Tucatinib in patients with HER2-positive advanced/metastatic breast cancer: A systematic literature review of real-world evidence

Abstract

Background

Recent years have seen rapid advances in human epidermal growth factor receptor 2 (HER2)-directed therapies for breast cancer. The shifting therapeutic landscape has presented patients and clinicians with a plethora of effective treatment options, but optimal treatment sequencing can be challenging. Tucatinib is a HER2-selective tyrosine kinase inhibitor that, when given in combination with trastuzumab and capecitabine, has demonstrated survival benefits for patients with HER2-positive (HER2+) metastatic breast cancer (MBC) pretreated with trastuzumab, pertuzumab, and trastuzumab-emtansine. Real-world data, which can supplement knowledge gained from clinical trials, has begun to emerge for tucatinib, but a comprehensive review of these studies has yet to be conducted.

Methods

A systematic literature review was conducted to identify studies published between January 2020 and January 2025 that evaluated the effectiveness and safety of tucatinib in routine clinical practice. Studies were eligible for inclusion if they utilized real-world data, involved patients with HER2+ MBC treated with tucatinib in any line of metastatic therapy, and assessed tucatinib's effectiveness, health-related quality of life (QoL), patient-reported outcomes (PROs), or safety.

Results

Of 468 unique references identified, 12 publications met the inclusion criteria, 3 of which were manuscripts and 9 of which were congress abstracts. Included studies were heterogeneous in sample size, the lines of therapy assessed, and outcomes reported. Tucatinib-based treatment outcomes in the post-trastuzumab deruxtecan (T-DXd) setting and for patients with brain metastases were also reported. Despite variability among studies, outcomes were broadly consistent with clinical trial findings.

Conclusion

Overall, the available real-world evidence supports the clinical effectiveness of tucatinib-based therapies in HER2+ MBC, including heavily pretreated populations, patients with prior exposure to T-DXd, and those with brain metastases. However, gaps remain in the real-world data regarding the safety profile of tucatinib and its impact on health-related quality of life in routine clinical practice.

Graphical abstract

Highlights

• •There have been multiple recent advances in HER2-directed therapy for patients with HER2+ MBC.
• •In RCTs, tucatinib, a HER2-selective TKI, showed benefit in patients pretreated with trastuzumab, pertuzumab and a trastuzumab-based ADC.
• •This systematic literature review summarizes real-world evidence for tucatinib effectiveness and safety.
• •Tucatinib-based therapy was effective in patients with pretreated HER2+ MBC, including post T-DXd and patients with brain metastases.
• •Additional, robust real-world evidence is needed, especially for safety and HRQoL.

1. Introduction

Historically, breast cancers overexpressing human epidermal growth factor receptor 2 (HER2), a cell-surface receptor tyrosine kinase, were associated with aggressive disease and poor prognosis [1,2]. Although HER2-positive tumors account for only 13% of all breast cancers, they represent over 25% of metastatic breast cancer (MBC) cases, reflecting their aggressive biology [3,4]. Furthermore, up to 50% of patients develop brain metastases [5], which are associated with poor outcomes [6,7]. Despite these challenging characteristics, current 5-year survival percentages for patients with HER2+ MBC are higher than those for other subtypes, at 47 and 41% for cases with and without hormone receptor (HR) expression, respectively [8]. Improved survival outcomes for patients with HER2+ are attributable in large part to continual advancement in HER2-directed therapies over the past 2 decades [9].

HER2+ MBC therapy was revolutionized by the development of the anti-HER2 monoclonal antibody trastuzumab, and later pertuzumab, which target the HER2 protein, preventing hetero-dimerization with HER-family proteins, inhibiting oncogenic downstream signaling, and promoting antibody-dependent cytotoxicity [1,2,[10], [11], [12]]. HER2-targeted therapies have continued to evolve with the introduction of HER2-directed antibody-drug conjugates (ADCs) that deliver cytotoxic payloads with precision to specific cell types [13]. Stably linking the microtubule-inhibitory agent DM1 to trastuzumab, trastuzumab emtansine (T-DM1) provided progression-free survival (PFS) and overall survival (OS) benefits versus lapatinib plus capecitabine for patients with HER2+ MBC who had progressed on trastuzumab plus a taxane [14]. More recently, trastuzumab deruxtecan (T-DXd) has been developed by conjugating a topoisomerase I inhibitor payload to trastuzumab via a cleavable linker [15]. In second-line and later settings, T-DXd had superior clinical outcomes to T-DM1 (PFS: HR 0.28, P < 0.001; OS: HR 0.64, P = 0.0037) or physician's choice of treatment (PFS: HR 0.36, P < 0.0001; OS: HR 0.66, P = 0.0021) [[16], [17], [18], [19]]. Current guidelines recommend trastuzumab, pertuzumab, plus a taxane (THP) in first-line (1L) therapy and T-DXd as the preferred option in 2L [[20], [21], [22]]. More recently, PFS gains in the DESTINY-Breast09 trial have led to US FDA approval of T-DXd combined with pertuzumab in 1L as well [23,24].

In parallel with monoclonal antibody-based therapies, the development of small molecule tyrosine kinase inhibitors (TKIs; i.e., lapatinib, neratinib, pyrotinib, and tucatinib) has shown clinically relevant activity for patients with HER2+ MBC [[25], [26], [27], [28], [29]]. Small molecule TKIs are cell membrane-permeable and are effective in blocking phosphorylation of the C-terminus of the HER2 protein, thus suppressing downstream signaling and cell proliferation [30]. Tucatinib is highly specific for HER2 because it spares other HER-family members, and has therefore potential for lower epidermal growth-factor receptor-related dermatologic and gastrointestinal toxicities [31]. In a heavily pretreated population of patients with disease progression on trastuzumab, pertuzumab, and T-DM1, the triplet of tucatinib, trastuzumab, and capecitabine demonstrated extended PFS and OS versus placebo, trastuzumab and capecitabine (PFS: median 7.8 versus 5.6 months, HR 0.54, P = 0.001; OS: median 24.7 versus 19.2 months, HR 0.73, P = 0.004), even for patients with brain metastases including stable and active disease (intracranial central nervous system-PFS: median 9.9 versus 4.2 months; HR 0.39, P < 0.001; OS: median 21.6 versus 12.5 months; HR 0.60, P < 0.001) [28,32,33].

While recent expansion in the number of effective, well-tolerated HER2+ MBC therapy options has been a boon for patients, it has introduced challenges related to optimal therapeutic sequencing [34]. Due to feasibility and cost constraints, it is highly unlikely that future randomized controlled trials (RCTs) will be conducted to directly compare the many combinations and sequences of HER2-targeted treatments that are now available. Despite being limited by the lack of randomization of patients to treatment groups, the value of real-world evidence (RWE) is being increasingly recognized and utilized in informing clinical and policy decisions, including by regulatory agencies and guidelines [[35], [36], [37], [38]]; RWE may provide supplementary decision-making guidance to clinicians in situations where RCT data are unavailable. Real-world studies using patient data from routine clinical practice can yield insights into treatment sequencing outcomes, and safety for post-approval therapies in patients with a broader range of demographic and disease characteristics [34,39]. With the emergence of T-DXd as a new 1L option, understanding subsequent treatment outcomes is an important line of inquiry.

This systematic literature review aims to provide a comprehensive overview of existing RWE for the clinical effectiveness and safety of tucatinib. Although the main objective of this manuscript is to systematically summarize and evaluate all tucatinib-related RWE, recent developments in the field have led us to devote additional attention to outcomes for patients with disease progression on T-DXd. Gaps in knowledge regarding tucatinib effectiveness and safety are identified and discussed, along with considerations for future inquiries.

2. Materials and methods

2.1. Data sources and searches

This systematic review was conducted in accordance with the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) collaboration [40] and has been registered in the International Prospective Register of Systematic Reviews (PROSPERO: CRD420251005401). A comprehensive literature search was conducted using PubMed, EMBASE, and Cochrane Library. Search dates ranged from January 1, 2020, to January 15, 2025. Gray literature searches in Google Scholar and targeted searches of relevant congress proceedings (listed in Supplementary Material) covering the previous 2-year period (2023–2024) were also conducted. The following keywords were used in the searches: breast cancer, breast carcinoma, tucatinib, and Tukysa (see Supplementary Table 1 for full search strings).

