Advancing precision medicine in patients with metastatic breast cancer: a narrative review on the role of multi-gene profiling and molecular tumor board

Introduction

Background

Metastatic breast cancer (mBC) remains one of the leading causes of cancer-related mortality worldwide and continues to represent a significant clinical challenge for the global oncology community (1,2).

Although substantial progress has been achieved in early detection and the treatment of early-stage disease, mBC is still considered incurable, and therapeutic management in this setting focuses on prolonging survival while preserving quality of life (3).

The current therapeutic landscape includes chemotherapy, endocrine therapy, targeted agents, and immunotherapy. Treatment decisions are primarily guided by the tumor’s molecular characteristics, particularly hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) status, allowing for the classification of mBC into three major biological subtypes: HR⁺/HER2^–, HER2⁺, and triple-negative (TN) BC. Each subtype is managed according to distinct therapeutic algorithms tailored to its molecular profile (4).

Despite therapeutic advances, many patients eventually experience disease progression, underscoring the urgent need for more effective and personalized treatment strategies (5).

In this context, precision medicine has emerged as a promising approach that integrates genetic, environmental, and lifestyle factors to guide the prevention and treatment of disease (6).

New clinical trial designs, such as basket, umbrella, and platform trials, aim to improve the efficiency of developing personalized therapies based on genomic information. Basket trials test a single drug in patients with different tumor histologies that share the same molecular target. Conversely, umbrella trials evaluate multiple drugs within a single tumor type, stratifying patients into subgroups according to their specific molecular targets. Platform trials are adaptive study designs that allow the evaluation of multiple drugs targeting different molecular alterations in each disease, with the flexibility to add or remove drugs based on interim results (7).

Rationale and knowledge gap

The traditional “one-size-fits-all” approach has proven inadequate for addressing the biological complexity and heterogeneity of mBC. The emergence of next-generation sequencing (NGS) technologies and the identification of novel molecular targets have laid the foundation for personalized treatment strategies, enabling the transition toward precision oncology (8). Despite these advances, the implementation of precision medicine in routine clinical practice remains limited, hindered by multiple organizational, regulatory, and economic barriers (9). These challenges include limited access to genomic testing, insufficient integration of molecular tumor boards (MTBs), and disparities in reimbursement and infrastructure across healthcare systems.

To address these limitations, emerging models of precision oncology, including Precision Pro, Dynamic Precision, and Intelligent Precision, are redefining paradigms of clinical research and care delivery. Precision Pro, Dynamic Precision, and Intelligent Precision represent advanced iterations of the new experimental trial models. The basket trial model aligns with the concept of Intelligent Precision, which leverages artificial intelligence (AI) and big data to analyze large-scale genomic and clinical datasets. This approach refines therapeutic decision-making by shifting from treating groups of tumors sharing the same genomic alteration to employing predictive algorithms that identify individual patients most likely to benefit from specific therapies. The umbrella trial model corresponds to Dynamic Precision, introducing flexibility in clinical trials by allowing treatment selection to adapt in real time to emerging genomic data. Finally, the platform trial approach is comparable to Precision Pro, which aims to optimize treatment selection through increasingly refined patient stratification using multi-omic approaches, including genomics, proteomics, metabolomics, and epigenetics. Integrating these three approaches has the potential to revolutionize disease management by accelerating the transition from static models to personalized, dynamic, and intelligent systems (7). Recent prospective studies have further highlighted the clinical utility of multi-omic profiling in guiding treatment decisions for heavily pretreated mBC patients. In particular, Pierobon et al. demonstrated that a multi-omic approach can effectively inform therapeutic selection in patients with refractory disease, supporting the practical application of precision oncology in complex clinical settings (10).

Objective

Despite the promising outlook of precision medicine, a substantial knowledge gap persists regarding the real-world implementation of multigene genomic profiling in patients with mBC, as well as its actual impact on clinical outcomes (11).

Although genomic and transcriptomic analyses have significantly advanced our understanding of tumor biology, translating these findings into actionable clinical strategies is still inconsistent across clinical settings.

In this context, MTBs, multidisciplinary teams dedicated to integrating molecular data with clinical decision-making, have emerged as pivotal enablers of precision oncology. However, their structure, workflow, and clinical impact remain heterogeneous, underscoring the need for further consolidation and standardization to ensure consistent and equitable access to innovative targeted therapies (12,13). Numerous challenges continue to hinder the widespread adoption of precision oncology approaches, including long turnaround times for molecular test results, limited availability and reimbursement of targeted drugs, infrastructural disparities across healthcare systems, and the fragmentation of clinical and genomic data integration (14).

Although recent systematic reviews and real-world evidence (RWE) have confirmed that MTBs can enhance treatment appropriateness and, to a lesser extent, improve clinical outcomes, their overall impact remains limited and inconsistent across institutions (15,16). These findings highlight the urgent need for robust, scalable, and sustainable frameworks to support clinical decision-making and ensure timely access to personalized treatment options. Establishing a comprehensive and adaptive precision oncology infrastructure is essential to accelerate the transition toward targeted, dynamic, and intelligent care models in mBC.

