Evidence and gaps in tumor-agnostic therapies for breast cancer: a narrative review

Introduction

Tumor-agnostic therapies have redefined oncology by enabling treatment selection based on shared molecular alterations rather than tissue of origin (1,2). This paradigm targets genomic drivers that promote tumorigenesis across histologies and has established a biomarker-centered framework for precision medicine (1,2). Since 2017, multiple U.S. Food and Drug Administration (FDA) approvals have validated this approach, demonstrating that targeting specific molecular alterations can produce durable clinical responses across diverse tumor types (3,4). These approvals have also influenced regulatory strategies and accelerated the development of biomarker-defined therapies using basket and platform trial designs (5,6).

Application of tumor-agnostic therapy in breast cancer remains limited. Standard management is defined by hormone receptor and human epidermal growth factor receptor 2 (HER2) status, with treatment decisions based on estrogen receptor (ER), progesterone receptor (PR), and HER2 expression rather than rare genomic alterations (7). Next-generation sequencing has identified a subset of breast cancers harboring tumor-agnostic biomarkers, including microsatellite instability-high (MSI-H), mismatch repair deficiency (dMMR), tumor mutational burden-high (TMB-high), NTRK and RET fusions, BRAF V600E mutations, and HER2 overexpression (1,8). These alterations occur at low frequency and are most often identified in metastatic or treatment-refractory disease. Targeted therapies directed at these biomarkers may provide additional treatment options beyond conventional endocrine, cytotoxic, or HER2-directed therapies.

Breast cancer is a biologically heterogeneous disease composed of molecular subtypes with distinct genomic and immunologic profiles, including hormone receptor–positive/HER2-negative, HER2-positive, and triple-negative disease. Tumor-agnostic biomarkers are unevenly distributed across these subtypes and remain rare in all groups (1,6). Representation of breast cancer in tumor-agnostic registrational trials has been minimal. Several pivotal studies included no breast cancer patients, while others included only small subgroups, limiting the ability to define histology-specific efficacy, durability, and safety. Clinical application therefore relies on extrapolation from non-breast tumor types and requires careful interpretation of biological context (1,6).

This review evaluates all FDA-approved tumor-agnostic therapies in the context of breast cancer. The biological rationale, regulatory framework, and clinical evidence supporting these approvals are synthesized, with emphasis on biomarker prevalence, trial representation, and breast-specific outcomes. A guideline-informed approach to molecular testing is outlined, and key evidence gaps are identified. The objective is to define the current role of tumor-agnostic therapies in breast cancer and to establish priorities for future research, including generation of histology-specific data and optimization of precision oncology strategies. We present this article in accordance with the Narrative Review reporting checklist (available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-2025-1-69/rc).

Methods

This narrative review was conducted using a structured literature search of PubMed/MEDLINE, Embase, and ClinicalTrials.gov, together with U.S. FDA regulatory documents and prescribing information, from database inception through November 1, 2025. Search terms included combinations of “tumor-agnostic”, “tissue-agnostic”, “histology-agnostic”, “basket trial”, “precision oncology”, and “FDA approval”, along with biomarker-specific terms and associated therapies, including MSI-H/dMMR (pembrolizumab, dostarlimab), tumor mutational burden (TMB) (pembrolizumab), NTRK fusions (larotrectinib, entrectinib, repotrectinib), BRAF V600E (dabrafenib, trametinib), RET fusions (selpercatinib), and HER2 overexpression (trastuzumab deruxtecan). Searches were supplemented by cross-referencing FDA approval summaries, prescribing information, and pivotal trial publications to ensure completeness and regulatory accuracy. The detailed search strategy is summarized in Table 1.

Inclusion criteria were restricted to FDA-approved tumor-agnostic therapies supported by prospective clinical trials or registration-enabling analyses. Eligible studies reported clinical outcomes including objective response rate (ORR), duration of response (DoR), progression-free survival (PFS), or overall survival (OS), and provided data on biomarker definition, trial design, or breast cancer representation. Extension cohorts, subgroup analyses, and biomarker prevalence studies were included when they contributed breast-specific or mechanistic insights. Histology-specific studies without tumor-agnostic relevance, preclinical-only studies, and non-English publications were excluded unless used for contextual comparison.

Study selection was performed through sequential screening of titles, abstracts, and full texts by multiple authors. Screening and data extraction were conducted independently, with discrepancies resolved through discussion and consensus. Data extraction focused on trial design, patient population, biomarker definition, clinical outcomes, and the extent of breast cancer representation within each dataset.

Molecular landscape of tumor-agnostic biomarkers in breast cancer

Conceptual framework and biology

Tumor-agnostic therapy refers to treatment selection based on shared molecular alterations rather than the anatomic site of tumor origin. This approach has emerged from advances in next-generation sequencing that enable identification of oncogenic drivers across histologies, allowing patients with different tumor types to receive the same targeted therapy when a common biomarker is present (1,2). Regulatory approval of these therapies has been supported by basket trial designs, which enroll patients across multiple cancer types based on a defined molecular alteration rather than histology (5,6).

Current tumor-agnostic approvals target a limited set of biologically validated pathways that function as lineage-independent oncogenic drivers. Immune checkpoint inhibitors, including programmed cell death protein 1 (PD-1) blockade, leverage the immunogenicity of MSI-H and dMMR tumors, which accumulate frameshift insertion-deletion mutations that generate neoantigens and activate innate immune signaling pathways, producing an inflamed tumor microenvironment enriched in tumor-infiltrating lymphocytes (3,4). TMB-high tumors are also approved for PD-1 blockade, although TMB functions as a surrogate for neoantigen load generated primarily through single-nucleotide variants and demonstrates variable predictive value across tumor types (3,4). Targeted therapies act on specific oncogenic dependencies, including tropomyosin receptor kinase (TRK) signaling in NTRK fusion-positive tumors, mitogen-activated protein kinase (MAPK) pathway activation in BRAF V600E-mutant cancers, and RET kinase activation in RET fusion-driven disease (9,10). Antibody-drug conjugates such as trastuzumab deruxtecan target HER2-expressing tumor cells by delivering a membrane-permeable cytotoxic payload, with a bystander effect that allows the released drug to affect adjacent tumor cells regardless of HER2 expression (11).

These therapies demonstrate that select molecular alterations can function as dominant oncogenic drivers across tumor types. The predictive value of these biomarkers is influenced by tumor lineage, co-mutation patterns, immune contexture, and tumor microenvironment, indicating that tumor-agnostic activity exists along a spectrum rather than as a purely histology-independent phenomenon (12).

Biomarker epidemiology in breast cancer

The prevalence of tumor-agnostic biomarkers in breast cancer is low and varies across molecular subtypes and detection methods, limiting broad clinical applicability. MSI-H and dMMR phenotypes are rare, with prevalence estimates generally below 2% and often less than 1% depending on assay methodology and cohort selection (8). These alterations occur less frequently than in colorectal or endometrial cancers and are identified across receptor-defined subtypes, including ER-positive disease. Enrichment is observed in tumors with increased lymphocytic infiltration and in hereditary cancer syndromes (8).

TMB-high, defined as ≥10 mutations per megabase, is identified in approximately 5% of breast cancers overall and is more frequently observed in metastatic disease, with reported rates approaching 5–8% (4). TMB-high status occurs across receptor-defined subtypes with relatively similar frequency. Median TMB values are higher in triple-negative breast cancer. TMB-high breast cancers are predominantly driven by APOBEC-mediated mutagenesis, accounting for the majority of hypermutated cases, while dMMR represents a smaller but clinically relevant subset (4). These mutational processes differ biologically, with APOBEC-driven tumors characterized by specific nucleotide substitution patterns and dMMR tumors characterized by frameshift mutations.

Gene fusions involving neurotrophic tyrosine receptor kinase (NTRK) are exceedingly rare in breast cancer, with prevalence estimates well below 1% (9). Detection rates vary by platform, with RNA-based sequencing identifying higher frequencies than DNA-based assays. Secretory breast carcinoma represents a distinct subtype defined by the ETV6-NTRK3 fusion, in which NTRK rearrangements are nearly universal and confer sensitivity to TRK inhibition. NTRK fusions in non-secretory breast cancer are heterogeneous and uncommon (9).

