Multi-gene expression assays in breast cancer: a literature review

Introduction

Breast cancer remains the most common cancer among women globally, with an estimated 2.3 million new cases and 685,000 deaths in 2020 alone (1). This malignancy exhibits extreme heterogeneity, both inter- and intratumorally, which poses significant challenges for accurate prognosis and effective treatment. Traditional clinicopathological factors, such as tumor size, lymph node status, and histological grade, have long been the cornerstone of breast cancer management. These factors, while providing some level of prognostic information, often lack the precision required to stratify patients accurately. For example, two patients with similar tumor sizes and lymph node involvement may have vastly different clinical outcomes, highlighting the limitations of relying solely on these traditional markers (2).

The advent of multi-gene expression assays has revolutionized the field of breast cancer research and clinical practice. These assays analyze the expression levels of multiple genes simultaneously, providing a more comprehensive molecular portrait of the tumor. By measuring the activity of genes involved in various biological processes, such as cell proliferation, apoptosis, and angiogenesis, multi-gene expression assays can offer more refined prognostic and predictive information. This molecular fingerprint can help clinicians make more informed decisions regarding treatment selection, potentially reducing overtreatment in low-risk patients and ensuring appropriate therapy for those at higher risk (3).

Given the growing number of gene expression platforms and their integration into clinical guidelines, it is essential to comprehensively understand their comparative strengths, limitations, and implications for clinical decision-making. This review aims to provide a detailed synthesis of the major multigene expression assays used in early-stage breast cancer—namely Oncotype DX, MammaPrint, and Prosigna—and to examine their roles in guiding adjuvant therapy, minimizing overtreatment, and supporting precision medicine initiatives. By evaluating the current evidence and highlighting ongoing challenges such as cost, technical variability, and population-specific validation gaps, this article seeks to inform both clinical practice and future research directions in the field of molecular oncology and health policy. We present this article in accordance with the Narrative Review reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-803/rc).

Methods

Literature search strategy

A comprehensive literature search was carried out to identify relevant studies on multi-gene expression assays in breast cancer. The search spanned multiple databases, including PubMed, Embase, and Web of Science. The search terms used were a combination of Medical Subject Headings (MeSH) and free-text words related to breast cancer, multi-gene expression assays, Oncotype DX, MammaPrint, Prosigna, prognosis, treatment selection, limitations, and future directions. For example, search strings included “breast cancer AND multi-gene expression assays”, “Oncotype DX AND breast cancer prognosis”, and “MammaPrint AND treatment selection in breast cancer”. The search was restricted to articles published in English between 2000 and 2024 to ensure the relevance and timeliness of the information.

Study selection criteria

Following the initial literature search, the titles and abstracts of all identified articles were screened to determine their potential relevance. Articles were included if they focused on the development, validation, clinical utility, limitations, or future directions of multi-gene expression assays in breast cancer. Studies that presented original data, review articles, meta-analyses, and clinical guidelines were considered eligible. Articles were excluded if they were not related to breast cancer, did not involve multi-gene expression assays, were published in a language other than English, or were conference abstracts without full-text availability.

After the title and abstract screening, the full-text articles of the potentially relevant studies were retrieved and carefully reviewed. Two independent reviewers evaluated each article based on the predefined inclusion and exclusion criteria. Any discrepancies in the review process were resolved through discussion or by consulting a third reviewer.

Data extraction and synthesis

For the included studies, relevant data were extracted using a standardized data extraction form. The extracted data encompassed the study design, patient characteristics (such as age, tumor subtype, and stage), details of the multi-gene expression assay used (genes analyzed, scoring system), main findings related to prognosis prediction, treatment selection, limitations, and any proposed future directions.

The extracted data were then synthesized in a narrative manner. For studies reporting similar findings, the results were grouped together and presented concisely. When available, statistical data such as hazard ratios, recurrence rates, and survival outcomes were summarized to provide quantitative evidence for the clinical utility of multi-gene expression assays. The limitations of the assays were discussed based on the common issues reported across multiple studies, and the future directions were derived from the emerging trends and research gaps identified in the reviewed literature. The search strategy summary was presented in Table 1.