The aim of this article is to provide an up-to-date and critical narrative review of the application of precision oncology in mBC. Specifically, we examine the therapeutic opportunities enabled by genomic profiling, the evolving role of MTBs in supporting integrated clinical decisions, and the major barriers limiting their broader implementation. Finally, we discuss innovative strategies designed to overcome these obstacles, with the goal of promoting equitable and timely access to personalized treatments and advancing the transition toward increasingly targeted, dynamic, and intelligent oncology care models. We present this article in accordance with the Narrative Review reporting checklist (available at https://bc.amegroups.com/article/view/10.21037/bc-25-18/rc).

Methods

We performed a narrative review based on a critical and up-to-date synthesis of the literature on precision oncology in mBC (Table 1). A comprehensive search was conducted in PubMed, Embase, medRxiv, major oncology society guidelines and clinicaltrial.gov up to April 2025. Search terms included “metastatic breast cancer”, “precision oncology”, “multi-gene profiling”, “Molecular Tumor Board”, and related synonyms. We included peer-reviewed articles, clinical trials, and guidelines published in English from 2013 to 2025, focusing on clinical applications, barriers, and strategies for implementation. Case reports and editorials were excluded. Two reviewers independently screened titles and abstracts, with full-text review of selected articles; discrepancies were resolved by consensus. References of included articles were also screened for additional studies. An initial search yielded approximately 3,000 articles; after applying predefined eligibility criteria, 52 studies were retained for inclusion in this review.

Precision oncology in mBC: current landscape and future perspectives

Tumor-agnostic therapies and European Society for Medical Oncology (ESMO) frameworks

In recent years, some drugs have been approved according to a tumor-agnostic model, where efficacy is independent of tumor type. The ESMO Precision Medicine Working Group (PMWG) has proposed a classification of molecularly guided treatment options (MGTO) based on therapeutic effect against driver molecular alterations, identifying three categories: tumor-agnostic, effective regardless of tumor type (e.g. TRK inhibitors in tumors harboring NTRK gene fusions); tumor-modulated, efficacy partially dependent on tumor type (e.g. PARP inhibitors in tumors harboring BRCA1/2 mutation/homologous recombination deficiency); and tumor-limited, efficacy restricted to specific tumor types (e.g. PI3K inhibitors in BC with PIK3CA mutation). The ESMO Tumour-Agnostic Classifier and Screener (ETAC-S) is a tool that assesses the agnostic potential of MGTO, and is based on 3 minimum requirements: therapy must have shown an objective response in at least one in five patients [objective response rate (ORR) ≥20%] in two-thirds of the tumor types studied (and in ≥4 tumor types), with at least five evaluable patients in each tumor type, in the setting of refractory disease (17).

The role and workflow of MTBs

The advent of comprehensive genomic profiling and MGTO has led to the establishment of MTBs.

MTBs are multidisciplinary teams that integrate genomic, genetic, and clinical data to formulate evidence-based therapeutic recommendations. They typically include oncologist, surgeon, geneticist, pathologist, pharmacist, radiologist or radiation oncologist, molecular biologist, data manager, research nurse, bioinformatician, ethicist, and the patient or patient representative (12,18).

The medical oncologist plays a central role by contextualizing molecular findings within the patient’s clinical background, including prior therapies, disease trajectory, and comorbidities. The pathologist ensures the adequacy and integrity of tumor samples and provides essential histopathological context, which often influences the interpretation of genomic results. Given the complexity of NGS data, the bioinformatician is indispensable in processing and filtering raw data, prioritizing clinically relevant alterations, and distinguishing actionable mutations from variants of uncertain significance (VUS). The molecular biologist supports the technical interpretation of biomarker findings and provides insights into potential limitations or artifacts related to the testing platform. The clinical geneticist contributes by interpreting germline variants, clarifying their implications in the context of hereditary cancer syndromes, and offering guidance for both patients and their relatives. Pharmacists contribute by assessing the availability of targeted therapies, considering regulatory approvals, off-label use, and access through expanded access programs, and by evaluating potential drug interactions. Data managers and research nurses ensure that clinical data are accurately recorded and that recommendations are traceable over time. Moreover, the inclusion of ethicists or patient representatives helps ensure that ethical considerations and patient values are embedded in therapeutic decision-making (19).

The MTB process typically unfolds in distinct, yet interrelated, phases. It begins with comprehensive data collection, encompassing clinical, radiological, histological, and genomic information, obtained from either tissue or liquid biopsies as appropriate. Once molecular results are available, the interpretation phase follows, applying validated frameworks to assess clinical actionability. The ESMO Scale for Clinical Actionability of molecular Targets (ESCAT) provides a widely accepted system for ranking genomic alterations according to the strength of supporting clinical evidence (20). More recently, the ETAC-S has been introduced to evaluate the tumor-agnostic potential of molecularly guided therapies and inform drug development strategies (17). Following molecular interpretation, the multidisciplinary team convenes to evaluate all viable treatment options, including approved drugs, off-label therapies supported by clinical or preclinical evidence, or potential enrolment in clinical trials such as basket, umbrella, or platform trials. Once consensus is reached, the MTB formulates a treatment recommendation, which is then communicated to the treating oncologist and discussed with the patient. A hallmark of the MTB approach is its dynamic and adaptive nature (Figure 1) (21). Recommendations are not fixed, but are subject to revision in response to disease progression, new molecular findings, or emerging clinical evidence. This continuous re-evaluation supports a model of adaptive oncology and contributes to real-world learning by linking therapeutic decisions to patient outcomes. Digital tools can enhance MTB efficiency, enabling remote collaboration and the development of virtual MTBs (VMTBs) (22).