2.2. Study selection and data extraction

Two researchers (MKR and KMW) independently screened the abstracts of all identified publications for inclusion and reviewed full texts of articles from the title/abstract screening for eligibility. Reasons for exclusion were recorded; any inconsistencies between these two researchers were resolved by consensus with a third researcher (RCB). Data from eligible studies, including abstracts, full-text manuscripts, and posters and oral presentations from congress proceedings, were independently extracted by MKR and KMW and verified by RCB and CR.

2.3. Eligibility criteria, outcomes and data analysis

References were eligible for inclusion if they utilized real-world data, included a patient population that received tucatinib in any metastatic line, and evaluated tucatinib effectiveness and/or health-related quality of life (HRQoL), patient-reported outcomes (PROs), or safety in HER2+ MBC. All data were summarized descriptively. Details on eligibility criteria, real-world study definition, study selection and data extraction as well as study quality evaluation are provided in the Supplementary Material.

2.4. Study quality evaluation

Study quality was assessed using the Newcastle-Ottawa Scale (NOS) [41] and the International Society of Pharmacoeconomics and Outcomes Research (ISPOR)–Academy of Managed Care Pharmacy (AMCP)–National Pharmaceutical Council (NPC) questionnaire [42]. Additional details are provided in the Supplementary Material.

3. Results

3.1. Search results

The database search yielded 678 references, with 21 additional references identifed through other sources (Fig. 1). After removal of duplicates (n = 231), 468 references remained. During title and abstract screening, 450 references were excluded; the most frequent reasons for exclusion were ineligible publication type (e.g., a review or trial-in-progress; n = 183) and ineligible study design (e.g., an interventional study; n = 156) (for details see Fig. 1 and Supplementary Table 2). A total of 12 references (three full-text articles, nine congress abstracts) were ultimately included in this systematic literature review. Congress publications (posters or oral presentations) were obtained for eight of the nine abstracts and additionally used for data extraction. Although a 10th congress abstract was identified during data extraction, the presentation used the same data set as the full-text manuscript [43,44]. As such, the unique data reported in the poster was included as a supplement to the full text article.

Fig. 1.

PRISMA flowchart. ^aOnly publications that were primary research articles or abstracts were included; reviews, guidelines, and trials-in-progress were excluded.

3.2. Study characteristics

Table 1 summarizes the characteristics of the 12 included RWE studies. All 12 studies included patients with HER2+ advanced/MBC, were retrospective in design and evaluated tucatinib in 2L or later [[44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55]]. One study also reported outcomes for patients treated with tucatinib in 1L (n = 23) [52]. Most studies were conducted in the United States (n = 10); the other two studies were conducted in France [44] and Australia [48]. Real-world data sources included the Flatiron Health Research database (n = 4) [45,47,52,54], the Komodo Health Sentinel database (n = 1) [49], the Merative MarketScan database (n = 1) [55], French cancer centers (n = 1) [44], and single institutions (n = 5) [46,48,50,51,53]. Although the four studies that used the Flatiron Health Research database had different time frames and sample sizes, there is potential for some population overlap. Similarly, such an overlap is likely present in the single center studies reported by Kathri and colleagues [46,51]. Across eight studies with data available, median follow-up durations ranged from 8.2 to 24.7 months for the overall populations (Table 1).

Table 1.

Characteristics of included RWE studies (N = 12).

First author, year of publication (Congress, if applicable)	Location	Publication type	Estimated study duration, mo (median follow-up)	Study inclusion criteria	Patients treated with tucatinib, N (% receiving TTC)	Treatment groups analyzed
Kaufman PA, 2023aa [45] (SABCS 2022)	US	abstract/poster	∼67 (10)	Pts receiving tucatinib outside of a clinical trial setting	183 (74%)	Overall 1L–4L+ Post T-DXd
Khatri VM, 2024 [46] (American Radium Society 2024)	US	abstract/oral presentation	∼36 (24.7)	Pts received TTC plus stereotactic radiosurgery	34 (not reported)	Overall Age ≤50 years Age >50 years
Anders C, 2024a [47] (EBCC 2024)	US	abstract/poster	72 (11.5)	Pts receiving TTC in 2L + or post T-DXd	89 (100%)	Overall 3L Post T-DXd
Blackley E, 2024 [48] (SABCS 2023)	Australia	abstractb	not reported	Pts treated with TTC with tumor and plasma samples	11 (100%)	Overall CNS disease
Anders C, 2023a [49] (ASCO 2023)	US	abstract/poster	∼69 (9)	Pts diagnosed with MBC who initiated treatment with tucatinib	528 (71%)	Overall 1L–4L+ Post T-DXd
Moore H, 2024 [50] (SABCS 2023)	US	abstract/poster	∼14 (>6c)	Pts receiving tucatinib + trastuzumab ± capecitabine	23 (87%)	Tucatinib + trastuzumab ± capecitabine 2L+3L Post T-DXd Presence/type of brain metastases
Khatri VM, 2023 [51]	US	manuscript	34 (20.8)	Pts received TTC plus stereotactic radiosurgery	22 (100%)	Overall Age ≤50 years
Frenel JS, 2024 [44]	France	manuscript	>28 (11.6)	Pts received TTC after prior T-DXd	101 (100%)	Overall Reason for T-DXd discontinuation Duration of previous T-DXd Tucatinib in 3–4L Tucatinib in 5L+ Timing after T-DXd Presence/type of brain metastases
Kaufman PA, 2023ba [52]	US	manuscript	72 (12)	Pts received: tucatinib in any line; TTC in 2L or 3L after 1+ HER2-directed therapy; tucatinib immediately following T-DXd in any line	216 (74%)	Overall 1L–4L Approved TTC in any line Approved TTC in 2L, 3L, or 4L Brain metastases Post T-DXd
Shagisultanova E, 2024 [53] (ESMO BC 2024)	US	abstract/poster	not reported	Pts received sequential TKIs	23 (not reported)	Tucatinib follow by tucatinib Tucatinib followed by other TKI Other TKI followed by tucatinib
Tarantino P, 2024a,d [54] (SABCS 2024)	US	abstract/poster	46 (not reported)	Pts received TTC immediately following T-DXd	95 (100%)	Overall By HER2 and HR status Prior T-DXd overall Prior T-DXd by HER2- and HR-status Reason for T-DXd discontinuation
Anders C, 2023b [55] (AMCP 2023)	US	abstract/poster	56 (8.2)	Not specified	111 (80%)	Overall 1L–4L+

1L, first line; 2L, second line; 3L, third line; 4L+, fourth line or higher; 5L+, fifth line or higher; CNS, central nervous system; HER2, human epidermal growth factor receptor 2; HR, hormone receptor; MBC, metastatic breast cancer; mets, metastases; mo, months; pts, patients; TKI, tyrosine kinase inhibitor; T-DXd, trastuzumab deruxtecan; TTC, tucatinib + trastuzumab + capecitabine; US, United States.

^aPublications based on overlapping data cuts from the Flatiron Health Database.

^bCongress publication requested from primary author but not received.

^cFollow-up was at least 6 months for all patients, but median follow-up was not reported.

^dThe manuscript describing these data [56] was published after our literature search date range.

3.3. Quality assessment

According to the NOS, all three included manuscripts were of moderate quality (5–6 points) [44,51,52]. Based upon the ISPOR–AMCP–NPC questionnaire, one manuscript scored 12 [51], whereas the remaining two manuscripts scored 17 [44] and 21 [52].

3.4. Patient characteristics

Across the 12 RWE studies, overall 1436 patients (range: 11–528) received tucatinib for the treatment of HER2+ MBC. Eleven studies reported patient age with medians ranging from 49 to 57 years (minimum: 25; maximum: 85). Baseline disease characteristics varied by study, and reporting was limited in some publications. Eleven studies defined the patient population as having MBC; the other study included patients with locally advanced disease but did not specify the number of patients with stage III disease [50]. Time from primary diagnosis to tucatinib treatment initiation was not reported in any study. Eleven studies provided information about brain metastases at treatment start with frequencies ranging from 39 to 100%; three of these studies reported whether brain metastases were active or stable (Supplementary Table 3). Six studies reported other metastases at study start (i.e., liver [39–52% of patients], lung [40–53%], or bone [52–69%]). Six studies reported the frequency of de novo MBC at diagnosis, which ranged from 30 to 70% [44,45,49,52,53,55], HR status was reported in four studies, with positivity rates ranging from 45 to 74% [44,46,51,53]. One study reported median time from MBC diagnosis to treatment initiation (33 months) [55]. Disease-free interval (DFI) was reported in one study; 11 and 51% of patients had a DFI of <24 and ≥ 24 months, respectively [54].