Key content and findings

Types of multi-gene expression assays

Oncotype Dx

Oncotype DX is one of the most well-studied and widely used multi-gene expression assays in clinical practice. It was developed by Genomic Health and is specifically designed for estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative breast cancer patients. The assay analyzes the expression of 21 genes, which are selected based on their known biological functions related to breast cancer progression. These genes include those involved in cell cycle regulation (e.g., Ki-67), invasion (e.g., MMP11), and hormonal response [e.g., ER and progesterone receptor (PR)] (4).

The Oncotype DX assay generates a recurrence score (RS) from 0 to 100 to guide adjuvant therapy in early-stage, HR-positive, HER2-negative, node-negative breast cancer. A low RS [≤10] indicates excellent prognosis with endocrine therapy alone, with a 10-year distant recurrence rate below 2%. A high RS [≥26] suggests a significantly increased recurrence risk, and chemotherapy is generally recommended (5,6). In contrast, patients with an intermediate RS [11–25] represent a more complex group, requiring treatment decisions based on age, tumor characteristics, and patient preference. The TAILORx trial showed that endocrine therapy alone was sufficient for most patients in this category. However, subgroup analysis indicated that women aged ≤50 with an RS of 16–25 may receive modest benefit from chemotherapy, particularly as RS increases. These findings support the use of Oncotype DX to guide personalized therapy and reduce overtreatment (2).

MammaPrint

MammaPrint, developed by Agendia, evaluates the expression of 70 genes to classify breast cancer patients into low-risk and high-risk groups for distant metastasis. Unlike Oncotype DX, MammaPrint is applicable to all breast cancer subtypes, regardless of ER, PR, or HER2 status. The 70-gene signature was initially identified through a large-scale gene-expression profiling study of over 295 breast cancer patients, followed by validation in multiple independent cohorts (3).

In early-stage breast cancer, MammaPrint has been shown to be highly effective in identifying patients at low risk of developing distant metastases. In the MINDACT trial, which included over 6,600 early-stage breast cancer patients, the 70-gene signature was able to accurately identify patients who could safely forego chemotherapy, without compromising overall survival. Among patients classified as having high clinical risk but low genomic risk, the 5-year distant metastasis-free survival rate was 94.7% in those who did not receive chemotherapy, with an absolute difference of less than 1.5% compared to those who did. These findings support the notion that chemotherapy can be safely omitted in patients with discordant risk profiles—clinically high but genomically low—without compromising oncologic outcomes (7). Reflecting this evidence, the National Comprehensive Cancer Network (NCCN) Breast Cancer Guidelines recommend MammaPrint as a Category 1 test for HR-positive, HER2-negative early-stage breast cancer with 0–3 positive lymph nodes, particularly in cases where clinical and genomic risk assessments are incongruent. Additionally, the assay is endorsed for identifying ultra-low-risk patients who may benefit from de-escalation of systemic therapy (8). Although MammaPrint is not yet formally endorsed in Japan’s national guidelines, the Pan-Asian European Society for Medical Oncology (ESMO)-adapted consensus acknowledges the applicability of MINDACT data to Asian populations and supports its clinical integration in comparable risk settings (9).

Prosigna

Prosigna, a 50-gene assay developed by Nanostring Technologies (Seattle, WA, USA), provides both a risk-of-recurrence score and classifies tumors into four intrinsic molecular subtypes: luminal A, luminal B, HER2-enriched, and basal-like. The assay uses a novel digital counting technology to measure gene expression levels, offering high sensitivity and reproducibility (10).

The subtyping capability of Prosigna is particularly useful in cases where immunohistochemistry results are equivocal or when further refinement of the risk assessment within each subtype is needed. For instance, within the luminal subtypes, Prosigna can distinguish between luminal A-like tumors, which typically have a more indolent course and better prognosis, and luminal B-like tumors, which are more aggressive and may require more intensive treatment. In a study of over 1,000 breast cancer patients, Prosigna-determined subtypes were associated with distinct clinical outcomes, further validating its utility in clinical decision-making (11). In the most recent version of the NCCN Clinical Practice Guidelines (8), Prosigna is recommended as a consideration-worthy multigene assay, specifically applicable to patients with HR-positive, HER2-negative early-stage breast cancer with 0–3 positive lymph nodes. It is recognized for its prognostic utility and assigned a Category 2A level of evidence.