RET fusions and BRAF V600E mutations occur in a very small fraction of breast cancers, typically less than 1%, and are usually identified through comprehensive genomic profiling (10). These alterations are established oncogenic drivers in other malignancies but remain rare in breast cancer and are primarily relevant in molecularly selected cases.

HER2 amplification or overexpression is more common, occurring in approximately 15–20% of invasive breast cancers, and represents the most prevalent biomarker among tumor-agnostic targets (11). Additional tumors demonstrate low-level HER2 expression, now classified as HER2-low, which has therapeutic relevance in the context of antibody–drug conjugates. HER2-directed therapies are well established in breast oncology, and tumor-agnostic approval of HER2-targeted agents extends this approach beyond traditional histology-specific indications (11).

Tumor-agnostic biomarkers in breast cancer are rare and unevenly distributed across subtypes. Molecular testing is most appropriate in metastatic or treatment-refractory disease, where identification of actionable alterations may inform treatment selection.

Clinical evidence for tumor-agnostic therapies in breast cancer

Pembrolizumab for MSI-H/dMMR tumors

Pembrolizumab became the first drug to receive tumor-agnostic FDA approval in 2017 for adult and pediatric patients with unresectable or metastatic MSI-H or dMMR solid tumors that had progressed after prior therapy and lacked satisfactory treatment options (13). This approval established a regulatory precedent demonstrating that a shared biomarker, rather than tissue of origin, could support histology-independent treatment selection.

The approval was supported by pooled data from five clinical trials: KEYNOTE-016, KEYNOTE-164, KEYNOTE-158, KEYNOTE-012, and KEYNOTE-028. Across these studies, 149 patients with MSI-H or dMMR tumors representing 15 tumor types were included, with an overall response rate of 39.6% [95% confidence interval (CI): 31.7% to 47.9%] (14). Complete responses occurred in 7% of patients, 78% of responses were ongoing at 6 months or longer, and DoR ranged from 1.6+ to 22.7+ months (14). Subsequent KEYNOTE-158 analyses expanded the non-colorectal MSI-H/dMMR dataset and confirmed durable activity across multiple histologies without changing conclusions regarding breast cancer representation (4,15,16).

Breast cancer representation in the pivotal pembrolizumab MSI-H/dMMR approval datasets was absent. No breast cancer patients were reported in KEYNOTE-016, KEYNOTE-164, KEYNOTE-158, KEYNOTE-012, or KEYNOTE-028, and breast histologies were not identified in any published tumor-type breakdown or FDA approval summary (3,4,13-17). The total number of breast cancer patients included in the approval-supporting datasets was 0, corresponding to 0% of the study population. No breast-specific efficacy outcomes, including ORR, response category breakdown, DoR, PFS, or OS, were reported.

The absence of breast cancer patients is consistent with the underlying biology of this disease. MSI-H and dMMR are rare in breast cancer, with prevalence estimates generally below 2% in unselected cohorts and substantially lower in specific subtypes (18-20). In a cohort of 1,635 breast cancers, 1.9% were mismatch repair deficient, while in a study of 440 triple-negative breast cancers, 0.2% were dMMR by immunohistochemistry and no MSI-H cases were detected by polymerase chain reaction (18,20). These rates limit the feasibility of enrolling MSI-H/dMMR breast cancer patients in tumor-agnostic registrational trials.

Pembrolizumab remains biologically relevant for the rare patient with confirmed MSI-H or dMMR breast cancer, and current practice guidelines recognize the tumor-agnostic indication in this setting despite the absence of direct breast-specific trial data (21). Clinical application in breast oncology is based entirely on extrapolation from non-breast cohorts, and no direct estimate of efficacy, durability, or subgroup-specific outcomes in breast cancer is available from registrational evidence.

Pembrolizumab for MSI-H/dMMR tumors represents a landmark tumor-agnostic approval with durable activity across multiple malignancies. In breast cancer, its application is constrained by the extreme rarity of the biomarker and the absence of breast cancer patients in the approval-supporting datasets. Future progress will depend on registry-based data and case aggregation to define outcomes in this rare but potentially actionable subset.

Larotrectinib for NTRK fusion-positive tumors

Larotrectinib was the first selective tropomyosin receptor kinase inhibitor to receive tumor-agnostic FDA approval in 2018 for patients with solid tumors harboring an NTRK gene fusion, establishing proof of concept that targeting a shared oncogenic driver can produce durable responses independent of tumor histology (9,22). The approval was supported by an integrated analysis of three early-phase trials, LOXO-TRK-14001, SCOUT, and NAVIGATE, which enrolled patients with NTRK fusion-positive solid tumors across multiple histologies (9,23). In the initial analysis of 55 patients, the ORR was 75% (95% CI: 61% to 85%), with a 13% complete response rate and a 62% partial response rate (9). Median DoR and PFS were not reached at initial reporting (9).

The expanded pooled analysis included 159 patients, with 153 evaluable for response, and demonstrated an investigator-assessed ORR of 79% (95% CI: 72% to 85%), a 16% complete response rate, median DoR of 35.2 months, median PFS of 28.3 months, and median OS of 44.4 months (23). The current FDA label dataset includes 339 patients with NTRK fusion-positive tumors and represents the most comprehensive evidence base for this approval (22).

Breast cancer representation increased with expanded enrollment. In the initial 55-patient dataset, 1 breast cancer patient was included (1.8%) (9,21). In the 159-patient dataset, 5 breast cancer patients were included (3.1%) (21,23). In the 339-patient FDA label dataset, 14 breast cancer patients were included, representing 4.1% of the total population (22).

Within the breast cancer subgroup, 6 patients had secretory breast carcinoma and 8 had non-secretory breast cancer (22). The overall ORR in breast cancer was 57% (8 of 14 patients; 95% CI: 29% to 82%) (22). Response rates differed by histologic subtype. Secretory breast carcinoma demonstrated an ORR of 83% (5 of 6 patients), while non-secretory breast cancer demonstrated an ORR of 38% (3 of 8 patients) (22).

Durability of response was substantial within this subgroup. DoR ranged from 7.4 to 58.2+ months among all breast cancer patients (22). In secretory breast carcinoma, DoR ranged from 11.1 to 58.2+ months. In non-secretory breast cancer, DoR ranged from 7.4 to 12.5+ months (22). PFS and OS were not reported separately for the breast cancer subgroup (22).

Secretory breast carcinoma is characterized by the ETV6-NTRK3 fusion in the majority of cases and demonstrates high sensitivity to TRK inhibition (24,25). NTRK fusions in non-secretory breast cancer are rare and heterogeneous, involving diverse fusion partners and NTRK genes, which may contribute to lower response rates (22).

Breast cancer patients comprised a small proportion of the overall dataset, and subgroup-specific reporting was limited. Complete versus partial response breakdown was not reported for breast cancer patients. PFS, OS, central nervous system activity, prior lines of therapy, and fusion partner distribution within the breast cancer subgroup were not reported (22). Trial-level attribution of breast cancer patients across the individual studies was not available.

Larotrectinib demonstrates high and durable activity in NTRK fusion-positive tumors. Breast cancer representation remains limited to 14 patients within a 339-patient dataset. Clinical activity is highest in secretory breast carcinoma. Activity in non-secretory breast cancer is variable and based on small numbers. Treatment decisions require molecular confirmation and clinical context.

Entrectinib for NTRK fusion-positive tumors

Entrectinib, a multikinase inhibitor targeting TRKA, TRKB, TRKC, ROS1, and ALK, received tumor-agnostic FDA approval in 2019 for adults and pediatric patients 12 years of age and older with NTRK fusion-positive solid tumors (26,27). The approval was based on an integrated efficacy analysis of three trials, ALKA-372-001, STARTRK-1, and STARTRK-2, which together formed the primary efficacy-evaluable dataset of 54 adults with NTRK fusion-positive solid tumors (27,28). The ORR was 57% (31 of 54 patients; 95% CI: 43% to 71%), including a 7% complete response rate and a 50% partial response rate (27). Median DoR was 10 months, median PFS was 11 months, and median OS was 21 months (27). The current FDA prescribing information reflects the same 54-patient dataset with extended follow-up and reports an ORR of 59% (95% CI: 45% to 72%), a 13% complete response rate, and a 46% partial response rate, with DoR ranging from 2.8 to 47.8+ months (26).