3.5. Treatment history

Nine studies reported on prior treatments, with the median number of prior lines ranging from 2 to 4 (minimum number of prior lines: 0; maximum number of prior lines: 15) (Table 2) [44,45,47,49,50,[52], [53], [54], [55]]. Six studies reported on prior treatment with trastuzumab (77–100% of patients) and pertuzumab (69–89%) [44,47,49,52,53,55], and four studies reported on prior TKI treament (17–33%) [44,47,50,52]. Seven studies reported the proportion of patients who received prior T-DM1 (39–91%) [44,47,49,50,52,53,55] and nine reported on prior T-DXd treatment (0–100%) [44,45,47,49,50,[52], [53], [54], [55]], with seven studies analyzing the use of tucatinib immediately after T-DXd with median T-DXd treatment durations ranging from 3.7 to 8.9 months [44,45,47,49,52,54,55]. Tucatinib was generally used in heavily pretreated settings (3L to 5L) in this context. Patient baseline characteristics in the post T-DXd subgroups were only partially reported, so corresponding subgroup analyses must be interpreted with care.

Table 2.

Treatment history (n = 9).

Study	N lines of treatment, median (range)	Prior trastuzumab, n (%)	Prior pertuzumab, n (%)	Prior TKIc,d, n (%)	Prior T-DM1, n (%)	Prior T-DXd, n (%)
Kaufman PA, 2023a [45]	2 (0–7)	–	–	–	–	29 (16)
Anders C, 2024 [47]	3 (2–10)	89 (100)	79 (89)	lapatinib/neratinib 25 (28)	78 (88)	34 (38)
Anders C, 2023a [49]	2 (IQR, 1–3)	425 (80)	378 (72)	–	231 (44)	68 (13)
Moore H, 2024 [50]	3 (0–6)	–	–	lapatinib: 2 (9) neratinib: 7 (30)	17 (74)	3 (13)
Frenel JS, 2024a [44]	4 (2–15)	100 (99)	82 (81)	33 (33)	94 (91)	101 (100)
Kaufman PA, 2023ba [52]	2 (0–10)	167 (77)	149 (69)	lapatinib/neratinib 37 (17)	109 (51)	41 (19)
Shagisultanova E, 2024b [53]	–	22 (96)	17 (74)	–	9 (39)	0
Tarantino P, 2024c [54]	4 (1–15)	–	–	–	–	95 (100)
Anders C, 2023b [55]	2 (0–7)	103 (93)	89 (80)	–	65 (59)	19 (17)

TKI, tyrosine kinase inhibitor; T-DM1, ado-trastuzumab emtansine; T-DXd, trastuzumab deruxtecan.

^aFull manuscript.

^bIncludes other TKIs besides tucatinib.

^cMixed treatments, not only tucatinib; all 95 patients receiving tucatinib did so immediately following T-DXd.

^dTKIs include lapatinib, neratinib, and pyrotinib.

3.6. Effectiveness outcomes

The included studies varied by the tucatinib combination therapies for which outcomes were assessed, but most patients received the registered tucatinib triplet (tucatinib, trastuzumab, and capecitibine [TTC]). Five studies evaluated patients who all received TTC [44,47,48,51,54], and two studies did not report the type of tucatinib-based regimen [46,53]; in the remaining 5 studies 71 to 87% of patients received TTC [45,49,50,52,55] (Table 1). Of the total 1436 patients across all studies receiving tucatinib, 1098 (76%) received TTC, 281 (20%) received tucatinib monotherapy or in combination with trastuzumab, capecitabine, or other agents, and 57 (4%) received a tucatinib-based regimen that was not specified. Only one study had comparator treatments [54]. Tucatinib effectiveness outcomes assessed varied among the included studies; the reported data are summarized in Table 3.

Table 3.

Effectiveness outcomes in included studies (N = 12).