Figure 1 Workflow of MTB. MTB, molecular tumor board.

Key pillars for the development and sustainability of MTBs include technical solutions for data visualization and interpretation; interoperability, through the creation of platforms for storing and exchanging genomic and clinical data across different institutions; continuous learning mechanisms, involving the systematic collection of real-world data on treatment outcomes to improve future therapeutic decisions; access to clinical trials; legal considerations; patient selection criteria; standardization of decisions (e.g., ESCAT); official recognition; local leadership; and the establishment of international MTB networks (14).

Tissue and liquid biopsy in genomic profiling

MTBs also play a critical role in selecting the most appropriate method for identifying molecular targets, taking into account factors such as tissue availability, sample quality, and the type of genomic alteration. While tumor tissue remains the standard source for molecular profiling, liquid biopsy, particularly the analysis of circulating tumor DNA (ctDNA), has emerged as a valuable complementary tool, especially when tissue is insufficient or inaccessible. Notably, liquid biopsy may offer advantages in capturing tumor heterogeneity and detecting mechanisms of acquired resistance, particularly in patients experiencing disease progression during or after targeted therapies (19,23). In a study by Iams et al. involving a cohort of patients with mBC, the concordance rate for variant detection between tissue and blood samples was 59.4%. Interestingly, 19.7% of variants were detected exclusively through ctDNA, while 20.9% were uniquely identified in tumor tissue. These findings highlight the complementary nature of the two approaches and suggest that the combined use of tissue and liquid biopsy may provide a more comprehensive assessment of the tumor’s genomic landscape (24).

Actionable genomic alterations in mBC

Ultimately, the integration of genomic profiling into the management of mBC aims to translate molecular findings into actionable clinical decisions. Several genomic alterations, such as those involving PIK3CA, BRCA1/BRCA2 and ESR1, are now well-established biomarkers that guide therapeutic selection and inform treatment sequencing in this setting. Once a molecular alteration is identified, its clinical value must be defined. The ESCAT ranks alterations from tier I (clinically validated and approved) to tier X (no evidence) (20,25).

The management of mBC treatment is reported in Tables 2-4, according to ESMO Guidelines (26).

The integration of molecular profiling into the multidisciplinary management of mBC remains a central challenge in advancing precision oncology. In this context, the use of sequential or combinatorial targeted therapies requires a structured framework to prioritize clinically actionable genomic alterations, taking into account both variant clonality and interactions within relevant biological pathways. For instance, in patients with HR⁺/HER2^– mBC progressing after endocrine therapy plus CDK4/6 inhibitors, the identification of mutations in PIK3CA, ESR1, or alterations in PTEN/AKT pathways can guide the use of agents such as alpelisib, capivasertib, or selective estrogen receptor degraders (SERDs) (8,27-29). The choice between combination or sequential strategies depends not only on clinical evidence but also on co-occurring alterations that may influence therapeutic resistance, such as PTEN loss, a well-recognized mechanism of resistance to PI3K inhibitors (30). Liquid biopsy provides an additional layer of insight by capturing the dynamic clonal evolution of tumors in real time. Recent data suggest that approximately 20% of genomic alterations are detectable exclusively through ctDNA, highlighting the value of integrating liquid and tissue-based analyses for comprehensive tumor characterization (24). This approach is particularly relevant in scenarios requiring rapid therapeutic decision-making, such as combining a SERD in response to emerging ESR1 mutations with a PI3K or AKT inhibitor in the presence of concurrent alterations (27-29).

Tumor mutational burden (TMB) and immunotherapy

Among tumor-agnostic alterations, high TMB (TMB-H; ≥10 Mut/Mb) can confer eligibility for programmed death-1 (PD-1)/programmed death ligand-1 (PD-L1) checkpoint inhibitors, such as pembrolizumab, although TMB-H does not always predict immunotherapy response, and tissue-based TMB assessment is preferred for reliability (8). The USA Food and Drug Administration (FDA) has approved pembrolizumab for the treatment of all solid tumors with TMB-H. This approval was based on an exploratory analysis of the KEYNOTE-158 study, a single-arm, multi-cohort phase II study of pembrolizumab for previously treated, advanced solid tumors of any histology with TMB-H. Although the FDA approval applies to all solid tumor histologies with TMB-H, it should be noted that the patient population considered in the KEYNOTE-158 study did not include patients with BC.

Large-scale molecular screening programs and RWE

The Phase II TAPUR basket study evaluated pembrolizumab in a cohort of 28 patients with mBC exhibiting TMB-H. Among the enrolled patients, disease control and objective response were observed in 37% and 21%, respectively. The median progression-free survival (PFS) was 10.6 weeks, and the median overall survival (OS) was 30.6 weeks. No correlation was observed between PFS and the degree of TMB. These findings indicate that pembrolizumab demonstrates antitumor activity in heavily pretreated mBC patients with TMB-H. The results support the recent FDA approval of pembrolizumab for the treatment of patients with unresectable or metastatic solid tumors characterized by TMB-H, who have no satisfactory alternative treatment options. This approval encompasses both HR-positive and TNBCs with TMB-H (30).

In contrast, the study by McGrail et al. reported that TMB-H failed to predict response to immune checkpoint blockade in mBC. Specifically, TMB-H tumors achieved an overall response rate of 15.3%, significantly lower than that of low TMB tumors (odds ratio =0.46). Comparable results were observed when analyzing OS, further confirming the limited predictive value of TMB in this cohort (31).