Breast cancer representation in the approval dataset consisted of six patients, corresponding to 11% of the 54-patient population (26,27). Mammary analogue secretory carcinoma of the salivary gland accounted for an additional 7 patients and represents a distinct entity despite histologic similarity (26,27). The breast cancer subgroup demonstrated an ORR of 83% (5 of 6 patients; 95% CI: 36% to 100%) (26,27). DoR ranged from 4.2 to 42.3+ months (26). Median DoR, PFS, and OS were not reported for the breast cancer subgroup.

Histologic subtype, receptor status, and disease characteristics were not reported for the breast cancer patients. Secretory and non-secretory breast cancer were not distinguished. Hormone receptor status, HER2 status, prior lines of therapy, and disease stage were not described. NTRK fusion gene and fusion partner distribution were not reported for the breast cancer subgroup, although the overall dataset included NTRK1, NTRK2, and NTRK3 fusions with ETV6-NTRK3 identified in 46% of patients (26). Trial-level attribution across ALKA-372-001, STARTRK-1, and STARTRK-2 was not available.

Intracranial activity was reported in the overall dataset. Twelve patients had baseline central nervous system disease by investigator assessment, with 11 confirmed by blinded independent central review (27). Among patients with measurable central nervous system metastases, intracranial ORR was 55% (27). The current FDA label reports intracranial responses in 3 of 4 patients with measurable central nervous system metastases (26). Central nervous system involvement and intracranial outcomes were not reported for the breast cancer subgroup.

An expanded integrated analysis including 121 patients demonstrated an ORR of 61.2%, a 15.7% complete response rate, median DoR of 20.0 months, median PFS of 13.8 months, and median OS of 33.8 months (28). Breast cancer-specific outcomes were not reported in this expanded dataset.

Entrectinib demonstrates activity in NTRK fusion-positive breast cancer, with 5 of 6 patients responding in the approval dataset. Interpretation is limited by small sample size, wide confidence interval, and absence of histologic and molecular characterization. Clinical application relies on confirmation of NTRK fusion and consideration of individual patient factors.

Pembrolizumab for TMB-high tumors

Pembrolizumab received tumor-agnostic FDA approval in 2020 for patients with unresectable or metastatic solid tumors with TMB-high, defined as 10 mutations per megabase or greater as determined by the FoundationOne CDx assay (4,13). The approval was based on a prespecified biomarker analysis of the phase II KEYNOTE-158 trial, which enrolled patients across 10 tumor-specific cohorts (4,28).

In the KEYNOTE-158 efficacy population, 790 treated patients were evaluable for TMB, and 102 patients (13%) met criteria for TMB-high status (4). The ORR in this cohort was 29% (95% CI: 21% to 39%), including 4% complete responses and 25% partial responses (4). Median DoR was not reached, with responses ranging from 2.2+ to 34.8+ months (4). Fifty-seven percent of responses lasted 12 months or longer, and 50% lasted 24 months or longer (4,13). Median PFS was 2.1 months, and median OS was 11.7 months (4).

Breast cancer was absent from this registrational dataset. Breast cancer was not an eligible tumor type in KEYNOTE-158, and 0 breast cancer patients were included in the approval-supporting cohort (4,13,21,28). No breast-specific efficacy outcomes, including ORR, DoR, PFS, or OS, were reported.

TMB-high status occurs in a minority of breast cancers. Reported prevalence ranges from approximately 3.8% to 8.4%, with higher rates observed in metastatic compared with primary disease (29). TMB-high breast cancers are present across receptor-defined subtypes. Enrichment has been reported in lobular compared with ductal histology (29). APOBEC-mediated mutagenesis represents the dominant underlying mutational process, accounting for the majority of hypermutated tumors, while dMMR accounts for a smaller subset (29). These biological differences distinguish TMB-high breast cancer from MSI-H or dMMR tumors and may influence response to immune checkpoint inhibition.

Prospective clinical data in breast cancer are limited to the TAPUR basket study. Twenty-eight patients with metastatic breast cancer and high TMB were treated with pembrolizumab (21,30). The ORR was 21% (95% CI: 8% to 41%), and the disease control rate was 37% (21,30). Median PFS was 10.6 weeks, and median OS was 30.6 weeks (21,30). These results were derived from a small, non-randomized cohort.

Pembrolizumab for TMB-high tumors is included in tumor-agnostic treatment recommendations despite the absence of breast cancer patients in the registrational trial (21). Clinical application in breast oncology is based on extrapolation from non-breast cohorts and limited prospective data. The predictive value of a threshold of 10 mutations per megabase varies across tumor types. APOBEC-driven hypermutation in breast cancer represents a distinct biological context compared with mismatch repair–deficient tumors.

Pembrolizumab for TMB-high tumors represents an important tumor-agnostic approval. Relevance in breast cancer is limited by the absence of breast cancer patients in KEYNOTE-158 and modest activity observed in small prospective cohorts. Further study is required to define which TMB-high breast cancers derive meaningful benefit from immune checkpoint inhibition.

Dostarlimab for dMMR tumors

Dostarlimab-gxly received tumor-agnostic FDA accelerated approval in 2021 for adults with recurrent or advanced deficient mismatch repair solid tumors that had progressed after prior therapy and had no satisfactory alternative treatment options (31,32). The approval was based on the GARNET trial, a multicohort, single-arm, open-label study in which the dMMR population from Cohorts A1 and F formed the regulatory basis for approval (31).

The FDA approval dataset included 209 patients with dMMR recurrent or advanced solid tumors. Cohort A1 included 103 patients with dMMR endometrial cancer, and Cohort F included 106 patients with non-endometrial dMMR tumors across multiple histologies (31). The ORR was 41.6% (95% CI: 34.9% to 48.6%), including a 9.1% complete response rate and a 32.5% partial response rate (31). Median DoR was 34.7 months, and 95.4% of responders maintained response for at least 6 months (31). PFS and OS were not reported in the FDA approval dataset (31).

Breast cancer representation consisted of a single patient. One breast cancer patient was included in Cohort F, representing 0.5% (1 of 209) of the efficacy population (31). This patient achieved a complete response with a DoR of 16.8+ months that remained ongoing at the time of data cutoff (31). An ORR cannot be calculated for a subgroup of one patient.

Histologic subtype, hormone receptor status, HER2 status, Lynch syndrome status, prior lines of therapy, and central nervous system involvement were not reported for this patient (31). PFS and OS were not reported. Adverse events attributable to this patient were not described (31). Trial-level context beyond cohort assignment was not provided.

The prevalence of dMMR in breast cancer is low. In a cohort of 1,635 breast cancers, 1.9% were mismatch repair deficient (18). In a series of 440 triple-negative breast cancers, 0.2% were dMMR by immunohistochemistry, and no cases were MSI-H by polymerase chain reaction testing (20). Additional genomic analyses report similar low frequencies across datasets (19). These rates limit the ability to enroll patients with dMMR breast cancer in tumor-agnostic registrational studies.

Dostarlimab demonstrates activity across dMMR solid tumors. Evidence in breast cancer is limited to a single patient with a complete and durable response. Pembrolizumab MSI-H/dMMR approval datasets included zero breast cancer patients. The dostarlimab dataset represents the only tumor-agnostic dMMR approval with any breast cancer representation. Treatment decisions in breast cancer rely on molecular confirmation and clinical context (21).

Dabrafenib plus trametinib for BRAF V600E-mutated tumors

Dabrafenib plus trametinib received tumor-agnostic FDA accelerated approval in 2022 for adult and pediatric patients with unresectable or metastatic solid tumors harboring a BRAF V600E mutation who had progressed following prior treatment and had no satisfactory alternative options (33-37). The approval was supported by data from the Rare Oncology Agnostic Research (ROAR) basket trial and the National Cancer Institute Molecular Analysis for Therapy Choice (NCI-MATCH) Subprotocol H, with additional support from pediatric studies and prior disease-specific approvals in other BRAF V600E-driven cancers (36,37).