Study	Group analyzed	Median rwPFS, mo (95%CI)	PFS rate, % (95%CI)	Median rwCNS PFS, mo (95%CI)	Median rwTTNT, mo (95%CI)	Median OS, mo (95%CI)	OS rate, % (95%CI)	rwDCR, %	rwTR, n (%)
Kaufman 2023a [45]	Overall (N = 183)				9.4 (6.8–11.9)	26.1 (20.2–NR)	6-mo: 86.1 (81.0–91.6) 9-mo: 79.8 (73.8–86.4) 12-mo: 74.2 (67.3–81.9)
	2L+3L any Tucatinib regimen (n = 94)				9.5 (6.8–13.6)	NR (NR–NR)	6-mo: 92.8 (87.4–98.5) 9-mo: 90.0 (83.6–96.9) 12-mo: 84.1 (75.6– 93.5)
	2L+3L Tucatinib triplet (n = 70)				9.5 (6.5–13.6)	NR (NR–NR)	6-mo: 93.5 (87.6–99.9) 9-mo: 91.7 (85.0–99.0) 12-mo: 83.5 (73.3–95.2)
	post T-DXdb (n = 29) median prior T-DXd duration (range): 3.7 (0–21) mo				8.1 (4.0–NR)	12.6 (11.9–NR)	12-mo: 61.8 (43.2–88.5)
Khatri 2024 [46]	Overall (N = 34)	18.9 (12.2–NR)	24-mo: 37%	10.2 (7.4–16.1) 12-mo: 38.8% 24-mo: 16%		30.0 (22–NR)	12-mo: 94.1% 24-mo: 69%	DIC: Median: 6.8 mo (95% CI, 4.6–10.7) 12-mo: 33% 24-mo: 17% LC: Median: 40.4 mo 12-mo: 96.8% 24-mo: 94%
Anders 2024 [47]	Overall (N = 89)				8.4 (6.2–11.8)	24.9 (15.6–NR)
	3L (n = 28)				9.5 (5.1–15.7)
	post T-DXdb (n = 30) median prior T-DXd duration (IQR): 5.9 (4.3–11.1) mo				6.4 (4.5–11.9)	12.6 (11.9–NR)
Blackley 2024 [48]	Overall (N = 11)	9.6 (4.0–NR)				36.2 (10.4–NR)
	CNS disease (n = 7)	13.5 (3.9–24.4)				36.2 (10.4–NR)
Anders 2023a [49]	Overall (N = 528)				10.7 (9.4–13.1)
	2L+3L (n = 318)				11.5 (9.6–14.4)
	post T-DXdb (n = 61) median prior T-DXd duration (IQR): 5.3 (2.1–8.3) mo				7.5 (5.0–13.3)
Moore 2024 [50]	Overall: Tucatinib + Trastuzumab ± Capecitabine (N = 23)								rwPR: 6 (26%) rwSD: 8 (35%) rwPD: 7 (30%) ND: 2 (9%)
	BM (stable) (n = 9)								rwPR: 4 (44%) rwSD: 3 (33%) rwPD: 1 (11%) ND: 1 (11%)
	BM (active) (n = 3)								rwPR: 1 (33%) rwSD: 1 (33%) rwPD: 1 (11%) ND: 0
	No BM (n = 11)								rwPR: 1 (9%) rwSD: 4 (36%) rwPD: 5 (45%) ND: 1 (9%)
	Prior neratinib (n = 7)								rwPR: 1 (14%) rwSD: 2 (29%) rwPD: 3 (43%) ND: 1 (14%)
	Prior lapatinib (n = 2)								rwPR: 1 (50%) rwSD: 0 rwPD: 1 (50%) ND: 0
	Prior T-DM1 (n = 17)								rwPR: 4 (24%) rwSD: 7 (41%) rwPD: 4 (24%) ND: 2 (12%)
	Prior T-DXd (n = 3)								rwPR: 0 rwSD: 0 rwPD: 2 (67%) ND: 1 (33%)
Khatri 2023a [51]	TTC + stereotactic radiosurgery (N = 22)	Systemic PFS: NR Overall PFS: NR	Systemic PFS: 12-mo: 86% 24-mo: 57% Overall PFS: 12-mo: 40% 24-mo: 32%	11.3 (4.9–NR) 12-mo: 44.9% 24-mo: 44.9%		21.2 (17.3–NR)	12-mo: 84% 24-mo: 50%	Local failure: 7% 12-mo LC: 94% 24-mo LC: 81% Any intracranial failure: 50% Distant failure: 59% 12-mo DIC: 39% 24-mo DIC: 26%
Frenel 2024a [44]	Overall (N = 101) median prior T-DXd duration (range): 8.9 (1.4–25.8) mo	4.7 (3.9–5.6)	6-mo: 33.1 (24.8–44.3) 12-mo: 11.9 (6.4–22.1)		5.2 (4.5–7.0)	13.4 (11.1–NR)	6-mo: 77.0 (69.0–86.0) 12-mo: 57.5 (47.2–70.1)	64%	rwCR: 2 (2.2) rwPR: 27 (30.3) rwSD:28 (31.5) rwPD: 32 (36.0) Missing: 12 rwORRc: 29 (32.6)
	BMs (n = 39)	5.0 (3.9–6.8)	6-mo: 32.4 (20.4–51.3) 12-mo: 11.9 (4.8–29.3)		5.7 (4.7–8.6)	12.9 (9.6–NR)	6-mo: 76.0 (63.4–91.1)
	BM (active) (n = 16)	4.7 (3.0–7.3)	6-mo: 25.0 (10.7–58.4)		5.6 (4.4–NR)	12.4 (8.3–NR)	6-mo: 81.2 (64.2–100.0)	Intracranial: 66.7%	Intracranial: rwCR: 1 (6.7) rwPR: 2 (13.3) rwSD: 7 (46.7) rwPD: 5 (33.3) Missing:1 rwORRc: 3 (20.0)
	BM (non-active) (n = 23)	5.2 (3.4–8.0)	6-mo: 37.9 (22.3–64.7)		5.7 (4.7–NR)	17.0 (13.4–NR)	6-mo: 72.4 (55.7–94.2)
	No BM (n = 62)	4.4 (3.5–6.0)	6-mo: 33.6 (23.2–48.8) 12-mo: 10.6 (4.2–27.0)		4.8 (3.9–6.3)	19.9 (10.7–NR)	6-mo: 77.7 (67.7–89.2)
	Tuc as 3-4L (n = 37)	4.4 (3.4–8.0)	6-mo: 33.8 (21.0–54.2)		4.8 (3.9–9.9)	12.9 (10.2–NR)	6-mo: 79.6 (67.2–94.3)
	Tuc as 5L+ (n = 64)	4.7 (3.7–5.7)	6-mo: 32.8 (22.8–47.3)		5.3 (4.7–7.0)	13.4 (10.7–NR)	6-mo: 75.3 (65.2–87.1)
	Duration of prior T-DXd <6 mo (n = 23)	4.7 (3.2–5.7)	6-mo: 17.4 (7.1–42.4)		5.5 (4.8–NR)	10.6 (5.5–NR)	6-mo: 59.8 (42.6–84.1)
	Duration of prior T-DXd 6–12 mo (n = 36)	3.9 (2.9–6.0)	6-mo: 23.1 (12.2–43.9)		4.0 (3.5–6.1)	NR	6-mo: 82.1 (70.1–96.2)
	Duration of prior T-DXd >12 mo (n = 23)	5.6 (4.4–NR)	6-mo: 47.8 (31.2–73.3)		7.2 (5.2–NR)	NR	6-mo: 81.7 (66.9–99.7)
	T-DXd discontinued due to toxicity (n = 18)	7.3 (3.0–NR)	6-mo: 61.1 (42.3–88.3) 12-mo: 14.3 (2.8–73.3)		7.8 (3.1–NR)	NR	6-mo: 81.7 (64.9–100.0)	75.0	rwCR: 2 (12.5) rwPR: 4 (25.0) rwSD: 6 (37.5) rwPD: 4 (25.0) Missing: 2 rwORRc: 6 (37.5)
	T-DXd discontinued due to progression (n = 82)	4.4 (3.9–5.5)	6-mo: 28.4 (19.9–40.6) 12-mo: 10.9 (5.4–22.1)		5.2 (4.5–6.3)	12.9 (11.1–NR)	6-mo: 75.7 (66.8–85.9)	62.5	rwCR: 0 rwPR: 23 (31.9) rwSD: 22 (30.6) rwPD: 27 (37.5) Missing: 10 rwORRc: 23 (31.9)
	Tucatinib immediately after T-DXd (n = 86)	5.0 (4.2–6.0)	6-mo: 36.1 (26.9–48.4) 12-mo: 14.2 (7.7–26.0)		5.5 (4.8–7.2)	13.4 (11.9–NR)	6-mo: 78.0 (69.5–87.6) 12-mo: 60.5 (49.6–73.9)
	Tucatinib not immediately after T-DXd (n = 15)	3.5 (2.8–NR)	6-mo: 16.7 (5.0–56.1)		3.8 (2.8–NR)	11.1 (8.3–NR)	6-mo: 72.0 (51.9–99.8)
Kaufman 2023ba [52]	Overall (N = 216)				8.7 (6.8–10.7)	26.6 (20.2–NR)	12-mo: 75.3 (69.3–81.9) 24-mo: 55.5 (47.3–65.0)
	1L (n = 23)				12.8 (9.1–NR)	15.5 (13.8–NR)
	2L (n = 69)				9.8 (6.5–18.3)	NR (20.0–NR)
	3L (n = 44)				9.4 (5.7–13.6)	NR
	4L+ (n = 80)				6.5 (5.1–8.8)	16.6 (12.2–NR)
	Approved TTC in any line (n = 144)				8.8 (7.1–11.8)	26.1 (18.8–NR)	12-mo: 75.9 (68.4–84.1) 24-mo: 54.7 (44.7–66.9)
	Approved TTC in 2L (n = 52)				11.0 (6.8–18.3)	21.3 (17.1–NR)
	Approved TTC in 3L (n = 31)				9.5 (5.4–15.7)	NR
	Approved TTC in 2L and 3L (n = 83)				9.8 (6.8–14.1)	NR	12-mo: 85.0 (76.7–94.2) 24-mo: 63.8 (50.5–80.7)
	Approved TTC in 4L+ (n = 61)				8.1 (6.2–11.9)	16.6 (12.2–NR)
	post T-DXdb (n = 35) Median rwTTD of T-DXd 4.6 mo (95%CI 3.7-10.3)				8.1 (4.5–11.5)	13.9 (12.2–NR)	12-mo: 69.2 (53.7–89.0)
Tarantino 2024 [54]	HER2+, prior T-DXd, TTCb (n = 95)	4.7
	HER2+, PD on prior T-DXd, TTC (n = 74)	4.5
	HER2+, non-PD on prior T-DXd, TTC (n = 21)	6.2
	HER2+, prior T-DXd, ET	overall (n = 57): 4.4 PD on prior T-DXd (n = 32): 3.2 non-PD on prior T-DXd (n = 25): 10.2
	HER2+, Prior T-DXd, CT + anti-HER2 mAb(s)	overall (n = 56): 5.0 PD on prior T-DXd (n = 44): 4.7 non-PD on prior T-DXd (n = 12): 14.7
	HER2+, prior T-DXd, anti-HER2 mAb(s) ± TKI	overall (n = 34): 7.3 PD on prior T-DXd (n = 16): 5.2 non-PD on prior T-DXd (n = 18): 9.3
	HER2+, prior T-DXd, T-DM1	overall (n = 27): 4.1 PD on prior T-DXd (n = 20): 3.4 non-PD on prior T-DXd (n = 7): 15.7
	HER2+, prior T-DXd, SG	overall (n = 16): 2.3 PD on prior T-DXd (n = 13): 2.3 non-PD on prior T-DXd: not reported
Shagisultanova 2023 [53]	Tuc as 1st TKI (systemic disease)								rwCR: 1 (7.7) rwPR: 4 (30.8) rwSD: 8 (61.5) rwPD: 0
	Tuc as 2nd TKI (systemic disease)								rwCR: 3 (20.0) rwPR: 2 (13.3) rwSD: 8 (53.3) rwPD: 2 (13.3)
	Tuc as 1st TKI (CNS disease)								rwCR: 1 (14.3) rwPR: 1 (14.3) rwSD: 5 (71.4) rwPD: 0
	Tuc as 2nd TKI (CNS disease)								rwCR: 0 rwPR: 2 (25.0) rwSD: 5 (62.5) rwPD: 1 (12.5)
Anders 2023b [55]	Overall (N = 111)				7.7 (5.8–NR)
	2L+3L (n = 58)				8.8 (5.5–NR)
	4L+ (n = 51)				7.7 (5.3–NR)
	Post T-DXdb (n = 18) median prior T-DXd duration (range): 4.6 (3.5–5.9) mo				6.2 (4.3–NR)