Thus, TMB-H does not consistently predict response to immunotherapy. For instance, temozolomide, an alkylating agent, can induce a high mutational burden; however, glioma patients with TMB-H following temozolomide treatment do not respond to PD-1 inhibitors (32).

Moreover, genomic analysis aimed at assessing TMB from liquid biopsy may detect mutations arising from clonal hematopoiesis, extra-tumor sources, or factors such as smoking and aging. These alterations should be carefully filtered to avoid misleading results (33). Tissue-based TMB assessment is generally preferred for its higher reliability (34).

More complex scenarios arise when NGS reports reveal non-actionable genomic alterations (“private genomic alterations”) that must still be documented, as they may become clinically relevant in the future. In assessing the utility of broad molecular profiling in BC, molecular screening programs play a crucial role by systematically collecting and interpreting these alterations.

AURORA (Aiming to Understand the Molecular Aberrations in mBC), led by the Breast International Group (BIG), aims to enhance the understanding of mBC through comprehensive profiling of primary tumors, metastatic samples, and circulating cell-free DNA (cfDNA) extracted from plasma, collected from at least 1,000 patients. Analysis of the first 381 patients, either treatment-naïve for mBC or after only one line of therapy, revealed that most driver mutations were shared between primary and metastatic sites (88%) while only a minority of patients (10%) harbored at least one exclusive mutation in the metastatic sample. This contrasts with other studies that collected paired samples and reported a higher rate of acquired genomic alterations ranging from 45% to 73%. Recent evidence suggests that mutations in metastases are less associated with metastatic dissemination and tend to accumulate under the pressure of oncological therapies. Although more than 50% of patients exhibited ESCAT level I or II alteration, only 7% of patients in this cohort received therapy matched to their molecular profile (11). Similarly, the SAFIR02-BREAST study, another screening program, enrolled endocrine-resistant patients who had received at most one previous line of chemotherapy for the metastatic setting. Patients were required to have a tissue biopsy within 12 months of enrollment or, alternatively, a liquid biopsy. Among 1,462 patients who underwent genomic profiling, 238 were randomized to receive a targeted therapy based on their molecular profile (n=75 of 115 patients with ESCAT I/II alterations) or maintenance chemotherapy (n=40 of 115) (35).

The ROME study is a randomized, multi-basket, phase II study evaluating the feasibility, efficacy, and safety of targeted therapy versus standard of care in patients with solid tumors harboring actionable alterations identified through comprehensive genomic profiling and discussed by MTB. Following MTB review, patients with at least one actionable alteration were randomized 1:1 to targeted therapy at the MTB’s choice or standard therapy at the investigator’s choice. Overall, more than 1,700 patients were screened in 40 Italian centers: 897 had genomic alterations discussed at the MTB and 400 patients (22.5%) were randomized (targeted therapy/standard of care =200/200). Thirty-eight different histologies were enrolled, including BC (10%) (36). Preliminary results from the mBC cohort of the ROME trial were reported at the 2022 San Antonio Breast Cancer Symposium (SABCS). Complete mutational screening data are available for 62 patients with mBC (63% HR⁺/HER2⁻, 35% TN, 2% HR⁻/HER2⁺). Actionable mutations were identified in 34 patients (54%) and 28 patients (45%) were assigned to receive a targeted therapy following MTB discussion. The MTB recommended germline testing for 6 patients: 4 were confirmed to carry germline mutations (66%; 2 BRCA, 1 PALB2, 1 BRIP1), including cases with no prior indication for germline testing (37). Previously, Yadav et al. showed that the frequency of pathogenic variants in 9 BC related genes (ATM, BRCA1, BRCA2, CDH1, CHEK2, NF1, PALB2, PTEN, and TP53) in patients without first- to second-degree relatives with BC is close to 5%. This prompted the push to extend the National Comprehensive Cancer Network (NCCN) criteria for genetic testing to include all women diagnosed with BC at ≤65 years of age (38).

MTBs play a critical role in integrating molecular insights with clinical judgment, balancing efficacy, biological rationale, and potential toxicity (12,14).

Barriers to implementation and healthcare system strategies

Despite the growing availability of molecular profiling, data from large-scale screening initiatives such as AURORA, SAFIR02-BREAST, and the ROME trial, have shown that only a minority of patients with mBC actually receive genomically matched therapies, even when harboring ESCAT I/II alterations (11,35,36). Contributing factors include limited drug access, long turnaround times, and regulatory or reimbursement barriers (9,39). To address these limitations, several innovative solutions have been proposed, including VMTBs, AI-driven trial matching algorithms, and personalized reimbursement models, all designed to enhance the feasibility and timeliness of precision-based interventions (9,22,33). Therefore, the successful integration of precision oncology in mBC depends on the clear identification of actionable genomic targets and the removal of structural and regulatory barriers that hinder their implementation. Experiences from various healthcare systems suggest that combining clinical trial integration, early access programs, standardized recommendations, and real-world data collection can help ensure equitable and timely access to personalized therapies for patients with advanced disease. Molecular screening programs (e.g., AURORA, SAFIR02-BREAST, ROME trial) show that only a minority of patients receive genomics-matched therapies, due to limited availability of targeted therapies, access barriers, regulatory and reimbursement issues, and delays in testing and decision-making. Thus, the clinical implementation of precision oncology faces several barriers, including limited reimbursement for NGS testing and off-label therapies, complex regulatory procedures, and restricted access to clinical trials (14). Proposed solutions include optimizing turnaround times for genomic testing and MTB decisions (e.g., liquid biopsy, VMTBs), facilitating access to targeted drugs and clinical trials (e.g., personalized reimbursement, expanded access programs, AI-based patient-trial matching), and streamlining regulatory processes (9).

In summary, optimizing the management of patients with mBC requires effective integration of genomic data interpretation, personalized therapeutic decisions, and access to targeted therapies and clinical trials, under the guidance of MTBs.

Various healthcare systems have implemented effective strategies to overcome these obstacles.

In Italy, the MTB recommendation may not be foreseen by the Italian Medicines Agency (AIFA). Today, the methods of access to the drug, when not indicated according to AIFA, refer to Law 648/96, Law 326/2003 art.48 (5% fund), DM 09/07 /2017 (nominal therapeutic use), Class Cnn (class C drugs, not negotiated), clinical trial and Law 94/98 (off-label use). ESMO recommends using off-label drugs combined with genomics only if they have been developed in an access program and following a decision procedure at national or regional level. But this type of indication very often remains unapplied, for economic reasons (12).

In France, national programs such as SAFIR have successfully integrated genomic screening with clinical trials, ensuring access to targeted therapies through structured study frameworks (11).

In the USA, the TAPUR registry, promoted by the American Society of Clinical Oncology (ASCO), along with expanded access programs, has enabled the use of targeted agents outside of their approved indications. These initiatives have also contributed to the generation of RWE, supporting future regulatory evaluations (30).

In Germany, the inclusion of NGS testing within public insurance reimbursement packages has improved the accessibility and equity of molecular profiling (14).

Minimal residual disease (MRD) and ctDNA-guided adjuvant strategies

In recent years, significant clinical efforts have focused on evaluating whether ctDNA representing MRD after surgery and adjuvant therapy can guide the personalized intensification or de-escalation of adjuvant treatment. Several phase II and III trials have been designed to investigate this approach. The c-TRAK-TN study, one of the first to implement serial ctDNA monitoring in patients with high-risk early-stage TNBC, demonstrated the feasibility of MRD surveillance and early detection. However, challenges related to assay sensitivity and sampling frequency limited effective intervention, with few patients receiving treatment and no sustained ctDNA clearance after immunotherapy in most cases, underscoring the complexity of translating MRD detection into immediate clinical benefit (40). The phase II randomized DARE (DNA-guided second-line Adjuvant therapy for high RIsk, NCT04567420) trial investigates an escalation strategy involving the addition of palbociclib plus fulvestrant or other systemic therapies in HR⁺/HER2⁻ patients, selected based on ctDNA positivity after adjuvant treatment. The trial aims to determine whether MRD-guided therapy reduces the risk of clinical recurrence compared to standard management (41). The international, randomized phase III trial TREAT-ctDNA/EORTC-2129-BCG (NCT05512364) enrolls patients with HR⁺/HER2⁻ BC who test positive for ctDNA after adjuvant therapy. ctDNA-positive patients are randomized to either continue standard endocrine therapy or undergo a treatment modification or escalation, such as switching to a SERD like elacestrant. This trial represents a key effort to determine whether MRD-guided therapeutic intervention can translate into clinical benefits, including reduced metastatic events and improved metastasis-free survival (42). Several observational and proof-of-concept studies using commercial assays such as Signatera™ and Guardant Reveal have demonstrated that postoperative ctDNA detection is a strong predictor of distant relapse and can provide an earlier indication than conventional imaging. This finding has prompted numerous ongoing and planned trials investigating MRD-guided treatments. However, several challenges remain in the adjuvant setting, including low and variable ctDNA levels, assay sensitivity, sampling frequency, and biological confounders such as clonal haematopoiesis, which can lead to false-positive results. Moreover, it remains unclear whether early intervention based on MRD detection improves clinical outcomes or merely anticipates relapse without altering disease course (43). Dynamic changes in ctDNA levels closely mirror tumor burden and can precede radiological response, as shown in proof-of-concept studies (44).

Early ctDNA clearance is associated with improved clinical outcomes, whereas persistent or rising ctDNA levels predict rapid progression and poorer prognosis. Thus, ctDNA serves as a versatile biomarker with both prognostic and predictive value in early and metastatic settings alike, enabling the detection of emerging resistance mechanisms (40,45).

Despite these advances, randomized trials are still needed to definitively demonstrate that treatment decisions guided solely by ctDNA monitoring can improve clinical outcomes, particularly in the adjuvant setting (43,46,47).

AI and predictive models in mBC

Future efforts should focus on integrating molecular monitoring into a multidisciplinary, dynamic framework supported by digital technologies and real-world data collection, to fully realize the promise of precision medicine for patients with mBC. In recent years, AI has increasingly contributed to precision oncology, showing significant potential in mBC. Machine learning and deep learning algorithms enhance the analysis of complex clinical and genomic data, enabling the development of predictive models for risk stratification, treatment selection, and response monitoring. Several studies have demonstrated the effectiveness of AI models in predicting metastatic progression and prognosis. Explainable AI approaches have further facilitated the identification of genomic biomarkers associated with metastasis development, employing interpretability techniques such as SHapley Additive exPlanations (SHAP) to ensure transparency and clinical validity (48).