The adult efficacy dataset used for the tumor-agnostic approval included 131 patients with BRAF V600E-mutated non-colorectal solid tumors (34). The ORR by independent review was 41% (95% CI: 33% to 50%) (34). Ninety percent of patients had received prior systemic therapy (34). In the ROAR trial, anaplastic thyroid carcinoma demonstrated an ORR of 56%, median DoR of 14.4 months, median PFS of 6.7 months, and median OS of 14.5 months (36). In NCI-MATCH Subprotocol H, the ORR was 38%, with median PFS of 11.4 months and median OS of 28.6 months (37).

Breast cancer representation in the approval-supporting datasets was absent. No breast cancer patients were enrolled in the predefined ROAR cohorts, which included anaplastic thyroid carcinoma, biliary tract cancer, gastrointestinal stromal tumor, adenocarcinoma of the small intestine, low-grade glioma, high-grade glioma, hairy cell leukemia, and multiple myeloma (36). No breast cancer patients were reported in NCI-MATCH Subprotocol H, which enrolled patients across multiple tumor histologies but did not identify breast cancer among included tumor types (37,38). The FDA label lists all tumor types contributing to the approval dataset, and breast cancer is not included (3).

Breast cancer patients accounted for 0 of 131 patients in the adult efficacy dataset, corresponding to 0% representation. No breast-specific efficacy outcomes were reported. ORR, DoR, PFS, OS, response category breakdown, central nervous system involvement, and prior treatment history were not reported for breast cancer (34,36,37).

BRAF V600E mutations are rare in breast cancer. Prevalence is less than 1% across genomic profiling studies (39). A whole-genome sequencing analysis identified BRAF driver mutations in approximately 0.2% of breast cancers (39). Prior basket studies of BRAF-directed therapy have included small numbers of breast cancer patients, confirming the presence of this alteration but not generating sufficient data to define clinical benefit (40).

Dabrafenib plus trametinib is theoretically applicable to breast cancer under the tumor-agnostic FDA indication. Clinical use in breast oncology is based entirely on extrapolation from non-breast cohorts. No direct evidence exists to estimate efficacy or durability in breast cancer. The absence of breast cancer patients in registrational datasets and the low prevalence of BRAF V600E mutations limit clinical applicability.

Dabrafenib plus trametinib demonstrates activity across BRAF V600E-mutated solid tumors. Evidence in breast cancer is limited by the absence of enrolled patients in approval-supporting studies. Clinical relevance in breast oncology remains dependent on rare case identification and extrapolation from other tumor types.

Selpercatinib for RET fusion-positive tumors

Selpercatinib received tumor-agnostic FDA accelerated approval in 2022 for patients with advanced RET fusion-positive solid tumors that had progressed after prior therapy and had no satisfactory alternative treatment options (10,41). The approval was supported by the LIBRETTO-001 trial, a global phase 1/2 basket study evaluating RET fusion-positive malignancies across multiple histologies (10,42).

In the tumor-agnostic cohort excluding non-small-cell lung cancer and thyroid cancer, 45 patients were enrolled and 41 were evaluable for efficacy across 14 tumor types (10). The ORR by independent review was 44% (18 of 41 patients; 95% CI: 28% to 60%), including a 4.9% complete response rate and a 39% partial response rate (10). Median DoR was 24.5 months, with 67% of responses lasting at least 6 months (10). Median PFS was 13.2 months, and median OS was 18.0 months (10). The estimated 18-month OS rate was 51.7% (10).

Breast cancer representation consisted of 2 patients, corresponding to 4.9% (2 of 41) of the efficacy-evaluable population (41). Both patients achieved objective responses, resulting in an ORR of 100% (2 of 2 patients). One patient achieved a complete response and one achieved a partial response (41). The patient with a complete response had no measurable disease at baseline. DoR ranged from 2.3+ to 17.3 months, with one response ongoing at the time of data cutoff (41).

Histologic subtype, hormone receptor status, HER2 status, RET fusion partner, prior lines of therapy, and central nervous system involvement were not reported for the breast cancer subgroup (10,41). PFS and OS were not reported separately. No subgroup-level estimates beyond individual patient outcomes were provided.

RET fusions are rare in breast cancer. Prevalence is less than 1% across genomic datasets (43,44). Large sequencing studies report frequencies as low as approximately 0.04%, confirming the rarity of this alteration (43). RET overexpression occurs more frequently but does not confer sensitivity to selective RET inhibition (43,44).

Selpercatinib demonstrates activity in RET fusion-positive tumors. Evidence in breast cancer is limited to 2 patients. The small sample size precludes estimation of generalizable efficacy. Clinical application relies on identification of a RET fusion and individualized treatment decisions based on molecular findings (41).

Repotrectinib for NTRK fusion-positive tumors

Repotrectinib is a next-generation TRK inhibitor designed to overcome solvent-front and gatekeeper resistance mutations that limit the durability of first-generation TRK inhibitors. It received tumor-agnostic FDA accelerated approval in 2024 for patients with locally advanced or metastatic NTRK fusion-positive solid tumors (45-47). The approval was based on the TRIDENT-1 trial, a global phase I/II multicohort study evaluating tumors with NTRK, ROS1, and ALK rearrangements (45-47).

The NTRK fusion cohorts included 88 adult patients, with 40 in the TKI-naive cohort and 48 in the TKI-pretreated cohort (45). The ORR by independent review was 58% (95% CI: 41% to 73%) in TKI-naive patients, including a 15% complete response rate and a 43% partial response rate (45). The ORR in TKI-pretreated patients was 50% (95% CI: 35% to 65%), with no complete responses and a 50% partial response rate (45). Median DoR was not estimable in the TKI-naive cohort and was 9.9 months in the TKI-pretreated cohort (45). Eighty-three percent of TKI-naive responders and 42% of TKI-pretreated responders maintained response for at least 12 months (45).

Breast cancer representation consisted of 3 patients, corresponding to 3.4% (3 of 88) of the efficacy population (45). Two patients were treated in the TKI-naive cohort and one patient in the TKI-pretreated cohort (45). All three were classified as adenocarcinoma (45).

Breast-specific outcomes differed by treatment setting. In the TKI-naive cohort, both breast cancer patients experienced progressive disease, resulting in an ORR of 0% (0 of 2) (45). In the TKI-pretreated cohort, the single breast cancer patient achieved a partial response, resulting in an ORR of 100% (1 of 1) (45). DoR for this patient was 15.6+ months and remained ongoing at the time of data cutoff (45). The combined ORR across the breast cancer subgroup was 33% (1 of 3).

Histologic subtype beyond adenocarcinoma was not reported for the breast cancer patients (45). NTRK gene subtype and fusion partner were not specified (45). Hormone receptor status, HER2 status, prior lines of therapy, and central nervous system involvement were not reported (45). PFS and OS were not reported separately for the breast cancer subgroup (45).

An expanded analysis of TRIDENT-1 reported outcomes in larger cohorts with longer follow-up, including 51 TKI-naive and 69 TKI-pretreated patients (47). ORRs remained consistent at 59% and 48%, respectively (47). Median PFS was 30.3 months in TKI-naive patients and 7.4 months in TKI-pretreated patients (47). Breast cancer-specific outcomes were not reported in this expanded dataset (47).

Repotrectinib demonstrates activity in NTRK fusion-positive tumors in both treatment-naive and resistance settings. Evidence in breast cancer is limited to 3 patients. The absence of response in TKI-naive patients and the durable response in a TKI-pretreated patient reflect heterogeneity within this small subgroup. Clinical applicability in breast cancer remains constrained by small sample size and lack of detailed subgroup characterization.

Fam-trastuzumab deruxtecan for HER2 IHC 3+ tumors

Fam-trastuzumab deruxtecan (T-DXd), an antibody-drug conjugate targeting HER2, received tumor-agnostic FDA accelerated approval in 2024 for adult patients with unresectable or metastatic HER2-positive (immunohistochemistry 3+) solid tumors who had progressed on prior therapy and had no satisfactory alternative treatment options (11,48). The approval was based on data from DESTINY-PanTumor02, DESTINY-Lung01, and DESTINY-CRC02, which evaluated HER2-expressing malignancies across multiple non-breast tumor types (11,49,50).