+, positive; –, negative; 1L, first line; 2L, second line; 3L, third line; 4L+, fourth line (or higher); 5L+, fifth line or higher; BM, brain metastases; CI, confidence interval; CNS, central nervous system; CR, complete response; DCR, disease control rate; DIC, distant intracranial control; LC, local control; HR, hormone receptor; HER2, human epidermal growth factor receptor 2; MBC, metastatic breast cancer; mets, metastases; mo, months; ND, not determined; NE, not evaluable; NR, not reached; OS, overall survival; PD, progressive disease; PFS, progression-free survival; PR, partial response; RR, response rate; rw, real world; SD, stable disease; SG, sacituzumab govitecan; TKI, tyrosine kinase inhibitor; T-DXd, trastuzumab deruxtecan; TR, tumor response; TTC, tucatinib + trastuzumab + capecitabine; TTNT, time to next treatment; US, United States.

^aFull manuscript.

^b“Immediately” after prior T-DXd. .

^cORR (CR + PR).

3.6.1. Real-world progression-free survival (rwPFS)

Five studies reported median rwPFS [44,46,48,51,54] with three also reporting rwPFS rates at 6, 12, and/or 24 months (Table 3) [44,46,51]. Median rwPFS for overall study populations ranged from 4.7 to 18.9 months; one study reported that rwPFS was not reached [51]. One study showed that HR status in the Cox proportional hazards model was not associated with rwPFS [51]. The wide variation in rwPFS may be explained by differences in the lines of therapy in which tucatinib was received and small sample sizes (i.e., less than 40) in three of the studies [46,48,51]. One study provided rwPFS outcomes for patients treated with various agents immediately following T-DXd (TTC: 4.7 months; sacituzumab govitecan [SG]: 2.3 months; T-DM1:4.1 months; chemotherapy [CT] + anti-HER2 mAb(s): 5 months; anti-HER2 mAb(s) ± TKI: 7.3 months) [54]. However, these data were descriptive in nature, some of the subgroups had small sample sizes, and patient characteristics were not provided.

Although eight studies evaluated the use of tucatinib-based regimens after T-DXd, only one of these evaluated rwPFS, whereas most of the other studies evaluated real-world time to next treatment (rwTTNT). This study reported an overall median rwPFS of 4.7 months, a median number of 4 prior treatment lines in the metastatic setting, and explicitly assessed outcomes in the post T-DXd setting [44]. It found numerical differences in rwPFS depending on the duration of prior T-DXd treatment (median rwPFS with <6, 6–12 and > 12 months of prior T-DXd duration: 4.7, 3.9 and 5.6 months respectively), reason for T-DXd discontinuation (median rwPFS with discontinuation of T-DXd due to toxicity or progression: 7.3 and 4.4 months respectively), and if tucatinib treatment was started immediately after T-DXd treatment or not (median rwPFS with and without use immediately post T-DXd: 5.0 and 3.5 months respectively) [44].

3.6.2. Real-world central nervous system progression-free survival (rwCNS PFS)

In the two reports that assessed rwCNS PFS [46,51], likely representing a continuous report of the same retrospective cohort study evaluating patients with active brain metastases who received either concurrent or sequential treatment with stereotactic radiosurgery and tucatinib, investigators reported median rwCNS PFS of 10.2 and 11.3 months with local failure defined in line with the Response Assessment in Neuro-Oncology for Brain Metastases (RANO-BM) criteria [46,51]. In these 2 reports, 12- and 24-month rwCNS PFS rates were also reported (Table 3). One of these studies showed that HR status by univariate analysis was not associated with intracranial failure [51].

3.6.3. Real-world time to next treatment (rwTTNT)

Six of the included studies reported rwTTNT across different subgroups [44,45,47,49,52,55], as a proxy for PFS [57,58]. Generally, in the overall populations rwTTNT ranged from 5.2 to 10.7 months. As with rwPFS, the results depended on the intensity of prior treatment and the subgroup sample sizes, which ranged from 18 to 86; both factors may have contributed to the observed variability. One study found that tucatinib was generally more effective in prolonging rwTTNT if used in earlier lines of treatment (median rwTTNT with TTC in 2L: 11.0 months; median rwTTNT with TTC in fourth line or higher [4L+]: 8.1 months) [52]. A cross-study comparison supports this trend with studies in earlier lines of therapy showing higher median rwTTNT values (7.7–10.7 months) [45,47,49,52,55] as compared with a study in a later line of therapy (5.2 months) [44].

All six studies also investigated rwTTNT of tucatinib-based therapy immediately after use of T-DXd and found median rwTTNT ranging from 5.5 to 8.1 months.

3.6.4. Real-world tumor response (rwTR)

Three studies reported on rwTR (Table 3) [44,50,53]. rwTR varied depending on prior treatments received and reasons for discontinuation. In generally extensively pretreated populations, real-world complete and partial response rates ranged from 0 to 23.1% and 15.4–30.8%, respectively. Real-world stable and progressive disease rates ranged from 31.5 to 61.5% and 0–36.0%, respectively. Although not explicitly reported by the studies, these tumor response percentages correspond to manually calculated disease control rates ([complete response + partial response + stable disease]/n; DCR) ranging from 56.4 to 100%.

In the two studies that assessed general rwTR in patients with brain metastases, at least two-thirds of the population demonstrated stable disease or better responses [44,53]. Two studies reported on intracranial real-world disease control rate (rwDCR), with a local control rate of 81–94% at 24 months (RANO-BM criteria; combination with stereotactic irradiation) [46,51] and intracranial complete and partial responses of 6.7 and 13.3% (RECIST-based criteria; no information on local interventions) [44].

In patients treated with prior T-DXd, those who discontinued T-DXd due to toxicity tended to have better rwTR (rwORR: 37.5%; complete response [CR]: 12.5%) than patients who discontinued due to progression (rwORR: 31.9%; CR: 0%) in one study [44].

3.6.5. Overall survival (OS)

Seven studies reported median OS, which ranged from 13.4 to 36.2 months in the overall study populations [[44], [45], [46], [47], [48],51,52]; the study reporting OS of 36.2 months included only 11 patients [48].

Four studies reported median OS in patients who received tucatinib-based therapy immediately following T-DXd [44,45,47,52]; median OS ranged from 12.6 to 13.9 months, and subgroup samples sizes ranged from 29 to 86 patients.

3.7. Safety

Only one study reported adverse events (AEs) [50]. Among the overall population treated with TT ± C (n = 23), 100% of patients experienced at least one AE, six patients (26%) had grade 3 or higher AEs, and three patients (13%) had grade 3/4 hepatotoxicity. The most common AEs reported were fatigue (70%; grade ≥3: 4%), palmar-plantar erythrodysesthesia (61%; grade ≥3: 0%), nausea (61%; grade ≥3: 9%), and anemia (61%; grade ≥3: 0%); 52 and 35% of patients experienced diarrhea and vomiting (any grade), respectively; with 1 patient (4%) having experienced grade ≥3 vomiting and no patients with grade ≥3 diarrhea. No serious AEs were reported.

Real-world time to treatment discontinuation (rwTTD) was reported in five studies with tucatinib in different lines of therapy, all of which included patient subgroups with prior T-DXd (Table 4) [45,47,49,52,55]. Overall, median rwTTD ranged from 5.9 to 8.5 months and varied across studies by treatment line and prior T-DXd use. Reasons for treatment discontinuation were reported in two studies, in which AEs or toxicity led to discontinuation in 9% [25,50]% [44] of patients, respectively. Another study reported that 75% of patients discontinued treatment due to progression [44]. Only one study reported on dose modifications (reductions, increases, interruptions, or delays); nine of 23 (39%) patients experienced a drug-drug interaction with tucatinib leading to treatment modification in 6 (26%) patients [50].