Clinical utility

Prognosis prediction

Multi-gene expression assays have significantly enhanced the accuracy of predicting breast cancer recurrence. In HR-positive, HER2-negative disease, numerous large-scale studies have demonstrated the superiority of the Oncotype DX recurrence score over traditional clinicopathological factors in predicting distant recurrence. For example, the NSABP B-20 trial, which included over 6,500 node-negative, HR-positive breast cancer patients, found that the Oncotype DX RS was a strong independent predictor of recurrence, even after adjusting for other known prognostic factors (12). The RxPONDER trial demonstrated that the Oncotype DX RS predicts chemotherapy benefit in HR-positive, HER2-negative breast cancer with 1–3 positive lymph nodes. Postmenopausal women with RS ≤25 did not benefit from chemotherapy, while premenopausal women showed improved outcomes, highlighting the role of menopausal status in guiding treatment. Based on the results of the TAILORx and RxPONDER trials, the current NCCN guidelines designate the Oncotype DX multigene assay as a preferred test and recommend its incorporation into treatment algorithms under specific clinical scenarios. For example, in HR-positive, HER2-negative early breast cancer, postmenopausal women with one to three positive lymph nodes and an RS <25 may reasonably forgo adjuvant chemotherapy. To enhance clinical decision-making, the RSClin tool integrates the RS with key clinicopathologic factors, including patient age, tumor size, and histologic grade. Compared to RS alone, RSClin provides a more accurate and individualized estimate of both recurrence risk and the potential benefit of adjuvant chemotherapy, particularly in node-positive patients or those with intermediate-risk profiles. The tool generates two primary outputs: the 10-year risk of distant recurrence and the absolute benefit of chemotherapy. By quantifying the expected absolute risk reduction associated with chemotherapy across different risk categories, RSClin refines treatment recommendations. For instance, a patient with an RS of 21 may initially appear to fall within an intermediate-risk group; however, if additional clinical factors such as high tumor grade and large tumor size are present, RSClin may reclassify the patient as having a higher risk of recurrence (ROR), suggesting a meaningful potential benefit from chemotherapy (5,13).

Similarly, MammaPrint has been validated in multiple multi-center trials for its ability to accurately identify low-risk patients in early-stage breast cancer. The EORTC 10041/BIG 03-04 MINDACT trial, as mentioned earlier, demonstrated that patients with a low-risk MammaPrint score had a very low risk of distant recurrence, regardless of their clinical risk classification based on traditional factors (7). These assays provide a more nuanced understanding of a patient’s long-term prognosis, allowing for more personalized treatment planning.

Prosigna, a gene expression assay based on the PAM50 intrinsic subtype classification, is clinically used to guide prognosis and adjuvant treatment decisions in early-stage breast cancer, particularly in hormone receptor-positive, HER2-negative patients. The assay provides two key outputs: the intrinsic molecular subtype—categorizing tumors into luminal A, luminal B, HER2-enriched, and Basal-like—and the ROR score, which estimates the 10-year risk of distant recurrence. Studies such as TransATAC and ABCSG-8 have demonstrated the prognostic utility of PAM50 subtypes, showing that luminal A tumors have significantly lower 10-year distant recurrence rates compared to luminal B tumors (14,15). These findings support the use of Prosigna for refining risk stratification and minimizing overtreatment. For example, in a postmenopausal patient with hormone receptor-positive, HER2-negative, T2N0 breast cancer, chemotherapy might be considered based on clinical features. However, if Prosigna identifies the tumor as luminal A with a low ROR score (e.g., 28), endocrine therapy alone may be sufficient, thus safely avoiding unnecessary chemotherapy.

Treatment selection

The information provided by multi-gene expression assays is invaluable in guiding treatment decisions. In ER-positive, HER2-negative breast cancer, the Oncotype DX recurrence score is a key factor in determining whether a patient will benefit from chemotherapy. Patients with a low RS can often achieve excellent long-term outcomes with endocrine therapy alone, sparing them the significant side-effects of chemotherapy, such as nausea, hair loss, and increased risk of infections. In contrast, patients with a high RS are more likely to derive substantial benefit from chemotherapy in addition to endocrine therapy (2).

For patients with triple-negative breast cancer, a more aggressive subtype with limited treatment options, there are currently no widely-adopted multi-gene assays for routine use. However, ongoing research is focused on developing gene signatures that can predict response to chemotherapy or novel targeted therapies. For example, some studies have identified gene signatures associated with increased sensitivity or resistance to platinum-based chemotherapy in triple-negative breast cancer, which could potentially be used to select the most appropriate treatment for individual patients (16).