The combined approval dataset included 192 patients with HER2 IHC 3+ tumors (11). DESTINY-PanTumor02 included 111 patients and demonstrated an ORR of 51.4% (95% CI: 41.7% to 61.0%), with a complete response rate of 2.7% and a partial response rate of 48.6% (11). Median DoR was 19.4 months (11). DESTINY-Lung01 included 17 patients with HER2 IHC 3+ tumors and demonstrated an ORR of 52.9% (95% CI: 27.8% to 77.0%), including a 5.9% complete response rate and a 47.1% partial response rate, with a median DoR of 6.9 months (11,49). DESTINY-CRC02 included 64 patients and demonstrated an ORR of 46.9% (95% CI: 34.3% to 59.8%), with all responses classified as partial responses and a median DoR of 5.5 months (11,50).

Breast cancer representation was absent across all tumor-agnostic approval datasets. Breast cancer was explicitly excluded from DESTINY-PanTumor02, and no breast cancer patients were included in DESTINY-Lung01 or DESTINY-CRC02() (11,48). Breast cancer accounted for 0 of 192 patients, corresponding to 0% of the approval population. No breast-specific efficacy outcomes were reported, including ORR, DoR, PFS, or OS (11).

HER2 overexpression occurs in approximately 15% to 20% of breast cancers (11). T-DXd has established efficacy in breast cancer through disease-specific trials. Clinical activity has been demonstrated in HER2-positive, HER2-low, and HER2-ultralow breast cancer populations, leading to multiple FDA approvals based on randomized phase 3 data (50).

The tumor-agnostic approval is intended for HER2 IHC 3+ tumors outside of breast cancer and other malignancies with existing disease-specific indications. Clinical use in breast cancer is based on disease-specific approvals rather than the tumor-agnostic indication. Treatment selection in breast oncology follows established guidelines supported by randomized clinical trial data.

T-DXd demonstrates activity across HER2 IHC 3+ solid tumors. Evidence in breast cancer is derived entirely from disease-specific trials. The tumor-agnostic dataset does not include breast cancer patients. Interpretation of tumor-agnostic approvals requires consideration of existing histology-specific evidence.

Breast cancer representation and key efficacy outcomes across all tumor-agnostic approvals are summarized in Table 2 in the same order as discussed above.

Diagnostics and clinical implementation

The clinical integration of tumor-agnostic therapies in breast cancer requires a selective, guideline-aligned diagnostic strategy that accounts for the low prevalence of actionable tumor-agnostic biomarkers and the clinical context in which they provide benefit. Standard breast cancer management already incorporates routine biomarker testing for ER, PR, and HER2 status across all disease stages (7). Tumor-agnostic biomarkers represent a separate category of alterations identified through broader molecular profiling rather than routine histopathologic assessment.

Comprehensive genomic profiling is recommended in patients with recurrent or stage IV breast cancer to identify both breast-specific and tumor-agnostic therapeutic targets (21). Multigene panel testing using next-generation sequencing enables detection of actionable alterations including PIK3CA, ESR1, BRCA1/2, AKT1, and PTEN, as well as tumor-agnostic biomarkers such as MSI-H/dMMR, TMB-high status, NTRK fusions, RET fusions, and BRAF V600E mutations (1,21,50). Routine use of comprehensive sequencing in early-stage breast cancer is not standard practice, as the yield of clinically actionable tumor-agnostic alterations in this setting is low and clinical utility has not been established (21).

The choice of sequencing platform should be tailored to the biomarker class. DNA-based next-generation sequencing reliably detects point mutations, small insertions and deletions, copy number alterations, TMB, and microsatellite instability (4,50). RNA-based sequencing provides higher sensitivity for gene fusion detection, particularly for NTRK and RET rearrangements (21,25). DNA-based assays may miss fusions involving large intronic regions or low-expression transcripts, especially for NTRK2 and NTRK3 (25). RNA-based sequencing detects expressed fusion transcripts and provides direct evidence of functional rearrangements (25). Integrated DNA and RNA panels provide the most comprehensive molecular assessment when available and reduce the need for sequential testing (21).

Biomarker-specific testing approaches remain clinically relevant. MSI-H and dMMR status can be assessed using immunohistochemistry, PCR-based assays, or next-generation sequencing (4,21). Testing is recommended in the metastatic setting as part of comprehensive profiling rather than being limited to selected clinical scenarios (21). TMB is measured through validated sequencing platforms, with thresholds dependent on the assay used (4). The FoundationOne CDx assay serves as the FDA-approved companion diagnostic for pembrolizumab in TMB-high tumors using a cutoff of 10 mutations per megabase (4,13). Tissue-based testing is preferred for TMB and MSI assessment (21). Circulating tumor DNA assays may identify these alterations, but confirmation with tissue testing is appropriate when clinically actionable findings are detected (21).

Detection of NTRK and RET fusions requires sequencing approaches capable of identifying structural rearrangements (9,10). RNA-based sequencing is preferred for fusion detection due to higher sensitivity (21,25). Screening assays such as pan-TRK immunohistochemistry may be used in selected settings but require confirmatory testing due to limited specificity (21). BRAF V600E mutations are reliably detected through DNA-based sequencing panels (33,34). HER2 status continues to be assessed using immunohistochemistry with reflex in situ hybridization for equivocal results, consistent with established guidelines (11).

The timing of molecular testing should be individualized by breast cancer subtype and clinical course. Comprehensive genomic profiling is appropriate early in the treatment course for metastatic triple-negative breast cancer due to limited therapeutic options and the potential for biomarker-driven therapy (21). In hormone receptor-positive metastatic breast cancer, targeted testing such as PIK3CA mutation assessment is appropriate during first-line therapy, with broader genomic profiling considered at the time of endocrine resistance (21). Repeat biopsy or re-profiling at disease progression may identify acquired alterations, including gene fusions or hypermutation, that are not present in earlier disease stages (1,50).

Molecular testing results should be interpreted within a multidisciplinary framework that includes oncologists, pathologists, and molecular tumor boards (5,6). Rare tumor-agnostic alterations require careful interpretation of clinical actionability, supporting evidence, and therapeutic context (1,5). Treatment decisions should incorporate molecular findings alongside disease burden, prior therapies, performance status, and availability of standard treatment options (21). Enrollment in biomarker-driven clinical trials remains a preferred strategy when feasible (6).

Real-world implementation of comprehensive molecular profiling is limited by multiple barriers. Access to next-generation sequencing varies across institutions, particularly in community settings (50). RNA-based fusion testing is not uniformly available, which may lead to under-detection of actionable alterations (25). Turnaround time for sequencing can delay treatment decisions in rapidly progressive disease (50). Insurance coverage and reimbursement challenges remain common barriers to testing (50). Tissue adequacy and biopsy feasibility can further limit testing, particularly in heavily pretreated patients (21).

A targeted and guideline-informed approach to molecular testing provides a practical framework for identifying patients who may benefit from tumor-agnostic therapies in breast cancer. Broader access to high-quality genomic testing, improved standardization of sequencing assays, and integration of molecular tumor boards into routine care will be essential to optimize implementation of precision oncology strategies (1,50).

Strengths and limitations of the evidence

Tumor-agnostic therapy approvals represent a conceptual advance in precision oncology by demonstrating that treatments directed at shared molecular drivers can achieve clinically meaningful responses independent of tissue of origin. MSI-H and dMMR tumors accumulate neoantigens through defective DNA repair, enabling sensitivity to immune checkpoint blockade, while NTRK fusions encode constitutively active oncogenic drivers across histologies (3,9). Across pivotal trials, immune checkpoint inhibitors have demonstrated ORRs of approximately 30% to 42% in MSI-H/dMMR tumors, including 33.8% in the most recent KEYNOTE-158 analysis and 41.6% in the GARNET trial (3,9). TRK inhibitors have demonstrated ORRs ranging from 57% to 79% in NTRK fusion-positive cancers, including 75% in the initial larotrectinib dataset and 57% with entrectinib (3,9). These findings establish proof of concept for biomarker-driven therapy across tumor types. Detection of these biomarkers is supported by validated diagnostic platforms, including next-generation sequencing assays and FDA-approved companion diagnostics such as FoundationOne CDx for TMB assessment (4,13,50).