Table 4.

Time to treatment discontinuation and treatment persistence (n = 5).

Study	Group	rwTTD, median (range) [months]	Treatment persistence, %
			6 mo	9 mo	12 mo	18 mo	24 mo
Kaufman PA, 2023a [45]	tucatinib-based overall (N = 183)	8.1 (5.7–9.5)		53	42	30
	tucatinib-based 2L+3L (n = 94)	8.8 (6.3–11.3)		56	40	27
	TTC 2L+3L (n = 70)	8.6 (6.3–11.3)		54	40	24
	tucatinib-based post T-DXdb (n = 29)	8.1 (3.6–11.9)			69
Anders C, 2024 [47]	TTC (N = 89)	5.9 (5.0–9.4)	53		44	26	15
	3L TTC (n = 28)	9.4 (5–15.6)
	TTC post T-DXdb (n = 30)	6.4 (3.6–11.6)	63		33
Anders C, 2023a [49]	tucatinib-based overall (N = 528)	8.5 (7.2–9.3)	65	55	46	35
	tucatinib-based 2L+3L (n = 318)	9.1 (7.7–9.9)	67	57	49	37
	tucatinib-based post T-DXdb (n = 61)	7.3 (3.2–9.5)	58	61	40	NA
Kaufman PA, 2023ba [52]	tucatinib-based overall (N = 216)	6.5 (5.4–8.8)	61		40	29	21
	tucatinib-based 1L (n = 23)	9.1 (5.1–NR)
	tucatinib-based 2L (n = 69)	8.2 (6.3–18.2)
	tucatinib-based 3L (n = 44)	9.4 (5.6–13.6)
	tucatinib-based 4L+ (n = 80)	4.8 (3.5–7.1)
	approved TTC in any line (n = 144)	8.1 (5.7–9.5)	61		40	27	11
	approved TTC in 2L (n = 52)	8.6 (6.3–18.8)
	approved TTC in 3L (n = 31)	9.4 (5.4–15.6)
	approved TTC in 2L+3L (n = 83)	9.4 (6.3–14.1)	68		42	26	14
	approved TTC in 4L+ (n = 80)	5.4 (4.2–8.8)
	tucatinib-based post T-DXdb (n = 35)	6.4 (3.5–9.4)
Anders C, 2023b [55]	tucatinib-based overall (N = 111)	6.2 (5.0–10.7)	50	47	50
	tucatinib -based 2L+3L (n = 58)	7.4 (5.0–NR)	55	52	53
	tucatinib-based 4L+ (n = 51)	6.1 (4.2–NE)
	tucatinib-based post T-DXdb (n = 86)	5.5 (4.2–NE)

1L, first line; 2L, second line; 3L, third line; 4L(+), fourth line (or higher); NE, not evaluable; NR, not reported; rwTTD, real-world time to treatment discontinuation; T-DXd, trastuzumab deruxtecan; TTC, tucatinib + trastuzumab + capecitabine.

^aFull manuscript.

^b“Immediately” after prior T-DXd.

3.8. Health-related quality of life (QoL)

No studies were identified that assessed real-world QoL or PROs.

4. Discussion

In recent years, a series of advances have reshaped the HER2+ MBC therapeutic landscape, substantially improving survival outcomes, and with some patients who achieve complete responses in the 1L setting being potentially cured [9]. Nevertheless, clear treatment sequencing guidance for patients who have progressed on or are intolerant to standard-of-care options, especially in later treatment lines, remains an unmet medical need [34]. This is especially true after introduction of T-DXd into the current treatment algorithm, with a lack of data from RCTs describing outcomes for patients following this ADC treatment. Because of the rapid pace of change in the HER2+ MBC treatment space, RCTs comparing different sequences of existing treatment options are unlikely to be conducted. Therefore, this systematic literature review sought to help fill this knowledge gap by identifying RWE studies that have evaluated tucatinib effectiveness and safety in routine clinical practice in general, as well as after T-DXd specifically.

Our literature search identified a modest number of studies (N = 12) that described real-world clinical outcomes for patients with HER2+ MBC who received tucatinib across a wide range of treatment lines (range of median prior lines: 2–4). Clinical benefit with tucatinib was observed in terms of rwPFS, rwTTNT, rwTR, and OS, although there was substantial variability among studies, possibly attributable to differences in patient and pretreatment characteristics and/or small sample sizes in some cases. Overall, the data summarized herein indicate that tucatinib, in combination with trastuzumab and capecitabine, is an effective treatment option for patients with pretreated HER2+ MBC. However, key evidence gaps remain, including more robust real-world safety, patient-reported outcomes, and treatment sequencing data.

Caution should be exercised in direct comparison of real-world with clinical trial data due to lack of randomization and differences in patient population age, demographics, overall health status, metastases patterns, and prior treatment patterns. However, it is important to place the data reported here in the context of findings from HER2CLIMB, the pivotal trial evaluating TTC after use of trastuzumab, pertuzumab and T-DM1 [28]. The median rwPFS reported in the included retrospective studies ranged from 4.7 to 18.9 months with corresponding median OS ranging from 13.4 to 36.2 months with tucatinib treatment in the 3–5L setting. These data are broadly in line with HER2CLIMB, which reported a median PFS of 7.8 months (95% confidence interval [CI], 7.5–9.6) and a median OS of 24.7 months (95% CI, 21.6–28.9) with the use of TTC in a heavily pretreated population [28,32]. Key challenges to interpreting rwPFS include that the treating physician determines progression, usually based on imaging, laboratory, and clinical findings, and without imposing a strict assessment schedule [59]. Additionally, there is a need for chart review to generate the rwPFS endpoint. Such an approach contrasts with a clinical trial, which uses a rigid assessment frequency and typically applies RECIST. Six of the included studies thus reported rwTTNT, a surrogate for rwPFS that is more easily abstracted from charts than rwPFS, and is influenced by both progression and toxicity [57,58]. The reported median rwTTNT was less variable than rwPFS and ranged from 5.2 to 10.7 months [44,45,47,49,52,55]. In the study that reported both rwPFS and rwTTNT, rwTTNT was generally in line with rwPFS [44], supporting the use of this parameter as a PFS-surrogate in real-world analyses. Taken together, and acknowledging the differences among studies in pre-treatment situations, the real-world progression and survival data are generally consistent with the clinical trial findings, supporting a benefit for patients seen in routine clinical practice.

Brain metastases are a common occurrence in HER2+ MBC [5], and associated with neurocognitive symptoms, reduced QoL and poor survival outcomes [60,61]. Remarkably, in each of the studies that assessed subgroups of patients with brain metastases, tumor response and survival outcomes were similar to those without brain metastases [44,48,50]. Although the consistency of this finding among the included studies supports the existing evidence from HER2CLIMB demonstrating a strong clinical benefit in this vulnerable population [62], there was a possibility that some patients may have received tucatinib in earlier lines of therapy because of the presence of brain metastases and those without brain metastases may have experienced progression on more previous lines of therapy, making cross-population comparisons difficult. Exploratory analysis from HER2CLIMB found that median OS for patients with HER2+ MBC and brain metastases was numerically shorter, but similar to the overall population (21.6 months versus 24.7 months, respectively) [32,33]. Taken as a whole, these findings, coupled with the putative permeability of tucatinib across the blood–brain barrier [63], support tucatinib as a viable option for patients with HER2+ MBC that has spread to the brain.