Similarly, predictive models using retrospective patient data have anticipated the onset of de novo bone metastases at diagnosis, potentially enabling earlier therapeutic interventions (49).

Future perspectives and challenges

The available evidence confirms that precision oncology has already begun to transform the therapeutic landscape of mBC, offering more personalized and effective treatment options. However, significant barriers remain, including regulatory constraints, organizational challenges, and disparities in drug accessibility. Unlike previous reviews, this work adopts a critical and forward-looking perspective. Rather than providing a static overview of the current state of the field, it proposes innovative models of clinical research such as Precision Pro, Dynamic Precision, and Intelligent Precision, while emphasizing the central and strategic role of the MTB as a hub for clinical-genomic integration and decision-making governance.

A particular focus is placed on tumor-agnostic biomarkers, including TMB, which are examined not only for their potential to expand therapeutic opportunities but also for the limitations that currently hinder their broader clinical implementation. In addition, the paper outlines practical strategies to overcome ongoing challenges in the field. These include the use of liquid biopsy to capture tumor heterogeneity, the application of AI-driven algorithms for more efficient clinical trial matching, the development of innovative reimbursement models, and the expansion of access through dedicated regulatory programs.

In this context, the article serves as a programmatic contribution, offering a vision for the future of precision oncology that extends beyond technological and pharmacological innovation. It highlights the importance of governance, institutional collaboration, and equitable access as essential pillars for translating the promise of genomic profiling and targeted therapies into tangible, widespread benefits for patients with mBC. Only through the integration of these elements can precision medicine realize its full potential in routine clinical practice.

Strengths and limitations

This review provides a broad and clinically oriented synthesis of precision oncology in mBC, covering key areas such as multigene profiling, MTBs, liquid biopsy, tumor-agnostic biomarkers, MRD, and AI. A particular strength of the review is its forward-looking perspective, which highlights innovative frameworks and emphasizes system-level solutions to overcome barriers in implementation.

However, several limitations should be acknowledged. As a narrative rather than systematic review, it is subject to selection bias and relies on heterogeneous sources, ranging from randomized trials to observational data. Clinical outcomes such as OS and quality of life receive less emphasis compared to biomarker-driven strategies. Moreover, despite the wealth of genomic insights, real-world clinical translation remains limited, with only a minority of patients receiving genomically matched therapies.

In sum, the review offers a comprehensive and critical overview, but its conclusions should be interpreted considering the methodological constraints of a narrative approach and the ongoing challenges in clinical implementation.

Conclusions

Despite significant scientific and technological advances, the implementation of precision medicine in mBC remains limited by regulatory, organizational, and economic barriers. Inconsistent reimbursement policies, unequal access to comprehensive genomic testing, and limited availability of targeted therapies contribute to considerable disparities across treatment centers and geographic regions (9,12).

Liquid biopsy represents a promising approach to accelerate molecular profiling and dynamically monitor tumor evolution. Its minimally invasive nature and ability to capture real-time molecular changes—particularly under therapeutic pressure—make it an attractive complement to tissue-based testing. However, broader clinical adoption will require the development of standardized protocols, investment in training of healthcare personnel, and the integration of liquid biopsy into existing diagnostic pathways (19,23). MTBs are essential for the clinical interpretation of complex genomic data and for guiding personalized treatment decisions. However, their implementation remains uneven and often limited by resource constraints. The development of VMTBs and the formation of international collaborative networks could help address these gaps by facilitating the dissemination of expertise, harmonizing practices, and promoting equitable access to molecular decision-making tools (14,22).

In parallel, the integration of AI tools for automated patient-trial matching holds promises for improving clinical trial enrolment and reducing the time from molecular diagnosis to treatment initiation. Early experiences suggest that AI-driven solutions can enhance the scalability and efficiency of trial allocation in real-world settings (14).

Ultimately, the full realization of precision oncology in mBC depends not only on continued scientific innovation, but also on the adoption of innovative governance models, adaptive regulatory frameworks, and collaborative infrastructures that support data sharing and decision-making integration. A coordinated and multi-level approach, spanning healthcare systems, institutions, and stakeholders, is essential to ensure equitable, timely access to novel therapies and to translate the potential of genomic profiling into tangible clinical benefit for all patients.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://bc.amegroups.com/article/view/10.21037/bc-25-18/rc