The primary limitation is the scarcity of breast cancer-specific data across all tumor-agnostic approvals. A consistent pattern is observed. Pembrolizumab for MSI-H/dMMR tumors, pembrolizumab for TMB-high tumors, and dabrafenib plus trametinib for BRAF V600E tumors included 0 breast cancer patients in pivotal datasets. Dostarlimab included 1 patient, selpercatinib included 2, and repotrectinib included 3. Entrectinib included 6 patients, and larotrectinib included 14 (22). Across all nine approvals, fewer than 30 breast cancer patients were represented. No tumor-agnostic approval provides breast cancer-specific PFS, OS, or central nervous system outcomes. Clinical use in breast cancer is therefore based almost entirely on extrapolation from non-breast cohorts.

Tumor-agnostic activity is not fully histology-independent. Treatment response is influenced by tumor lineage, co-mutation patterns, and tumor microenvironment (12). APOBEC-mediated mutagenesis is the dominant driver of TMB-high breast cancer, while dMMR accounts for a minority of cases, indicating distinct biological mechanisms within the same biomarker-defined group (12). NTRK fusion-positive secretory breast carcinoma demonstrates high sensitivity to TRK inhibition due to the near-universal presence of ETV6-NTRK3 fusions, whereas non-secretory breast cancers harbor heterogeneous fusions with lower response rates (9,25). Histology-dependent variability has been observed across tumor-agnostic datasets. In HER2 IHC 3+ tumors treated with trastuzumab deruxtecan, ORRs ranged from 0% in pancreatic cancer to approximately 70% in cervical cancer despite a shared biomarker (11). These findings demonstrate that tumor-agnostic biomarkers do not confer uniform therapeutic sensitivity across tumor types.

Methodological limitations further constrain interpretation. All tumor-agnostic approvals are based on single-arm basket trials without randomized comparator arms, relying on surrogate endpoints such as ORR and DoR rather than progression-free or OS (5,6). Most approvals were granted under the accelerated approval pathway, with continued approval contingent on confirmatory trials. These designs are appropriate for rare biomarker-defined populations but limit the ability to assess comparative effectiveness. Small sample sizes and wide confidence intervals reduce precision, particularly in breast cancer subgroups.

Diagnostic variability introduces additional uncertainty. TMB is not standardized across sequencing platforms, with differences in panel size, gene selection, and bioinformatic pipelines affecting classification (50). Misclassification rates at the 10 mutations per megabase threshold have been reported in the range of 10% to 19% depending on tumor type and assay design (50). Detection of gene fusions depends on testing methodology. DNA-based sequencing has reduced sensitivity for fusion detection compared with RNA-based sequencing, with studies demonstrating that up to 67% of NTRK fusion-positive tumors may be identified only through RNA-based analysis (50). These limitations may result in both under-detection and misclassification of actionable biomarkers.

Selection bias affects generalizability. Basket trial populations are enriched for patients with favorable performance status and access to specialized centers, which may not reflect real-world populations. Access to genomic testing varies across institutions and patient populations (50). Socioeconomic disparities influence testing utilization, with delayed or absent testing more common among underserved populations (50). These factors limit equitable implementation of tumor-agnostic therapies.

Reporting limitations further weaken the evidence base. Breast cancer subgroup data are frequently incomplete or absent. Key outcomes including PFS, OS, central nervous system activity, response category breakdown, prior lines of therapy, receptor subtype, and fusion partner distribution are often not reported for breast cancer patients in tumor-agnostic datasets (22). This limits assessment of durability, safety, and clinical applicability.

Tumor-agnostic therapies represent a biologically valid and clinically promising strategy for rare molecular subsets of breast cancer. Integration into routine practice remains constrained by limited breast-specific data, methodological limitations of basket trials, diagnostic variability, and barriers to implementation. Future progress requires prospective registry data, biomarker-enriched clinical trials, standardization of molecular diagnostics, and expanded access to genomic testing.

Future directions

Future progress in tumor-agnostic breast oncology will be defined by earlier identification of actionable biomarkers, generation of breast-specific evidence, and integration of molecular diagnostics into routine care. Comprehensive next-generation sequencing in metastatic or treatment-refractory disease remains the most effective strategy for identifying rare genomic alterations that may qualify for tumor-agnostic therapies (50). Multigene panel testing is recommended in recurrent or stage IV breast cancer to identify both breast-specific and tumor-agnostic targets (21). Unexpected findings, including gene fusions, hypermutation, or rare driver mutations, require confirmation with orthogonal assays and multidisciplinary review to ensure analytic validity and clinical relevance (21,50). RNA-based sequencing should be incorporated when DNA-based profiling does not identify an oncogenic driver, as a substantial proportion of actionable fusions are detectable only at the transcript level (50).

Longitudinal molecular profiling represents a clinically relevant strategy in advanced disease. Repeat biopsy at disease progression can identify acquired alterations, including HER2 expression changes, ESR1 resistance mutations, emergent gene fusions, and hypermutated states arising under therapeutic pressure (1,21). Subtype-specific testing strategies remain important. Comprehensive sequencing is appropriate early in the treatment course for metastatic triple-negative breast cancer, while broader profiling in hormone receptor-positive disease is most informative at the time of endocrine resistance (21). Expansion of sequencing adoption is expected to increase identification of rare actionable alterations and broaden the application of precision oncology in breast cancer (50).

Generation of breast-specific evidence within tumor-agnostic frameworks remains the highest priority. Across all nine FDA-approved tumor-agnostic indications, fewer than 30 breast cancer patients are represented in pivotal datasets, with several approvals including no breast cancer patients (22). Dedicated breast cancer cohorts embedded within basket and platform trials are required to define response rates, durability, co-mutation patterns, and resistance mechanisms in rare subsets such as NTRK, RET, and BRAF V600E (9,10). Prospective registry-based data collection provides an additional pathway for evidence generation in these low-frequency populations.

Emerging biomarkers with cross-histology activity may expand the tumor-agnostic landscape. Fibroblast growth factor receptor alterations have been evaluated in the RAGNAR trial, in which erdafitinib achieved an ORR of approximately 30% across multiple tumor types with predefined FGFR alterations (50). Breast cancer representation in this dataset remains limited. Neuregulin-1 fusions represent a validated target, with zenocutuzumab demonstrating an ORR of 30% across tumor types in the eNRGy study and receiving FDA approval for NRG1 fusion-positive non-small-cell lung cancer and pancreatic cancer (50). Breast cancer representation was minimal, with limited responses observed. HER3-directed antibody-drug conjugates, including patritumab deruxtecan, have demonstrated activity in metastatic breast cancer across subtypes, with ORRs ranging from approximately 22% to over 50% in early-phase and phase 2 studies (50). Ongoing pan-tumor trials are evaluating these agents across multiple histologies.

Resistance-directed therapy represents a defined area of therapeutic development. Next-generation TRK inhibitors have demonstrated activity in tumors with acquired resistance to first-generation agents. Repotrectinib is approved for NTRK fusion-positive tumors and retains activity in the post-TRK inhibitor setting (45). Selitrectinib remains investigational and has demonstrated activity in patients with solvent-front and gatekeeper resistance mutations (50). Resistance mutations account for the majority of acquired resistance to first-generation TRK inhibitors, supporting sequential treatment strategies (50). Similar approaches are under development for RET-directed therapies.

Additional emerging strategies include targeting rare hypermutated states and tumor suppressor alterations. DNA polymerase epsilon and delta proofreading mutations generate ultrahigh TMB and may confer sensitivity to immune checkpoint inhibition independent of dMMR (50). Pharmacologic reactivation of tumor suppressor mutations represents a novel therapeutic paradigm. Rezatapopt, a selective p53 reactivator, has demonstrated early clinical activity across multiple tumor types, including breast cancer, in patients with TP53 Y220C mutations (50). This mutation occurs in approximately 1% of solid tumors and is currently being evaluated in phase 2 studies.

Optimization of treatment strategies requires rational sequencing and combination approaches. Sequential use of next-generation inhibitors following resistance to first-generation agents represents a clinically supported strategy in kinase-driven tumors (45). Integration of immune checkpoint inhibitors in hypermutated tumors and combination strategies with endocrine or PI3K pathway-directed therapies represent potential approaches to improve response durability. Genomically matched therapy has demonstrated improved PFS in biomarker-selected populations in studies such as SAFIR02-BREAST, supporting structured frameworks for clinical decision-making (50).

Equitable access to molecular diagnostics remains a major barrier to implementation. Utilization of next-generation sequencing is inconsistent across institutions, with disparities related to socioeconomic status, race, and insurance coverage (50). Patients with limited access to genomic testing experience delayed or absent identification of actionable biomarkers. Expansion of testing infrastructure, improved reimbursement, and standardized clinical workflows are required to address these disparities and ensure appropriate patient selection.

Tumor-agnostic therapies will remain niche interventions in breast cancer without substantial expansion of the evidence base. Current data include fewer than 30 breast cancer patients across all approved tumor-agnostic indications and do not define efficacy, durability, or safety in this population (22). Progress will depend on generation of histology-specific evidence within tumor-agnostic frameworks, expansion of molecular testing, and integration of precision oncology into routine care.

Conclusions

Tumor-agnostic therapies have redefined precision oncology by enabling treatment selection based on shared molecular alterations rather than tissue of origin (1,2). Across nine FDA-approved indications spanning immune checkpoint inhibitors, kinase inhibitors, and antibody-drug conjugates, these therapies have demonstrated durable and clinically meaningful responses in biomarker-defined populations (3,4,9,11). In breast cancer, clinical impact is constrained by the low prevalence of actionable tumor-agnostic biomarkers and limited representation of breast tumors in registrational datasets (22).

Available evidence supports a selective, guideline-informed approach to clinical implementation. Comprehensive next-generation sequencing is recommended in patients with metastatic or treatment-refractory breast cancer to identify rare but actionable alterations, including MSI-H or dMMR, TMB-high status, NTRK and RET fusions, BRAF V600E mutations, and HER2 overexpression (21,50). These biomarkers occur in a small fraction of cases and are most frequently identified in advanced disease settings (4,8,10). When detected, they may support use of tumor-agnostic therapies, although clinical decision-making must account for the limited quantity and granularity of breast-specific evidence (21).

A consistent pattern is observed across tumor-agnostic approvals. Pembrolizumab for MSI-H/dMMR tumors, pembrolizumab for TMB-high tumors, and dabrafenib plus trametinib for BRAF V600E tumors included zero breast cancer patients in pivotal datasets (13,21,34). Dostarlimab included one patient, selpercatinib included two, and repotrectinib included three (31,41,45). Entrectinib included six patients, and larotrectinib included fourteen (22). Across all nine approvals, fewer than thirty breast cancer patients are represented, and no approval provides breast cancer-specific PFS, OS, or central nervous system outcomes (22). Estimates of efficacy, durability, and safety in breast cancer remain imprecise and are derived primarily from extrapolation across tumor types.

Tumor-agnostic activity is biologically valid but not fully histology-independent. Treatment response is influenced by tumor lineage, co-mutation patterns, and tumor microenvironment (12). NTRK fusion-positive secretory breast carcinoma demonstrates high sensitivity to TRK inhibition due to the near-universal presence of ETV6-NTRK3 fusions, while non-secretory breast cancers harbor heterogeneous alterations with more variable responses (9,25). Hypermutated tumors may respond to immune checkpoint inhibition, although mechanisms differ between dMMR and APOBEC-driven mutagenesis in breast cancer (4,12). These findings support biological plausibility while reinforcing the need for cautious interpretation in the absence of robust disease-specific data.

Future progress depends on generation of breast-specific evidence and improved implementation of molecular diagnostics. Histology-enriched cohorts within basket trials, prospective registry studies, and biomarker-driven clinical trials are required to define outcomes in rare subsets such as NTRK, RET, and BRAF V600E (9,10). Standardization of molecular diagnostics, particularly for TMB and gene fusion detection, is necessary to ensure consistent patient selection across platforms (50). Expanded access to genomic profiling and integration of multidisciplinary molecular tumor boards are essential for appropriate clinical application (21).

Tumor-agnostic therapies represent a biologically sound and clinically relevant strategy for rare molecular subsets of breast cancer. Current evidence remains limited, with fewer than thirty breast cancer patients represented across all approvals and no randomized data specific to this disease. Integration into breast oncology requires careful patient selection, multidisciplinary interpretation, and continued evidence generation. The role of tumor-agnostic therapies in breast cancer will remain defined by rarity until histology-specific data and broader access to molecular diagnostics are established.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-2025-1-69/rc

Peer Review File: Available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-2025-1-69/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-2025-1-69/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Wu S, Thawani R. Tumor-agnostic therapy: A paradigm shift in precision oncology. Cancers 2025;17:801. [Crossref] [PubMed]
2. Kim J, Kim HS, Nam M, et al. Tissue-agnostic biomarkers in solid tumors: current approvals and emerging candidates. Cancer Metastasis Rev 2025;44:58. [Crossref] [PubMed]
3. Marcus L, Lemery SJ, Keegan P, et al. FDA Approval Summary: Pembrolizumab for the Treatment of Microsatellite Instability-High Solid Tumors. Clin Cancer Res 2019;25:3753-8. [Crossref] [PubMed]
4. Marabelle A, Fakih M, Lopez J, et al. Association of tumour mutational burden with outcomes in patients with advanced solid tumours treated with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-158 study. Lancet Oncol 2020;21:1353-65. [Crossref] [PubMed]
5. Woodcock J, LaVange LM. Master Protocols to Study Multiple Therapies, Multiple Diseases, or Both. N Engl J Med 2017;377:62-70. [Crossref] [PubMed]
6. Park JJH, Siden E, Zoratti MJ, et al. Systematic review of basket trials, umbrella trials, and platform trials: a landscape analysis of master protocols. Trials 2019;20:572. [Crossref] [PubMed]
7. Bai J, Gao Y, Zhang G. The treatment of breast cancer in the era of precision medicine. Cancer Biol Med 2025;22:322-47. [Crossref] [PubMed]
8. Cortes-Ciriano I, Lee S, Park WY, et al. A molecular portrait of microsatellite instability across multiple cancers. Nat Commun 2017;8:15180. [Crossref] [PubMed]
9. Drilon A, Laetsch TW, Kummar S, et al. Efficacy of Larotrectinib in TRK Fusion-Positive Cancers in Adults and Children. N Engl J Med 2018;378:731-9. [Crossref] [PubMed]
10. Subbiah V, Wolf J, Konda B, et al. Tumour-agnostic efficacy and safety of selpercatinib in patients with RET fusion-positive solid tumours other than lung or thyroid tumours (LIBRETTO-001): a phase 1/2, open-label, basket trial. Lancet Oncol 2022;23:1261-73. [Crossref] [PubMed]
11. Meric-Bernstam F, Makker V, Oaknin A, et al. Efficacy and Safety of Trastuzumab Deruxtecan in Patients With HER2-Expressing Solid Tumors: Primary Results From the DESTINY-PanTumor02 Phase II Trial. J Clin Oncol 2024;42:47-58. [Crossref] [PubMed]
12. Adams S, Gatti-Mays ME, Kalinsky K, et al. Current Landscape of Immunotherapy in Breast Cancer: A Review. JAMA Oncol 2019;5:1205-14. [Crossref] [PubMed]
13. US Food and Drug Administration. Keytruda (pembrolizumab) prescribing information. Updated 2026. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=125514
14. Le DT, Durham JN, Smith KN, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science 2017;357:409-13. [Crossref] [PubMed]
15. Maio M, Ascierto PA, Manzyuk L, et al. Pembrolizumab in microsatellite instability high or mismatch repair deficient cancers: updated analysis from the phase II KEYNOTE-158 study. Ann Oncol 2022;33:929-38. [Crossref] [PubMed]
16. Marabelle A, O'Malley DM, Hendifar AE, et al. Pembrolizumab in microsatellite-instability-high and mismatch-repair-deficient advanced solid tumors: updated results of the KEYNOTE-158 trial. Nat Cancer 2025;6:253-8. [Crossref] [PubMed]
17. Chakravarty D, Gao J, Phillips SM, et al. OncoKB: A Precision Oncology Knowledge Base. JCO Precis Oncol 2017;1:1-16. [Crossref] [PubMed]
18. Cheng AS, Leung SCY, Gao D, et al. Mismatch repair protein loss in breast cancer: clinicopathological associations in a large British Columbia cohort. Breast Cancer Res Treat 2020;179:3-10. [Crossref] [PubMed]
19. Horimoto Y, Thinzar Hlaing M, Saeki H, et al. Microsatellite instability and mismatch repair protein expressions in lymphocyte-predominant breast cancer. Cancer Sci 2020;111:2647-54. [Crossref] [PubMed]
20. Ren XY, Song Y, Wang J, et al. Mismatch Repair Deficiency and Microsatellite Instability in Triple-Negative Breast Cancer: A Retrospective Study of 440 Patients. Front Oncol 2021;11:570623. [Crossref] [PubMed]
21. Henry NL, Somerfield MR, Dayao Z, et al. Biomarkers for Systemic Therapy in Metastatic Breast Cancer: ASCO Guideline Update. J Clin Oncol 2022;40:3205-21. [Crossref] [PubMed]
22. US Food and Drug Administration. Vitrakvi (larotrectinib) prescribing information. Updated 2026. Accessed June
23. 2026. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=210861
24. Hong DS, DuBois SG, Kummar S, et al. Larotrectinib in patients with TRK fusion-positive solid tumours: a pooled analysis of three phase 1/2 clinical trials. Lancet Oncol 2020;21:531-40. [Crossref] [PubMed]
25. Yoshino T, Pentheroudakis G, Mishima S, et al. JSCO—ESMO—ASCO—JSMO—TOS: international expert consensus recommendations for tumour-agnostic treatments in patients with solid tumours with microsatellite instability or NTRK fusions. Ann Oncol 2020;31:861-72. [Crossref] [PubMed]
26. Maund SL, Sokol ES, Ang Houle A, et al. NTRK gene fusions are detected in both secretory and non-secretory breast cancers. Pathol Int 2022;72:187-92. [Crossref] [PubMed]
27. US Food and Drug Administration. Rozlytrek (entrectinib) prescribing information. Updated 2025. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=212725
28. Doebele RC, Drilon A, Paz-Ares L, et al. Entrectinib in patients with advanced or metastatic NTRK fusion-positive solid tumours: integrated analysis of three phase 1-2 trials. Lancet Oncol 2020;21:271-82. [Crossref] [PubMed]
29. Marcus L, Donoghue M, Aungst S, et al. FDA Approval Summary: Entrectinib for the Treatment of NTRK gene Fusion Solid Tumors. Clin Cancer Res 2021;27:928-32. [Crossref] [PubMed]
30. Barroso-Sousa R, Jain E, Cohen O, et al. Prevalence and mutational determinants of high tumor mutation burden in breast cancer. Ann Oncol 2020;31:387-94. [Crossref] [PubMed]
31. Alva AS, Mangat PK, Garrett-Mayer E, et al. Pembrolizumab in Patients With Metastatic Breast Cancer With High Tumor Mutational Burden: Results From the Targeted Agent and Profiling Utilization Registry (TAPUR) Study. J Clin Oncol 2021;39:2443-51. [Crossref] [PubMed]
32. US Food and Drug Administration. Jemperli (dostarlimab) prescribing information. Updated 2025. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=761174
33. Oaknin A, Tinker AV, Gilbert L, et al. Clinical Activity and Safety of the Anti-PD-1 Monoclonal Antibody Dostarlimab for Patients with Recurrent or Advanced dMMR Endometrial Cancer. Future Oncology 2021;17:3781-5. [Crossref] [PubMed]
34. US Food and Drug Administration. Tafinlar (dabrafenib) prescribing information. Updated 2025. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=202806
35. US Food and Drug Administration. Mekinist (trametinib) prescribing information. Updated 2025. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=217513
36. Subbiah V, Kreitman RJ, Wainberg ZA, et al. Dabrafenib plus trametinib in BRAFV600E-mutated rare cancers: the phase 2 ROAR trial. Nat Med 2023;29:1103-12. [Crossref] [PubMed]
37. Salama AKS, Li S, Macrae ER, et al. Dabrafenib and Trametinib in Patients With Tumors With BRAFV600E Mutations: Results of the NCI-MATCH Trial Subprotocol H. J Clin Oncol 2020;38:3895-904. [Crossref] [PubMed]
38. Owsley J, Stein MK, Porter J, et al. Prevalence of class I-III BRAF mutations among 114,662 cancer patients in a large genomic database. Exp Biol Med (Maywood) 2021;246:31-9. [Crossref] [PubMed]
39. Hyman DM, Puzanov I, Subbiah V, et al. Vemurafenib in Multiple Nonmelanoma Cancers with BRAF V600 Mutations. N Engl J Med 2015;373:726-36. [Crossref] [PubMed]
40. US Food and Drug Administration. Retevmo (selpercatinib) prescribing information. Updated 2025. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=213246
41. Duke ES, Bradford D, Marcovitz M, et al. FDA Approval Summary: Selpercatinib for the Treatment of Advanced RET Fusion-Positive Solid Tumors. Clin Cancer Res 2023;29:3573-8. [Crossref] [PubMed]
42. Ross DS, Gay LM, Elvin JA, et al. Comprehensive genomic profiling of 8,654 breast carcinoma reveals therapeutically targetable molecular subtypes beyond those defined by hormone-receptor expression. Ann Oncol 2016;27:VI70. [Crossref]
43. Paratala BS, Chung JH, Williams CB, et al. RET rearrangements are actionable alterations in breast cancer. Nat Commun 2018;9:4821. [Crossref] [PubMed]
44. US Food and Drug Administration. Augtyro (repotrectinib) prescribing information. Updated 2024. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2023/218213Orig1s000TOC.html
45. Drilon A, Camidge DR, Lin JJ, et al. Repotrectinib in ROS1 Fusion-Positive Non-Small-Cell Lung Cancer. N Engl J Med 2024;390:118-31. [Crossref] [PubMed]
46. Besse B, Lin JJ, Bazhenova L, et al. Repotrectinib in NTRK fusion-positive advanced solid tumors: a phase 1/2 trial. Nat Med 2026;32:682-9. [Crossref] [PubMed]
47. US Food and Drug Administration. Enhertu (trastuzumab deruxtecan) prescribing information. Updated 2025. Accessed June 22, 2026. Available online: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/761139Orig1s000TOC.html
48. Smit EF, Felip E, Uprety D, et al. Trastuzumab deruxtecan in patients with metastatic non-small-cell lung cancer (DESTINY-Lung01): primary results of the HER2-overexpressing cohorts from a single-arm, phase 2 trial. Lancet Oncol 2024;25:439-54. [Crossref] [PubMed]
49. Raghav K, Siena S, Takashima A, et al. Trastuzumab deruxtecan in patients with HER2-positive advanced colorectal cancer (DESTINY-CRC02): primary results from a multicentre, randomised, phase 2 trial. Lancet Oncol 2024;25:1147-62. [Crossref] [PubMed]
50. Modi S, Jacot W, Yamashita T, et al. Trastuzumab Deruxtecan in Previously Treated HER2-Low Advanced Breast Cancer. N Engl J Med 2022;387:9-20. [Crossref] [PubMed]
51. Mosele F, Remon J, Mateo J, et al. Recommendations for the use of next-generation sequencing (NGS) for patients with metastatic cancers: a report from the ESMO Precision Medicine Working Group. Ann Oncol 2020;31:1491-505. [Crossref] [PubMed]