An outstanding question is where tucatinib fits best in the shifting HER2+ MBC treatment landscape, particularly within the context of the growing availability of ADCs in both early and MBC. THP has been the long-standing preferred 1L treatment [[20], [21], [22]], but in-progress clinical trials may soon provide additional options. Interim results from the DESTINY-Breast09 trial have shown a clinically meaningful PFS benefit with T-DXd plus pertuzumab in 1L compared with THP (median PFS: 40.7 versus 26.9 months; HR 0.56; P < 0.00001) [23]. Despite having a manageable safety profile, T-DXd plus pertuzumab has been associated with serious AEs, and treatment burden can lead to early discontinuation in a relevant number of patients (discontinuation rate not attributable to progression/death: 34.6%) [23]. To ensure long-term tolerance to HER2-directed therapy, it has been suggested that an induction–maintenance strategy could be implemented to exploit the efficacy of T-DXd during an induction period followed by maintenance with a second drug combination such as trastuzumab plus pertuzumab [64]. Such a strategy is being tested in the phase 2 DEMETHER trial, which is currently recruiting [65]. Maintenance strategies following induction with the current standard of care are also being pursued in several studies. In the phase 3 PATINA trial, palbociclib and endocrine therapy were deployed alongside trastuzumab and pertuzumab as maintenance for patients with HR+/HER2+ MBC who had not progressed on standard induction with a CT-based regimen[66]. The study showed a considerable PFS benefit of the combination treatment versus standard of care with trastuzumab, pertuzumab, and endocrine therapy (median PFS: 44.3 versus 29.1 months; HR 0.75, P = 0.02). Moreover, the phase 3 HER2CLIMB-05 trial has demonstrated a significant PFS benefit when adding tucatinib to trastuzumab and pertuzumab as maintenance following induction with CT in patients with HER2+ MBC regardless of HR status (median PFS: 24.9 versus 16.3 months [placebo]; HR 0.641; P < 0.0001) [67]. Notably, PROs from the HER2CLIMB trial indicated that HRQoL was preserved with the addition of tucatinib to trastuzumab plus capecitabine, and HRQoL was extended with tucatinib in patients with brain metastases [61]. Tucatinib may therefore represent a useful tool in balancing efficacy and QoL for some patients with HER2+ MBC.

Although PFS gains conferred with T-DXd are substantial, the vast majority of patients will still progress on T-DXd when given in the 2L [18]. Data on the effectiveness of subsequent treatments, especially with ADCs, are mixed. One study evaluated rwPFS for patients receiving various therapies immediately following T-DXd and found that T-DM1 and SG were associated with the shortest median rwPFS (4.1 and 2.3 months, respectively), but limited subgroup sample sizes, unknown patient characteristics, and the absence of statistics that mitigate treatment selection bias (e.g., propensity score matching) preclude conclusive interpretation of these data [54]. In contrast, DESTINY-Breast03 demonstrated T-DXd efficacy following T-DM1, two ADCs that share the same targeting antibody, but utilize different payloads and conjugation schemes [17,18]. It is possible that downregulation of surface HER2 protein may be a relevant factor for ADC cross-resistance in some situations [[68], [69], [70]]. For tucatinib following T-DXd, the capability of tucatinib to increase surface levels of HER2 may be useful in resensitizing the tumor to extracellular HER2 targeting [71]. Data on outcomes for an ADC-tucatinib-ADC treatment sequence could therefore be highly informative. In our systematic review, a total of nine studies reported on the use of tucatinib in the post T-DXd setting with tucatinib-based treatments being given primarily in the 3L to 5L [44,45,47,49,50,[52], [53], [54], [55]]. The overall duration of prior T-DXd was relatively short (3.7–8.9 months), potentially highlighting the fact that these data were generated shortly after T-DXd became available, thereby representing a sample of early progressing patients. In this context, the observed effectiveness outcomes for tucatinib in the post-T-DXd subgroups (rwTR: 32.6%; median rwPFS: 4.7–5.0 months; median rwTTNT: 5.5–8.1 months; median OS: 12.6–13.9 months) looks promising and suggest a persistent sensitivity to HER2-directed interventions, especially TKIs. Indeed, an updated analysis for one of the studies included in this SLR [43] showed persistent rwPFS and OS benefits with TTC across subgroups, further supporting the use of tucatinib-based treatments following T-DXd [72]. Nevertheless, these studies are subject to important limitations such as use in advanced treatment lines and limited subgroup sizes.

A key shortcoming in current real-world evidence for tucatinib is lack of robust safety data. rwTTD, an endpoint that may be affected by safety as well as effectiveness, ranged from 5.9 to 8.5 months and was thus relatively consistent among the five studies in which it was reported [45,47,49,52,55]. Moreover, it was generally in line with the 7.3 month median duration of exposure reported in HER2CLIMB [28]. Indeed, it is noteworthy that in four studies, treatment persistence at 18 months ranged from 26 to 35% [45,47,49,52]. These data suggest that real-world treatment management is permitting adequate treatment duration, even in advanced lines of therapy. Two studies reported 9–25% of patients discontinuing tucatinib-based therapy due to toxicity [44,50], so it is essential to better understand what AEs are being experienced by patients in the real-world setting in order to further optimize treatment management. Only one publication, a congress abstract, reported detailed information on AEs [50]. At least one AE was experienced by all patients in the study, with 26% of patients experiencing severe AEs. Commonly reported AEs included palmar-plantar erythrodysesthesia, fatigue, nausea, and anemia, all of which might negatively impact patient QoL. However, to our knowledge, no real-world data on HRQoL exists for tucatinib, which represents another important data gap. Therefore, future work elucidating real-world safety and HRQoL for patients taking HER2-directed therapies will be crucial for informed treatment selection.

This review is subject to a number of limitations. Tucatinib was first approved in 2020, so as expected, there are limited real-world studies published to date. The majority of studies included were reported as congress abstracts, with only 3 full-length articles. Generally, the included references had small sample sizes, especially for subgroups, and there was not a strong diversity of data sources. Four of the 12 studies used the Flatiron Health Database and although there were differences in the data cuts used, there is a possibility for patient overlap. Reflecting the complexity of the HER2+ MBC treatment landscape, tucatinib was administered in its full triplet combination, but also without one or both combination partners in some cases. Moreover, there was relevant variability in lines of therapy and prior therapies as well as a lack of information on subsequent therapies. Comparability among studies was hindered by patient population heterogeneity and differences in outcomes assessed. Furthermore, as previously mentioned, the current T-DXd subgroups with limited duration of T-DXd treatment may represent patients who were early progressors. Therefore, subsequent therapy outcomes for patients with longer T-DXd treatment duration will be informative.

There are also general limitations of real-world data that affect our conclusions. Patients were not randomized to treatment groups and when treatment comparisons were made, no statistics were used to balance patient characteristics between treatment groups in the included studies. Tumor response and disease progression was not assessed via RECIST criteria, and there was the potential that measured and unmeasured confounders impacted the clinical outcomes observed. There was also the possibility of missing or inaccurate data in the databases on which these studies were based. Finally, generalizability of the findings is limited due to the country-specific data sources that were predominantly from the United States.

5. Conclusions

In routine clinical practice, tucatinib-based regimens demonstrate clinically meaningful effectiveness in the 3L–5L treatment settings, including following prior treatment with T-DXd. However, limitations in the currently available evidence preclude definitive clinical recommendations and highlight the need for additional robust RWE. This review has identified several key data gaps, including the absence of real-world safety data for tucatinib, a lack of comparative effectiveness studies, limited insight into treatment patterns within the evolving HER2-positive metastatic breast cancer landscape, and insufficient data on PROs and HRQoL for HER2-directed therapies more broadly. Generating such RWE will be essential to inform clinical decision-making and optimize outcomes for patients with HER2-positive metastatic breast cancer.

Data sharing

No new data were created or analyzed during this study. Data sharing is not applicable to this article.

Funding

This review was sponsored by Pfizer Inc, which had a role in review design, data collection, analysis and interpretation of data, writing of the report, and decision to submit the article for publication.

CRediT authorship contribution statement

Rodrigo Sanchez-Bayona: Conceptualization, Writing – original draft, Writing – review & editing. Carey Anders: Writing – original draft, Writing – review & editing. Rupert Bartsch: Writing – original draft, Writing – review & editing. David A. Cameron: Writing – original draft, Writing – review & editing. Eva M. Ciruelos: Writing – original draft, Writing – review & editing. Carmen Criscitiello: Writing – original draft, Writing – review & editing. Francois P. Duhoux: Writing – original draft, Writing – review & editing. Theodoros Foukakis: Writing – original draft, Writing – review & editing. Jean-Sebastien Frenel: Writing – original draft, Writing – review & editing. Joseph Gligorov: Writing – original draft, Writing – review & editing. Peter A. Kaufman: Writing – original draft, Writing – review & editing. Matteo Lambertini: Writing – original draft, Writing – review & editing. Nathalie LeVasseur: Writing – original draft, Writing – review & editing. Alicia Okines: Writing – original draft, Writing – review & editing. Frédérique Penault-Llorca: Writing – original draft, Writing – review & editing. Volkmar Müller: Writing – original draft, Writing – review & editing. Matthew T. Blahna: Conceptualization, Writing – original draft, Writing – review & editing. Edward Neuberger: Conceptualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing. Sofia Simon: Conceptualization, Writing – original draft, Writing – review & editing. Emanuele Zanucco: Writing – original draft, Writing – review & editing. Christian Rosé: Conceptualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing. Giuseppe Curigliano: Visualization, Writing – original draft, Writing – review & editing. Nadia Harbeck: Writing – original draft, Writing – review & editing.