Peer Review File: Available at https://bc.amegroups.com/article/view/10.21037/bc-25-18/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://bc.amegroups.com/article/view/10.21037/bc-25-18/coif). A.I. reports consulting fees from Seagen and Lilly, and support for attending meetings and/or travel from Novartis and AstraZeneca. The other 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. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Cardoso F, Paluch-Shimon S, Senkus E, et al. 5th ESO-ESMO international consensus guidelines for advanced breast cancer (ABC 5). Ann Oncol 2020;31:1623-49. [Crossref] [PubMed]
3. Gradishar WJ, Moran MS, Abraham J, et al. Breast Cancer, Version 3.2024, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2024;22:331-57. [Crossref] [PubMed]
4. Gennari A, André F, Barrios CH, et al. ESMO Clinical Practice Guideline for the diagnosis, staging and treatment of patients with metastatic breast cancer. Ann Oncol 2021;32:1475-95. [Crossref] [PubMed]
5. Turner NC, Neven P, Loibl S, et al. Advances in the treatment of advanced oestrogen-receptor-positive breast cancer. Lancet 2017;389:2403-14. [Crossref] [PubMed]
6. Garraway LA, Lander ES. Lessons from the cancer genome. Cell 2013;153:17-37. [Crossref] [PubMed]
7. Duan XP, Qin BD, Jiao XD, et al. New clinical trial design in precision medicine: discovery, development and direction. Signal Transduct Target Ther 2024;9:57. [Crossref] [PubMed]
8. Mosele MF, Westphalen CB, Stenzinger A, et al. Recommendations for the use of next-generation sequencing (NGS) for patients with advanced cancer in 2024: a report from the ESMO Precision Medicine Working Group. Ann Oncol 2024;35:588-606. [Crossref] [PubMed]
9. Crimini E, Repetto M, Tarantino P, et al. Challenges and Obstacles in Applying Therapeutical Indications Formulated in Molecular Tumor Boards. Cancers (Basel) 2022;14:3193. [Crossref] [PubMed]
10. Pierobon M, Robert NJ, Northfelt DW, et al. Multi-omic molecular profiling guide's efficacious treatment selection in refractory metastatic breast cancer: a prospective phase II clinical trial. Mol Oncol 2022;16:104-15. [Crossref] [PubMed]
11. Aftimos P, Oliveira M, Irrthum A, et al. Genomic and Transcriptomic Analyses of Breast Cancer Primaries and Matched Metastases in AURORA, the Breast International Group (BIG) Molecular Screening Initiative. Cancer Discov 2021;11:2796-811. [Crossref] [PubMed]
12. Irelli A, Chiatamone Ranieri S, Di Giacomo D, et al. Role of the Molecular Tumor Board for the Personalized Treatment of Patients with Metastatic Breast Cancer: A Focus on the State of the Art in Italy. Cancers (Basel) 2023;15:1727. [Crossref] [PubMed]
13. Agostinelli V, Torresi G, Tarantino V, et al. The Impact and Performance of the Molecular Tumor Board: Three-Year Activity in Precision Medicine for Treatment of Patients with Cancer from the Marche Region, in Italy. Oncol Ther 2025;13:201-15. [Crossref] [PubMed]
14. Bartels M, Chibaudel B, Dienstmann R, et al. Evolving and Improving the Sustainability of Molecular Tumor Boards: The Value and Challenges. Cancers (Basel) 2024;16:2888. [Crossref] [PubMed]
15. Di Pilla A, Cozzolino MR, Mannocci A, et al. The Impact of Tumor Boards on Breast Cancer Care: Evidence from a Systematic Literature Review and Meta-Analysis. Int J Environ Res Public Health 2022;19:14990. [Crossref] [PubMed]
16. Papalexis P, Georgakopoulou VE, Drossos PV, et al. Precision medicine in breast cancer Mol Clin Oncol 2024;21:78. (Review). [Crossref] [PubMed]
17. Westphalen CB, Martins-Branco D, Beal JR, et al. The ESMO Tumour-Agnostic Classifier and Screener (ETAC-S): a tool for assessing tumour-agnostic potential of molecularly guided therapies and for steering drug development. Ann Oncol 2024;35:936-53. [Crossref] [PubMed]
18. Ciliberto G, Canfora M, Terrenato I, et al. Bridging therapeutic opportunities: a survey by the Italian molecular tumor board workgroup of Alliance Against Cancer. J Exp Clin Cancer Res 2022;41:305. [Crossref] [PubMed]
19. Pinto C, Biffoni M, Popoli P, et al. Molecular tests and target therapies in oncology: recommendations from the Italian workshop. Future Oncol 2021;17:3529-39. [Crossref] [PubMed]
20. Mateo J, Chakravarty D, Dienstmann R, et al. A framework to rank genomic alterations as targets for cancer precision medicine: the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). Ann Oncol 2018;29:1895-902. [Crossref] [PubMed]
21. Giacomini P, Valenti F, Allegretti M, et al. The Molecular Tumor Board of the Regina Elena National Cancer Institute: from accrual to treatment in real-world. J Transl Med 2023;21:725. [Crossref] [PubMed]
22. Pishvaian MJ, Blais EM, Bender RJ, et al. A virtual molecular tumor board to improve efficiency and scalability of delivering precision oncology to physicians and their patients. JAMIA Open 2019;2:505-15. [Crossref] [PubMed]
23. Ignatiadis M, Sledge GW, Jeffrey SS. Liquid biopsy enters the clinic - implementation issues and future challenges. Nat Rev Clin Oncol 2021;18:297-312. [Crossref] [PubMed]
24. Iams WT, Mackay M, Ben-Shachar R, et al. Concurrent Tissue and Circulating Tumor DNA Molecular Profiling to Detect Guideline-Based Targeted Mutations in a Multicancer Cohort. JAMA Netw Open 2024;7:e2351700. [Crossref] [PubMed]
25. Condorelli R, Mosele F, Verret B, et al. Genomic alterations in breast cancer: level of evidence for actionability according to ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). Ann Oncol 2019;30:365-73. [Crossref] [PubMed]
26. Trapani D, Martins-Branco D, Curigliano G, et al. Updated treatment recommendations for systemic treatment: from the ESMO Metastatic Breast Cancer Living Guideline. Ann Oncol 2025;36:1414-8. [Crossref] [PubMed]
27. Burstein HJ, DeMichele A, Fallowfield L, et al. Endocrine and Targeted Therapy for Hormone Receptor-Positive, Human Epidermal Growth Factor Receptor 2-Negative Metastatic Breast Cancer-Capivasertib-Fulvestrant: ASCO Rapid Recommendation Update. J Clin Oncol 2024;42:1450-3. [Crossref] [PubMed]
28. Turner NC, Oliveira M, Howell SJ, et al. Capivasertib in Hormone Receptor-Positive Advanced Breast Cancer. N Engl J Med 2023;388:2058-70. [Crossref] [PubMed]
29. Turner NC, Im SA, Saura C, et al. Inavolisib-Based Therapy in PIK3CA-Mutated Advanced Breast Cancer. N Engl J Med 2024;391:1584-96. [Crossref] [PubMed]
30. 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]
31. McGrail DJ, Pilié PG, Rashid NU, et al. High tumor mutation burden fails to predict immune checkpoint blockade response across all cancer types. Ann Oncol 2021;32:661-72. [Crossref] [PubMed]
32. Strickler JH, Hanks BA, Khasraw M. Tumor Mutational Burden as a Predictor of Immunotherapy Response: Is More Always Better? Clin Cancer Res 2021;27:1236-41. [Crossref] [PubMed]
33. Magbanua MJM, Gumusay O, Kurzrock R, et al. Immunotherapy in Breast Cancer and the Potential Role of Liquid Biopsy. Front Oncol 2022;12:802579. [Crossref] [PubMed]
34. Miura K, Shukuya T, Greenstein R, et al. Ancestry-, Sex-, and Age-Based Differences of Gene Alterations in NSCLC: From the Real-World Data of Cancer Genomic Profiling Tests. J Natl Compr Canc Netw 2024;22:e247021. [Crossref] [PubMed]
35. Andre F, Filleron T, Kamal M, et al. Genomics to select treatment for patients with metastatic breast cancer. Nature 2022;610:343-8. [Crossref] [PubMed]
36. Botticelli A, Scagnoli S, Conte P, et al. The Rome trial from histology to target: The road to personalize targeted therapy and immunotherapy. Annals of Oncology 2024;35:S1202.
37. Botticelli A, Scagnoli S, Conte P, et al. Abstract P6-10-09: Mutational landscape of breast cancer patients in ROME trial: preliminary results. Cancer Res 2023;83:P6-10-09.
38. Yadav S, Hu C, Hart SN, et al. Evaluation of Germline Genetic Testing Criteria in a Hospital-Based Series of Women With Breast Cancer. J Clin Oncol 2020;38:1409-18. [Crossref] [PubMed]
39. Herrero Colomina J, Johnston E, Duffus K, et al. Real-world experience of Molecular Tumour Boards for clinical decision-making for cancer patients. NPJ Precis Oncol 2025;9:87. [Crossref] [PubMed]
40. Turner NC, Swift C, Jenkins B, et al. Results of the c-TRAK TN trial: a clinical trial utilising ctDNA mutation tracking to detect molecular residual disease and trigger intervention in patients with moderate- and high-risk early-stage triple-negative breast cancer. Ann Oncol 2023;34:200-11. [Crossref] [PubMed]
41. DNA-guided Second Line Adjuvant Therapy for High Residual Risk, Estrogen Receptor Positive, HER-2 Negative Breast Cancer (DARE). Available online: https://clinicaltrials.gov/study/NCT04567420
42. Elacestrant for Treating ER+/HER2- Breast Cancer Patients With ctDNA Relapse (TREAT ctDNA). Available online: https://clinicaltrials.gov/study/NCT05512364
43. Turner NC, Kingston B, Kilburn LS, et al. Circulating tumour DNA analysis to direct therapy in advanced breast cancer (plasmaMATCH): a multicentre, multicohort, phase 2a, platform trial. Lancet Oncol 2020;21:1296-308. [Crossref] [PubMed]
44. Dawson SJ, Tsui DW, Murtaza M, et al. Analysis of circulating tumor DNA to monitor metastatic breast cancer. N Engl J Med 2013;368:1199-209. [Crossref] [PubMed]
45. Lipsyc-Sharf M, McMahon CM, Yu C, et al. ctDNA and residual disease detection in triple-negative breast cancer. J Clin Oncol 2022;40:2297-306.
46. O'Leary B, Cutts RJ, Liu Y, et al. The Genetic Landscape and Clonal Evolution of Breast Cancer Resistance to Palbociclib plus Fulvestrant in the PALOMA-3 Trial. Cancer Discov 2018;8:1390-403. [Crossref] [PubMed]
47. Rossi G, Mu Z, Rademaker AW, et al. Cell-Free DNA and Circulating Tumor Cells: Comprehensive Liquid Biopsy Analysis in Advanced Breast Cancer. Clin Cancer Res 2018;24:560-8. [Crossref] [PubMed]
48. Yagin B, Yagin FH, Colak C, et al. Cancer Metastasis Prediction and Genomic Biomarker Identification through Machine Learning and eXplainable Artificial Intelligence in Breast Cancer Research. Diagnostics (Basel) 2023;13:3314. [Crossref] [PubMed]
49. Zhang WH, Tan Y, Huang Z, et al. Development and validation of an artificial intelligence model for predicting de novo distant bone metastasis in breast cancer: a dual-center study. BMC Womens Health 2024;24:442. [Crossref] [PubMed]