Limitation

Cost

One of the major barriers to the widespread adoption of multi-gene expression assays is their high cost. The cost of these tests can vary significantly depending on the assay and the geographical location, but it generally ranges from several thousand to tens of thousands of dollars. For example, the Oncotype DX assay can cost upwards of $4,000 in the United States, which may be prohibitively expensive for many patients, especially those without comprehensive insurance coverage (17). This high cost not only limits access to these valuable prognostic and predictive tools but also contributes to disparities in healthcare, as patients in resource-limited settings or with lower socioeconomic status may be unable to afford them (18).

Coverage, access, and equity

In September 2023, Japan’s Ministry of Health, Labour and Welfare approved reimbursement of the Oncotype DX^® Breast Recurrence Score test under the National Health Insurance system for patients with hormone receptor-positive, HER2-negative early-stage breast cancer, enabling publicly funded genomic risk stratification and the potential to reduce unnecessary chemotherapy (19). By contrast, other multigene assays such as Prosigna/PAM50 and MammaPrint are not currently reimbursed in Japan. In the United States, Oncotype DX is covered by all major commercial insurers and by federal programs including Medicare and Medicaid for eligible patients, effectively eliminating out-of-pocket costs for many (20). However, assays such as Prosigna and MammaPrint are not uniformly covered under Medicare at the federal level and are often paid out-of-pocket or only partially reimbursed through selected private plans.

Despite these policy advances, uneven coverage, affordability barriers, and limited representation of minority and socioeconomically disadvantaged populations in genomic research risk amplifying pre-existing health inequities. As Ramaswami et al. argue, precision medicine initiatives often preferentially benefit higher-socioeconomic status and majority groups, leaving underserved populations with reduced access to testing and fewer data-driven therapeutic recommendations. Embedding equity as a core implementation criterion—through inclusive cohort design, payer policies that minimize financial toxicity, and health-technology assessments that explicitly consider distributive justice—is essential to ensure that multigene expression assays genuinely advance precision oncology for all patients (21).

Technical variability

There is inherent technical variability in the performance of multi-gene expression assays. Differences in sample collection methods, such as the type of tissue biopsy (core needle biopsy vs. surgical excision), storage conditions (temperature and duration of storage), processing techniques (RNA extraction and amplification methods), and the specific laboratory platforms used for gene expression analysis (e.g., microarray vs. next-generation sequencing) can all potentially affect the results (22). Although efforts are underway to standardize these procedures, achieving consistent results across different laboratories remains a challenge. Inconsistent results can lead to confusion in clinical decision-making and undermine the reliability of these assays, potentially resulting in inappropriate treatment decisions (23).

Tumor heterogeneity may also contribute to technical variability. Sampling from different regions of the same tumor can result in distinct molecular characteristics, such as variations in ER, PR, HER2, Ki-67, programmed death-ligand 1 (PD-L1) expression, or BRCA mutations. Consequently, a single-site biopsy may not adequately reflect the overall tumor profile and may even misrepresent the actual driver mutations or biomarker distribution, potentially leading to suboptimal decisions in targeted or immunotherapy. Although several genomic tools offer valuable prognostic and predictive information, incorporating tumor heterogeneity into their interpretation remains essential. Future strategies should enable dynamic assessment of tumor heterogeneity and integrate this information into treatment planning and disease monitoring to enhance the effectiveness of precision medicine (24,25).

Additionally, in multifocal breast cancer, individual lesions may arise from different clonal origins. Desmedt et al. investigated 36 patients with ipsilateral multifocal breast cancer and found that even among lesions with similar pathological features (e.g., same grade and ER/HER2 status), approximately one-third exhibited genomic-level heterogeneity. In such cases, relying on a single lesion for molecular testing could result in missing critical driver mutations, thereby overlooking potential opportunities for targeted therapy. Therefore, even when lesions appear pathologically similar, multi-lesion sampling is recommended to ensure a comprehensive molecular profile for guiding precision oncology (26).

Limited generalizability

Some multi-gene expression assays may have restricted applicability to certain patient populations or subtypes. For example, the Oncotype DX assay was primarily developed and validated in ER-positive, HER2-negative breast cancer patients. Its utility in other subtypes, such as triple-negative or HER2-positive breast cancer, is less well-established. While some studies have attempted to apply Oncotype DX in these subtypes, the results have been less conclusive, highlighting the need for subtype-specific gene expression assays (2). Extrapolating the results of these assays to all breast cancer patients without proper validation can lead to inaccurate treatment decisions and suboptimal patient outcomes (27). Although Prosigna has been validated in multiple large clinical trials, the majority of supporting data are derived from European or North American populations. The limited representation of Asian patients in these studies raises concerns regarding the assay’s generalizability and predictive accuracy in Asian populations. Ethnic differences in tumor biology, treatment response, and gene expression profiles may influence the performance of multigene assays like Prosigna, underscoring the need for further validation in diverse cohorts (16,28). Additionally, the prognostic accuracy of Oncotype DX is lower in Black women; even with the same recurrence score, breast cancer-specific mortality remains significantly higher. Hoskins et al. emphasized the need to improve the representation of diverse populations in the development of genomic tools (29).

Clinical responsibility and risk management considerations

Gene expression assays can significantly influence clinical treatment decisions. The latest NCCN guidelines now include recommendations on the consideration of adjuvant systemic therapy based on various gene expression assays, including Oncotype DX, MammaPrint, Prosigna (PAM50), and EndoPredict (8). Nonetheless, clinicians must exercise final clinical judgment by integrating the broader clinical context—including pathological features, breast cancer subtype, patient age, menopausal status, and patient preference—rather than relying solely on assay results. If an inappropriate treatment decision is made based solely on assay findings without a clearly documented clinical rationale, it may raise concerns regarding medical negligence.

Many of these assays, particularly Oncotype DX and PAM50, were validated predominantly in Western populations, and their predictive accuracy in Asian or Black populations remains uncertain. For example, as previously noted, Oncotype DX has demonstrated reduced prognostic performance in Black women, potentially underestimating true risk (29). Additional concerns associated with gene expression assays include false-negative or false-positive results, ambiguous risk classifications, and pre-analytical vulnerabilities. For example, Prosigna (PAM50) requires formalin-fixed paraffin-embedded (FFPE) tissue; if sample quality is suboptimal or RNA degradation occurs, the results may be unreliable (30).

To mitigate these risks, it is recommended that patients receive a comprehensive informed consent prior to testing. This should include disclosure of the assay’s limitations—such as population-specific validation gaps and tissue quality dependencies. In cases where assay results are atypical or inconsistent with the clinical picture, physicians should explicitly explain the rationale behind their decisions, potential risks, and engage in shared decision-making with the patient.

Future directions

Integration with other omics data

The future of breast cancer research lies in the integration of multi-gene expression data with other omics data, such as genomics (somatic mutations), proteomics, and metabolomics. This integrated approach has the potential to provide a more comprehensive understanding of breast cancer biology and identify novel biomarkers and therapeutic targets. For example, integrating gene expression data with mutational profiling can help identify subgroups of patients with specific genetic alterations that are more responsive to certain therapies. In a recent study, patients with breast cancer harboring specific mutations in the PIK3CA gene were found to have distinct gene expression profiles, which could potentially be used to guide targeted therapy (31).

Development of novel assays

There is an urgent need for the development of new multi-gene expression assays that can better predict response to emerging therapies, such as immunotherapy and targeted agents for breast cancer. Currently, the majority of existing assays were developed based on traditional chemotherapy-based treatment paradigms. As the field of breast cancer treatment evolves, new assays are required to keep pace. Additionally, assays that can effectively monitor disease recurrence during follow-up or predict the development of resistance to existing therapies are areas of active research. For example, some studies are exploring the use of circulating tumor DNA (ctDNA)-based gene expression assays to detect early recurrence and monitor treatment response in breast cancer patients (32).

Recent studies have highlighted non-coding RNAs (ncRNAs), including the relatively understudied Y RNAs, as mechanistically informative biomarkers with potential to enhance risk stratification and personalized treatment in breast cancer. Y RNAs are involved in DNA replication, RNA quality control, and cellular stress responses. Aberrant expression and extracellular vesicle-mediated secretion of Y RNAs have been associated with tumor initiation, progression, and immune regulation, indicating their potential as minimally invasive diagnostic and prognostic markers (33,34). Concurrently, CRISPR-Cas technologies are revolutionizing breast cancer research through two major applications: large-scale functional interrogation of ncRNA regulatory networks to elucidate mechanisms of therapeutic resistance, and CRISPR-based detection platforms (e.g., SHERLOCK, DETECTR) that enable sensitive quantification of ncRNAs in liquid biopsies for real-time disease monitoring (35).

Rather than competing with established multigene assays such as Oncotype DX and Prosigna, ncRNA-based and CRISPR-enabled approaches should be considered complementary. These novel tools may help resolve intermediate-risk cases, identify functionally distinct subgroups, and expand the clinical utility of current recurrence scores by incorporating the regulatory complexity of the ncRNA landscape.

Conclusions

Multi-gene expression assays have transformed the landscape of breast cancer management, providing more accurate prognostic and predictive information. Despite their limitations, these assays have already had a significant impact on treatment decisions, reducing overtreatment in low-risk patients and improving overall outcomes. Table 2 compiles pivotal studies on multigene assays, outlining their development, validation, clinical utility, cost issues, technical variability, and future research directions. As technology continues to advance and our understanding of breast cancer biology deepens, multi-gene expression assays are expected to play an even more central role in personalized medicine for breast cancer patients. However, addressing the challenges of cost, technical variability, and limited generalizability will be crucial to ensure their widespread adoption and optimal use in clinical practice.

Acknowledgments

The authors gratefully acknowledge the support of the Melissa Lee Cancer Foundation. The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors also thank Dr. Morris Chang for his valuable contributions.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-803/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-803/prf

Funding: This work was supported in part by VGH-TPE (grant Nos. V110E-005-3, V111E-006-3, V112E-004-3 and V112C-013) and National Science and Technology Council (grant No. NSTC 111-2314-B-075-063-MY3).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-803/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. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin 2020;70:7-30. [Crossref] [PubMed]
2. Sparano JA, Gray RJ, Makower DF, et al. Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. N Engl J Med 2018;379:111-21. [Crossref] [PubMed]
3. van 't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530-6. [Crossref] [PubMed]
4. Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med 2004;351:2817-26. [Crossref] [PubMed]
5. Kalinsky K, Barlow WE, Gralow JR, et al. 21-Gene Assay to Inform Chemotherapy Benefit in Node-Positive Breast Cancer. N Engl J Med 2021;385:2336-47. [Crossref] [PubMed]
6. Mamounas EP, Tang G, Paik S, et al. 21-Gene Recurrence Score for prognosis and prediction of taxane benefit after adjuvant chemotherapy plus endocrine therapy: results from NSABP B-28/NRG Oncology. Breast Cancer Res Treat 2018;168:69-77. [Crossref] [PubMed]
7. Cardoso F, van't Veer LJ, Bogaerts J, et al. 70-Gene Signature as an Aid to Treatment Decisions in Early-Stage Breast Cancer. N Engl J Med 2016;375:717-29. [Crossref] [PubMed]
8. National Comprehensive Cancer Network (NCCN). NCCN Clinical Practice Guidelines in Oncology: Breast Cancer. Version 4.2025. Plymouth Meeting, PA: NCCN; April 30, 2025.
9. Lin CH, Yu J, Hsu GC, et al. Consensus statement on the management of breast cancer in Asia: the Pan-Asian adapted ESMO Clinical Practice Guidelines. Ann Oncol 2020;31:1213-33.
10. Nielsen TO, Hsu FD, Jensen K, et al. Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res 2004;10:5367-74. [Crossref] [PubMed]
11. Parker JS, Mullins M, Cheang MC, et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol 2009;27:1160-7. [Crossref] [PubMed]
12. Paik S, Tang G, Shak S, et al. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer. J Clin Oncol 2006;24:3726-34. [Crossref] [PubMed]
13. Sparano JA, Gray RJ, Makower DF, et al. Clinical and genomic risk to guide the use of adjuvant therapy for breast cancer: validation of the RSClin tool. J Clin Oncol 2021;39:557-66.
14. Dowsett M, Sestak I, Lopez-Knowles E, et al. Comparison of PAM50 risk of recurrence score with oncotype DX and IHC4 for predicting risk of distant recurrence after endocrine therapy. J Clin Oncol 2013;31:2783-90. [Crossref] [PubMed]
15. Gnant M, Filipits M, Greil R, et al. Predicting distant recurrence in receptor-positive breast cancer patients with limited clinicopathological risk: using the PAM50 risk of recurrence score in 1,478 postmenopausal patients of the ABCSG-8 trial. Ann Oncol 2014;25:339-45. [Crossref] [PubMed]
16. Dent R, Trudeau M, Pritchard KI, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res 2007;13:4429-34. [Crossref] [PubMed]
17. Smith-Bindman R, Miglioretti DL, Larson EB, et al. Use of 21-gene assay and other prognostic tests for breast cancer in a large integrated health care system. JAMA Oncol 2015;1:499-506.
18. Kuderer NM, Newman LA, Jeong JH, et al. Patterns of use and cost of gene expression profiling in breast cancer. J Natl Cancer Inst 2010;102:1466-74.
19. Exact Sciences. Japan approves reimbursement of the Oncotype DX Breast Recurrence Score® test. September 2023. Available online: https://www.exactsciences.com/newsroom/2023-09-Exact-Sciences-Japan-approves-reimbursement-oncotypeDX
20. Exact Sciences Precision Oncology. Insurance coverage for the Oncotype DX test – Medicare, Medicaid, and Commercial Insurance. Accessed July 2025. Available online: https://precisiononcology.exactsciences.com/oncotype-dx/healthcare-provider/coverage
21. Ramaswami R, Bayer R, Galea S. Precision Medicine from a Public Health Perspective. Annu Rev Public Health 2018;39:153-68. [Crossref] [PubMed]
22. Hicks DG, Symmans WF, Pusztai L. Standardization of multigene assays for breast cancer: a pathologist's perspective. Mod Pathol 2011;24:S26-S34.
23. Hayes DF, Thor AD, Dressler LG, et al. Tumor gene expression profiles in breast cancer: concordance among different microarray platforms. Breast Cancer Res Treat 2005;92:253-68.
24. Wadhwa B, Vasiyani H. Tumor heterogeneity in breast cancer and its implication in drug resistance. Clin Oncol 2017;2:1280.
25. Martelotto LG, Ng CK, Piscuoglio S, et al. Breast cancer intra-tumor heterogeneity. Breast Cancer Res 2014;16:210. [Crossref] [PubMed]
26. Desmedt C, Fumagalli D, Pietri E, et al. Uncovering the genomic heterogeneity of multifocal breast cancer. J Pathol 2015;236:457-66. [Crossref] [PubMed]
27. Pusztai L, Anderson K, Buyse M, et al. Considerations for clinical utility and scientific validation of genomic profiling in breast cancer. J Natl Cancer Inst 2011;103:1413-25.
28. Yamashita H, Koike K, Nakagawa T, et al. Prognostic value of the PAM50 gene expression profile in Japanese breast cancer patients. Clin Breast Cancer 2019;19:e530-e540.
29. Hoskins KF, Danciu OC, Ko NY, et al. Association of Race/Ethnicity and the 21-Gene Recurrence Score With Breast Cancer-Specific Mortality Among US Women. JAMA Oncol 2021;7:370-8. [Crossref] [PubMed]
30. Wallden B, Storhoff J, Nielsen T, et al. Development and verification of the PAM50-based Prosigna breast cancer gene signature assay. BMC Med Genomics 2015;8:54. [Crossref] [PubMed]
31. Alexandrov LB, Nik-Zainal S, Wedge DC, et al. Signatures of mutational processes in human cancer. Nature 2013;500:415-21. [Crossref] [PubMed]
32. Ma CX, Salunga R, Tuggle JT, et al. Gene expression profiles associated with tamoxifen resistance in breast cancer. Cancer Res 2004;64:9203-8.
33. Gulìa C, Signore F, Rigon G, et al. Non-coding RNAs in cancer: from biomarkers to therapeutic targets. Cancers (Basel) 2020;12:E3190.
34. Piergentili R, Malerba N, De Paolis V, et al. Y RNAs: recent developments on mechanisms and biological functions. Cells 2021;10:1566.
35. Passaro A, Xu L, Cheng Z, Mendez O, Martelli MP, Li J, et al. CRISPR-based liquid biopsy profiling of non-coding RNAs in breast cancer. Cell 2024;187:823-839.e10.