Declaration of competing interests

RS-B reports advisory/consulting/speaker fees from Roche, AstraZeneca, Novartis, Lilly, Daiichi Sankyo, Pfizer, Eisai, GlaxoSmithKline, Reveal Genomics, and Gilead; travel expenses from Pfizer, AstraZeneca, Gilead, Novartis, and Roche.

CA declares research funding from AstraZeneca, Caris Life Sciences, Elucida Oncology, Inc., G1 Therapeutics, Incyclix Bio, Lilly, Merck, Nektar, Novartis, Pfizer, Puma Biotechnology, Seagen Inc., Tesaro, and Zion Pharma; a compensated consultant role with AstraZeneca, Athenex, Eisai, Elucida, Genentech, Immunomedics, Ipsen, Novartis, Roche, and Seagen Inc.; honoraria from AstraZeneca, Athenex, Eisai, Elucida Oncology, Inc., Genentech, Immunomedics, Ipsen, Novartis, and Seagen Inc.; and royalties from UpToDate and Jones and Bartlet.

RB declares honoraria from AstraZeneca, Bristol Myers Squibb, Daiichi Sankyo, Eisai and Lilly; consulting/advisory roles with AstraZeneca, Daichi Sankyo, Eisai and Lilly; research funding from Daiichi Sankyo; and travel, accommodations and expenses from Merck Sharp & Dohme, Daiichi Sankyo, Novartis and Pfizer.

DAC reports consulting fees from AstraZeneca, Celgene, Daiichi Sankyo, Eisai, Eli Lilly, F. Hoffmann-La Roche, GlaxoSmithKline, Novartis, Oncolytics Biotech, Pfizer, Prima Biomed, Puma, Research Triangle Institute, Sanofi, Seattle Genetics, Synthon, and Zymeworks.

EMC reports advisory/consulting/speaker fees from Roche, AstraZeneca, Novartis, Lilly, Daiichi Sankyo, Pfizer, and Gilead; travel expenses from Pfizer, AstraZeneca, MSD, Novartis, and Roche.

CC has participated in advisory or consultancy roles and speakers' bureau engagements for Eli Lilly, Pfizer, Novartis, Roche, AstraZeneca, MSD, Daiichi Sankyo, Gilead, and Seagen.

FPD reports a postdoctoral research grant from Fondation Belge contre le Cancer; consulting fee from Amgen, AstraZeneca, Daiichi Sankyo, Eli Lilly, Gilead Sciences, Merck Sharp & Dohme, Novartis, Pfizer, Pierre Fabre, Roche and Seagen (paid to his institution); and travel support from Amgen, AstraZeneca, Daiichi Sankyo, Gilead Sciences, Pfizer, Roche and Teva.

TF reports institutional fees for consultancy to Roche, Astra Zeneca, Gilead Sciences, Novartis, Daiichi Sankyo; honoraria from UpToDate; research funding to his institution from Pfizer, AstraZeneca, Novartis and Veracyte.

J-SF declares consulting fees from Astra-Zeneca, Clovis, GlaxoSmithKline, Merck Sharp & Dohme, Daiichi Sankyo, Pfizer, Lilly, Novartis, Gilead, Seagen, and Exact Science; honoraria from Lilly, Novartis, Astra-Zeneca, Gilead, Daiichi Sankyo, Seagen, and Merck Sharp & Dohme; non-financial support from Pfizer, Lilly, Novartis, Astra-Zeneca, Clovis, GlaxoSmithKline, Gilead, Daiichi Sankyo, Seagen, and Merck Sharp & Dohme.

JG reports grants from Roche-Genentech, Eisai, Exact Science, Pfizer, and Mylan; and personal fees and travel support from Roche-Genentech, Novartis, Daiichi Sankyo, Merck Sharp & Dohme, Eisai, Exact Science, Pfizer, Mylan, Lilly, Pierre Fabre, and AstraZeneca.

PAK reports financial relationships with Amgen, AstraZeneca, Eisai, H3 BioMedicine, Lilly, Macrogenics, Pfizer, Polyphor, Roche/Genentech, and Sanofi.

ML reports advisory role for Roche, Lilly, Novartis, AstraZeneca, Pfizer, Seagen, Gilead, Merck Sharp & Dohme, Exact Sciences, Pierre Fabre, Menarini, Nordic Pharma; speaker honoraria from Roche, Lilly, Novartis, Pfizer, Sandoz, Libbs, Daiichi Sankyo, Takeda, Menarini, AstraZeneca; travel grants from Gilead, Daiichi Sankyo, Roche; research funding (to his Institution) from Gilead.

NL reports advisory board roles with: Astra Zeneca, Eli Lilly, Gilead, Knight Therapeutics, Merck, Novartis, Pfizer, Roche, Seagen, TerSera; speaker's honoraria from: Astra Zeneca, Daiichi Sankyo, Eli Lilly, Gilead, Merck, Knight Therapeutics, Merck, Novartis, Pfizer, Roche, Seagen, TerSera; research funds (to her institution): Abbvie, Astra Zeneca, Eli Lilly, Exact Sciences, Gilead, Pfizer, Roche.

AO reports research funding from Pfizer and Roche; honoraria for presentations from AstraZeneca, Esai, Gilead, Lilly, Pfizer, Roche and Seagen; advisory board fees from AstraZeneca, Pfizer, Roche and Seagen; and conference support from AstraZeneca, Lilly, Merck, Novartis and Roche.

FP-L reports speaker/consultancy honoraria from AstraZeneca, Daiichi Sankyo, Eisai, Pfizer, Merck Sharp & Dohme, Novartis, Roche, Seagen, Medscape, Gilead, Janssen, Lilly, Astellas, and Amgen; and travel grants from AstraZeneca, Roche, Pfizer, Daiichi Sankyo, Gilead, Merck Sharp & Dohme, and Novartis.

VM reports speaker honoraria from AstraZeneca, Daiichi Sankyo, Eisai, Pfizer, Merck Sharp & Dohme, Medac, Novartis, Roche, Seagen, onkowissen, high5 Oncology, Medscape, Gilead, Pierre Fabre, and I-MED Institute; consultancy honoraria from Roche, Pierre Fabre, PINK, ClinSol, Novartis, Merck Sharp & Dohme, Daiichi Sankyo, Eisai, Lilly, Seagen, Gilead, and Stemline; institutional research support from Novartis, Roche, Seagen, Genentech, and AstraZeneca; and travel grants from AstraZeneca, Roche, Pfizer, Daiichi Sankyo, and Gilead.

MTB, SS, EZ, CR are employees of and stockholders in Pfizer Inc.

EN reports no conflicts of interest.

GC reports receiving honoraria from Ellipses Pharma; consulting or advisory role for Roche/Genentech, Pfizer, Novartis, Lilly, Foundation Medicine, Bristol Myers Squibb, Samsung, AstraZeneca, Daiichi-Sankyo, Boehringer Ingelheim, GlaxoSmithKline, Seagen, Guardant Health, Veracyte, Celcuity, Hengrui Therapeutics, Menarini, and Merck; speakers' bureau for Roche/Genentech, Novartis, Pfizer, Lilly, Foundation Medicine, Samsung, Daiichi-Sankyo, Seagen, and Menarini; research funding from Merck (to his institution); and travel, accommodations, and expenses from Roche/Genentech, Pfizer, and Daiichi Sankyo.

NH reports consulting fees from Daiichi Sankyo, Exact Sciences, Gilead, F. Hoffmann-La Roche, and Sandoz; honoraria from Astra Zeneca, Daiichi Sankyo, Gilead, F. Hoffmann-La Roche, Lilly Deutschland, Menarini-Stemline, MSD, Novartis, Pfizer, Pierre Fabre Pharmaceuticals, Stemline Menarini, Viatris, and Zuelligpharma; and additional business interests in West German Study Group (WSG).

Appendix A. Supplementary data

The following is the Supplementary data to this article: