine aminotransferase elevations (11.7 vs 3.3%; P = 0.01). Docetaxel 75 mg/m 2 q3w was compared with nabpaclitaxel 100 mg/m 2 qw 3/4 followed by FEC with epirubicin at 100 mg/m 2 for stage II-III, HER2-negative breast cancer [19]. The overall pCR rate was 17%, with a higher pCR rate of 30% in patients with TNBC. Another trial that evaluated sequential nab-paclitaxel 260 mg/m 2 q3w and EC demonstrated a pCR rate of 5.7% in patients with stage II-III HER2-negative breast cancer (N = 53) [20].

HER2-positive
Three studies of nab-paclitaxel in early-stage HER2-positive breast cancer demonstrated highly consistent pCR rates, ranging from 45 to 49% (Table 1). Neoadjuvant carboplatin, nab-paclitaxel, and trastuzumab treatment in patients with stage II-III HER2-positive tumors resulted in a pCR rate of 45% in 55 evaluable patients, and a pCR plus RCB I rate of 50% in the ITT population [21]. A pCR rate of 52% was achieved in patients with ER-negative disease compared with a pCR rate of 40% in those with ER-positive disease. A phase II trial of neoadjuvant anthracycline followed by nab-paclitaxel 260 mg/m 2 q3w plus trastuzumab for operable HER2-positive breast cancer reported a pCR rate of 49% in the treated population [22]. Subgroup analysis demonstrated a higher pCR rate of 71% in patients with ER-negative tumors vs 36% in those with ER-positive disease. Similar results were achieved with neoadjuvant nab-paclitaxel 260 mg/m 2 every 2 weeks (q2w) followed by vinorelbine and trastuzumab for stage I-III HER2-positive breast cancer [23]. An overall pCR rate of 48.1% was reported, with a pCR rate of 68.8% in the hormone receptor-negative population and a pCR rate of 18.2% in the hormone receptorpositive population (n = 11).

Adjuvant treatment with nab-paclitaxel
Two studies of adjuvant nab-paclitaxel in early-stage disease were retrieved. The combination of nab-paclitaxel 100 mg/m 2 and capecitabine was compared with EC or cyclophosphamide, methotrexate, and fluorouracil (CMF) as adjuvant therapy for early-stage breast cancer in nonfrail elderly patients (age ≥65 years) in the phase II ICE II-GBG 52 trial [27]. After a median follow-up of almost 23 months, no significant difference in OS was observed between treatment arms (HR, 1.18; 95% CI 0.52-2.66). A greater percentage of patients experienced treatment discontinuations with nab-paclitaxel plus capecitabine than EC/CMF (35.8 vs 6.6%; P < 0.001). In both arms, the main reasons for discontinuation were adverse events followed by investigator or patient decision; however, grade ≥3 toxicities were less frequent with nab-paclitaxel plus capecitabine than EC/CMF (64.8 vs 90.9%; P < 0.001). Grade ≥3 hematologic events were less common (22.3 vs 88.4%; P < 0.001) and grade ≥3 nonhematologic events were more common (58.5 vs 18.7%; P < 0.001) with nab-paclitaxel plus capecitabine than with EC/CMF. The authors suggested that tolerability might have been better if a lower capecitabine dose were used in the nab-paclitaxel plus capecitabine arm.
Another ongoing phase III trial (Nordic trip, N = 1800), will provide additional information about the potential benefit of nab-paclitaxel plus EC as a treatment for early TNBC [29]. Patients will be randomized to receive adjuvant or neoadjuvant treatment in 1 of 3 arms: nabpaclitaxel followed by EC, nab-paclitaxel plus capecitabine followed by EC plus capecitabine, or nab-paclitaxel plus carboplatin followed by EC. The primary endpoint is invasive disease-free survival.

Studies of nab-paclitaxel in MBC Unselected (all subtypes)
Four studies of nab-paclitaxel in MBC of unselected subtype reported median OS ranging from 10.8 months with nab-paclitaxel 260 mg/m 2 q3w to 26.9 months with nabpaclitaxel 125 mg/m 2 qw 3/4 combined with cisplatin ( Table 2).
A phase II trial of nab-paclitaxel 125 mg/m 2 qw 3/4 and cisplatin for MBC reported an ORR (primary endpoint) of 67.1%, with an 80.6% ORR in the first-line setting and an 80% ORR for those who were not previously treated with taxanes [31]. The median PFS was 9.8 months (95% CI 8.1-11.6 months) by investigator assessment, and median OS was 26.9 months. Compared with patients who were pretreated with taxanes, those who had not previously received taxanes demonstrated longer PFS (median, 8.5 vs 11.2 months; P = 0.009) by investigator assessment and longer OS (median, not reached vs 16.7 months; P < 0.001). There were no significant differences in PFS according to molecular subtype. Treatment was well tolerated in most patients, with neutropenia being the most common cause for dose adjustment. Grade ≥3 neutropenia was the most common hematologic adverse event, affecting 84.9% of patients.
A randomized phase II/III trial of women with refractory MBC compared nab-paclitaxel 260 mg/m 2 q3w with paclitaxel concentrate for nanodispersion (PICN) 260 or 295 mg/m 2 q3w [32]. Comparing the equal-dose regimens of nab-paclitaxel and PICN, ORR and PFS were numerically greater for nab-paclitaxel vs PICN (ORR, 43 vs 35%; PFS, median, 7.8 vs 5.3 months [P not significant for either]). ORR and PFS were not significantly different for higher-dose PICN vs lower-dose PICN or vs nabpaclitaxel. The lower-dose PICN arm had lower rates of grade ≥3 adverse events compared with the higher-dose PICN arm and the nab-paclitaxel arm.
A number of regional retrospective analyses have also been conducted recently on the use of nab-paclitaxel for the treatment of unselected MBC. A retrospective analysis of patients with breast cancer in British Columbia who received nab-paclitaxel from 2007 to 2011 was performed [33]. Most patients had metastatic disease, and 2 had regional relapse only. Approximately onefourth of patients had prior taxane exposure. Time to relapse was significantly shorter in patients with prior exposure to adjuvant taxanes vs those without (median, 2.7 vs 4.5 years, P < 0.001). No significant differences in time to treatment failure (defined as time from first to last cycle of nab-paclitaxel) or dose reduction rates were found between these 2 groups. Thus, nab-paclitaxel may result in clinical benefit in patients with MBC regardless of whether they have had prior taxane exposure. A retrospective German survey of national chemotherapy practices demonstrated that nab-paclitaxel was used less frequently than paclitaxel and docetaxel for first-line treatment of MBC [34]. However, retrospective analysis of a US claims database of patients with MBC who were treated with taxanes (n = 2599 docetaxel; n = 1643 paclitaxel; n = 261 nab-paclitaxel) demonstrated that patients remained on nab-paclitaxel 50% longer than on other taxanes (127 vs 85 days; P < 0.05), possibly due to lower incidence of discontinuation for neutropenia with nab-paclitaxel vs docetaxel or paclitaxel (6.9% vs docetaxel, 29.4% or paclitaxel, 17.5%; P < 0.001) [35]. A separate analysis of a US claims database revealed that Brufsky Exp Hematol Oncol (2017) 6:7 nab-paclitaxel was most often administered as secondor later-line therapy, as monotherapy, or on a weekly schedule [36].

HER2-negative or TNBC
A phase II trial of nab-paclitaxel 100 mg/m 2 qw 3/4 with or without anti-death receptor 5 monoclonal antibody tigatuzumab (10 mg/kg loading, 5 mg/kg q2w) was performed in patients with metastatic TNBC [37]. ORR was 28% in the combination arm vs 38% in the nab-paclitaxel-alone arm. PFS was not significantly different between arms (median, 3.7 months for nab-paclitaxel monotherapy vs 2.8 months for the combination; P = 0.3152). Five patients in the combination arm demonstrated long-term PFS (334-1025+ days) compared with 1 patient in the nab-paclitaxel arm (1004+ days).
There was one study of nab-paclitaxel in patients with HER2-negative MBC (unselected for hormone receptor status) that reported median PFS [38]. None of the studies on HER2-negative breast cancer or TNBC reported OS.

Health economic analyses
Three cost-effectiveness analyses compared nab-paclitaxel with other taxanes in patients with MBC.
The Spanish COSTABRAX cost-effectiveness analysis demonstrated that nab-paclitaxel q3w was cost-effective compared with paclitaxel q3w as a second-line treatment for MBC [39]. Efficacy data from the phase III CA012 trial were used in a Markov model expanded to a time horizon of 5 years. The cost of life-year gained with nabpaclitaxel q3w vs paclitaxel q3w was €11,088, and the cost per quality-adjusted life-year (QALY) gained was €17,808. Compared with paclitaxel qw, nab-paclitaxel q3w showed a savings of €711 per patient. Another study combined cost data in China with a meta-analysis of 10 randomized phase III trials of nab-paclitaxel 260 mg/m 2 q3w or docetaxel 100 mg/m 2 q3w in comparison with paclitaxel 175 mg/m 2 q3w [40]. The cost per course of treatment was $19,752 for nab-paclitaxel, $8940 for paclitaxel, and $13,741 for docetaxel. The cost per QALY gained for nab-paclitaxel vs docetaxel as alternatives to paclitaxel was $57,900 vs $130,600. Thus, nab-paclitaxel appeared to be a more cost-effective alternative to docetaxel as initial therapy for MBC in a Chinese healthcare setting. The Italian COSTANza study, which used a Markov model, also suggested that nab-paclitaxel was cost-effective, with a gain of 0.165 QALY compared with paclitaxel [41].
A recent questionnaire-based study of healthcare providers (N = 22) in Sweden indicated that mean infusion times for nab-paclitaxel vs paclitaxel were 42.1 ± 20.7 vs 104.3 ± 43.3 min, respectively [42]. Total patient times in clinic per infusion were 82.2 ± 40.9 and 183.9 ± 34.8 min, respectively. The study suggested that the corresponding time required for a 12-week treatment of nab-paclitaxel q3w, nab-paclitaxel qw, and paclitaxel qw would be 2. 8, 8.4, and 20.9 h, respectively. The corresponding time for patient hospital visits would be 5.5, 16.4, and 36.8 h, respectively. Thus, nabpaclitaxel may require less time for drug administration, potentially reducing cost. Table 3 lists selected ongoing trials of nab-paclitaxel regimens for the treatment of MBC. There are several ongoing trials evaluating combinations of nab-paclitaxel with HER2-targeted agents for the treatment of metastatic HER2-positive breast cancer. The single-arm phase III PERUSE trial (NCT01572038; planned N = 1500) is evaluating the safety of first-line pertuzumab combined with trastuzumab and a taxane of the investigator's choice, as its primary endpoint, in patients with metastatic or locally recurrent HER2-positive breast cancer [https://clinicaltrials.gov/ct2/show/NCT01572038]. Secondary endpoints include PFS, OS, ORR, and quality of life. Interim safety results indicated that grade ≥3 adverse events occurred in 53.4% of patients in the docetaxel group (n = 320), 41.1% in the paclitaxel group (n = 331), and 26.7% in the nab-paclitaxel group (n = 45) [43]. The most common grade ≥3 toxicities included neutropenia (approximately 12, 6, and 2% in docetaxel, paclitaxel, and nab-paclitaxel groups, respectively) and diarrhea (approximately 9, 6, and 6%). The phase IIIb SAPPHIRE trial (NCT02019277; N = 50) is evaluating the safety and efficacy of trastuzumab combined with intravenous pertuzumab and a taxane of investigator's choice in patients with metastatic HER2-positive breast cancer [44,45]. Interim results showed that 50 patients had been enrolled, with the majority (72%) receiving nab-paclitaxel as the taxane of choice. Grade ≥3 adverse events were reported in 52% of patients, although toxicities were not categorized according to type of taxane received. In addition, the phase I/II STELA trial (NCT02073916; planned N = 45) will combine trastuzumab emtansine (T-DM1), nab-paclitaxel, and lapatinib to treat metastatic HER2-positive MBC (https://clinicaltrials.gov/ct2/show/NCT02073916, [46]).  A number of ongoing trials are also evaluating nabpaclitaxel in HER2-negative MBC. The phase II/III tnAcity trial (NCT01881230) is comparing nab-paclitaxel plus gemcitabine with nab-paclitaxel plus carboplatin as firstline treatment for metastatic TNBC. The phase II portion (N = 240) has 3 arms: nab-paclitaxel 125 mg/m 2 plus gemcitabine 1000 mg/m 2 on days 1 and 8 q3w, nab-paclitaxel 125 mg/m 2 plus carboplatin area under the curve (AUC) of 2 days 1 and 8 q3w, and gemcitabine 1000 mg/m 2 plus carboplatin AUC of 2 days 1 and 8 q3w (https://clinicaltrials.gov/ct2/show/NCT01881230, [47]). In the phase III portion (N = 550), the nab-paclitaxel plus gemcitabine or nab-paclitaxel plus carboplatin arm will be selected based on phase II trial results and compared with gemcitabine 1000 mg/m 2 plus carboplatin AUC of 2 q3w. The phase II SNAP trial (NCT01746225; planned N = 258) will evaluate different schedules of first-line nab-paclitaxel for the treatment of HER2-negative MBC (https://clinicaltrials.gov/ct2/show/NCT01746225, [48]). All patients will receive induction nab-paclitaxel 125 mg/m 2 qw 3/4 followed by nab-paclitaxel 150 mg/m 2 on days 1 and 15 of a 28-day cycle, 100 mg/m 2 qw 3/4, or 75 mg/m 2 qw. PFS will be assessed as the primary endpoint. An ongoing phase I/ II study (NCT01938833; planned N = 47) is evaluating the combination of nab-paclitaxel plus the histone deacetylase inhibitor romidepsin in recurrent or metastatic HER2-negative inflammatory breast cancer [https://clinicaltrials.gov/ct2/show/NCT01938833]. Results from the phase I portion (n = 9) demonstrated that the regimen was well tolerated and resulted in an ORR of 33%, including 1 complete response [49].

Discussion
Recent clinical data indicate that nab-paclitaxel is effective and safe across all stages of breast cancer. The results from trials in the neoadjuvant setting for early-stage TNBC or HER2-positive breast cancer were particularly encouraging. In TNBC, nab-paclitaxel monotherapy or in combination with other agents resulted in pCR rates ranging from 10.5 to 62%. In the phase III neoadjuvant GeparSepto trial, the largest difference in pCR was identified for patients with TNBC (nab-paclitaxel, 48.2% vs paclitaxel, 26.3%; P < 0.001), supporting the clinical benefit of nab-paclitaxel in early-stage TNBC [11]. The unmet need for the treatment of TNBC lends greater importance to these findings. Patients with early-stage HER2positive breast cancer also benefited from nab-paclitaxel treatment. Neoadjuvant nab-paclitaxel combined with trastuzumab and carboplatin, anthracycline, or vinorelbine demonstrated pCR rates in the breast and lymph nodes ranging from 45 to 49%, which is comparable to those observed for other current neoadjuvant therapies [50]. In addition, neoadjuvant nab-paclitaxel resulted in breast-conserving surgery in 71 to 77.5% of patients with early-stage breast cancer.
In the phase III CALGB 40502 study, patients with MBC treated with first-line nab-paclitaxel plus bevacizumab achieved a median PFS of approximately 9 months and a median OS of 23.5 months [30]. For reference, patients with MBC who received nab-paclitaxel 260 mg/m 2 q3w as first-or later-line therapy in a phase III trial demonstrated a median time to tumor progression of 5.3 months and a median OS of 15.0 months [5]. Neither PFS nor OS for the nab-paclitaxel plus bevacizumab arm of the CALGB 40502 trial was significantly different from that of the paclitaxel plus bevacizumab arm [30]. The 150 mg/m 2 dose of nab-paclitaxel was not optimal, with a higher percentage of patients in the nab-paclitaxel group developing hematologic and nonhematologic toxicities. Several ongoing trials are evaluating the potential clinical benefit of nab-paclitaxel in patients with MBC, particularly the HER2-positive and TNBC subpopulations.
AC doxorubicin + cyclophosphamide, AUC area under the curve, BC breast cancer, bev bevacizumab, carbo carboplatin, ddAC dose-dense AC, DLT dose-limiting toxicity, EC epirubicin + cyclophosphamide, FEC fluorouracil, epirubicin, and cyclophosphamide, gem gemcitabine, HER2 human epidermal growth factor receptor 2, iDFS invasive disease-free survival, is in situ, LABC locally advanced breast cancer, MBC metastatic breast cancer, MTD maximum tolerated dose, nab-P nab-paclitaxel, ORR overall response rate, pCR pathologic complete response, PD-L1 programmed death-ligand 1, PFS progression-free survival, PI3K phosphoinositide 3-kinase, q2w every 2 weeks, q3w every 3 weeks, qd daily, qw weekly, qw 3/4 first 3 of 4 weeks, R resection margin, RECIST Response Evaluation Criteria In Solid Tumors, RP2D recommended phase 2 dose, T primary tumor, TBD to be determined, T-DM1 trastuzumab emtansine, TNBC triple-negative breast cancer, trastuz trastuzumab, yp postneoadjuvant therapy a Pertains to MBC arms only Interim analyses from some of these trials have demonstrated promising results; once final, the findings from these trials will provide further insights into the role of nab-paclitaxel for the treatment of breast cancer across treatment settings and patient subsets. Recent efforts to maintain efficacy while limiting toxicity have focused on the optimization of nab-paclitaxel schedule and dose. The safety profiles of nab-paclitaxel-based regimens in the recent studies included in this review were consistent with those in past studies, including the registrational phase III trial. The most common grade 3/4 hematologic and nonhematologic adverse events associated with nab-paclitaxel were neutropenia and peripheral neuropathy, respectively. The majority of recent studies have examined weekly dosing, likely because an accumulation of data in the metastatic setting suggests an advantage over every-3-week dosing in balancing efficacy and tolerability. In addition, safety and treatment-exposure results of 2 large trials (GeparSepto in the neoadjuvant setting and CALGB 40502 in the metastatic setting) have suggested that nab-paclitaxel may be more feasible at a starting dose of 125 mg/m 2 compared with 150 mg/m 2 [11,30].

Future of nab-paclitaxel in breast cancer: nab-paclitaxel and immune therapy
In addition to the ongoing trials discussed above, there is interest in combining nab-paclitaxel with immuno-oncology agents (Table 3). Chemotherapy-induced cytotoxicity has been shown to activate the immune response and to release tumor antigens from cancer cells [51,52]. Preclinical data from mouse models of multiple solid tumor types suggested potential synergy between chemotherapy and immune checkpoint inhibitors [53,54]. Recent clinical data indicated that combining nab-paclitaxel with checkpoint inhibitors may be safe and effective in MBC. A phase Ib study of atezolizumab, a programmed death ligand 1 (PD-L1) inhibitor, combined with nab-paclitaxel qw (NCT01633970) demonstrated activity in 24 efficacyevaluable patients with metastatic TNBC (ORR 70.8%; stable disease in 20.8%) [55]. Five patients (16%) discontinued nab-paclitaxel due to toxicity (3 for peripheral neuropathy [1 each for grades 1, 2, and 3] and 1 each for fatigue and asthenia [both grade 2]). nab-Paclitaxel plus atezolizumab is currently being compared with nab-paclitaxel plus placebo as a first-line treatment for metastatic TNBC in the randomized phase III IMpassion130 trial (NCT02425891; Table 3) (https://clinicaltrials.gov/ct2/ show/NCT02425891, [56]). The PD-1 inhibitor nivolumab is also being evaluated in combination with nab-paclitaxel in MBC in an ongoing phase I trial (NCT02309177; planned N = 138) [https://clinicaltrials.gov/ct2/show/ NCT02309177]. The combination of atezolizumab and nab-paclitaxel is also being evaluated as a neoadjuvant regimen for the treatment of early-stage TNBC in an ongoing phase II trial (NCT02530489; planned N = 37) [https://clinicaltrials.gov/ct2/show/NCT02530489] and a phase III trial (NeoTRIPaPDL1; NCT02620280 [nab-paclitaxel plus carboplatin ± atezolizumab]; planned N = 272) [https://clinicaltrials.gov/ct2/show/NCT02620280]. Similarly, the PD-L1 inhibitor durvalumab combined with nab-paclitaxel is being examined as neoadjuvant therapy for early-stage TNBC in an ongoing phase I/II trial (NCT02489448; planned N = 61) [https://clinicaltrials. gov/ct2/show/NCT02489448]. Results from these trials will provide further rationale for combining nab-paclitaxel with immune therapies as an exciting new treatment approach for early-stage or metastatic breast cancer.

Conclusions
In addition to demonstrated efficacy in the already established setting of MBC, nab-paclitaxel appears to be an effective and well-tolerated neoadjuvant therapy for patients with early-stage breast cancer, particularly the HER2-positive and TNBC subgroups. Ongoing trials are evaluating nab-paclitaxel in all stages and subtypes of breast cancer. One anticipated future role of nab-paclitaxel is as a backbone chemotherapy, and ongoing trials of nab-paclitaxel combined with immune checkpoint inhibitors are particularly exciting, as these may provide more effective treatment regimens for early-stage and metastatic breast cancer.