What a Dutch population study of MammaPrint and Oncotype DX reveals about access, adoption and chemotherapy decisions
A short evidence blog from Odelle Technology
| 9% → 25% → 37% Use of gene-expression profiling among eligible patients: no reimbursement, retrospective reimbursement, and publicly known reimbursement |
The key finding at a glance
A national experiment hidden in routine care
Many people treat a reimbursement decision as the moment a medical technology gains access to a healthcare system. However, a new Dutch study shows why that description remains incomplete.
Joyce Meijer, Nienke Hermanns, Sabine Siesling and their colleagues analysed the Netherlands Cancer Registry to determine how reimbursement influenced the use of two gene-expression profile tests in early breast cancer: MammaPrint, the 70-gene signature developed by Agendia, and Oncotype DX, the 21-gene Breast Recurrence Score test from Exact Sciences.
The researchers included 138,765 women aged 18–69 years with primary invasive breast cancer without distant metastasis. All received treatment with curative intent in the Netherlands between January 2011 and June 2024. In addition, the registry captured care across 76 hospitals: 41 general hospitals, 27 top-clinical hospitals and eight academic hospitals.
Coverage created the possibility of access. It did not create uniform clinical use.
The technologies and the clinical decision
MammaPrint and Oncotype DX analyse gene activity within a breast tumour. Consequently, they can refine recurrence risk and support decisions about adjuvant chemotherapy. Clinicians use them most often when conventional clinicopathological factors leave uncertainty about whether chemotherapy offers enough additional benefit to justify its toxicity and burden.
MammaPrint classifies tumours through a 70-gene signature, and the international MINDACT trial prospectively tested its clinical utility. By contrast, Oncotype DX analyses 21 genes and reports a recurrence score that helps clinicians assess the likely value of chemotherapy in hormone-receptor-positive, HER2-negative early breast cancer.
Therefore, the practical value of these technologies extends beyond prognosis. Both tests aim to improve a treatment decision: they help clinicians identify patients who may safely avoid chemotherapy while preserving treatment for patients whose tumour biology indicates higher risk.
What reimbursement changed
The Dutch reimbursement history created a natural policy experiment. Specifically, the authors grouped eligible patients into three periods: no reimbursement, retrospective reimbursement and publicly known reimbursement.
| Reimbursement status | Eligible patients receiving a GEP | Interpretation |
| No reimbursement | 9% | Low baseline use |
| Retrospective reimbursement | 25% | A substantial rise, despite the decision not yet being publicly known at the time of care |
| Publicly known reimbursement | 37% | Further growth, but most eligible patients still remained untested |
The increase from 9% to 37% matters because it confirms that reimbursement can accelerate adoption. However, the reverse figure tells an even stronger story: after reimbursement became publicly known, 63% of eligible patients still did not receive a gene-expression test.
Importantly, the authors do not assume that every eligible patient should undergo testing. For example, some patients may reject chemotherapy regardless of the result, which would reduce the test’s value in shared decision-making. Nevertheless, patient preference alone is unlikely to explain such widespread non-use. Therefore, the authors call for research into guideline interpretation, regional policy, alternative decision tools and the reasons clinicians did not order a reimbursed test.
The result changed treatment
Among tested patients, the genomic risk result strongly influenced the chemotherapy decision:
MammaPrint low risk: 93% did not receive chemotherapy.
MammaPrint high risk: 83% received chemotherapy.
Oncotype DX low risk: 97% did not receive chemotherapy.
Oncotype DX high risk: 85% received chemotherapy.
The researchers also identified 13,003 patients who met the gene-expression-profile criteria but did not receive a test. Of these patients, 52% received chemotherapy. Because this was an observational study, the authors could not determine how many chemotherapy courses were unnecessary. Even so, the results reveal an important missed decision point: genomic testing might have helped some patients avoid chemotherapy.
Reimbursement also changed the competitive landscape
This study offers unusual commercial insight because it follows two competing tests through different reimbursement periods. Until 2021, clinicians used MammaPrint more frequently in every Dutch region. However, after Oncotype DX entered reimbursed care in August 2021 while MammaPrint remained outside national reimbursement, Oncotype DX use rose and MammaPrint use fell. Then, after reimbursement for MammaPrint became publicly known in October 2023, its use increased again.
Moreover, the adjusted analysis linked publicly known reimbursement with markedly higher odds of Oncotype DX use than the non-reimbursed period. Reimbursement did not act alone: MINDACT publications, alternative funding arrangements, guideline changes and differences in patient characteristics also influenced uptake. Nevertheless, the pattern shows that reimbursement can redistribute market share as well as expand access.
The unresolved implementation gap
Regional variation also persisted. Clinicians in the Southwest used Oncotype DX particularly often, whereas the Northwest recorded the highest MammaPrint use. Therefore, a national benefit-package decision did not remove local differences in hospital policy, professional preference or implementation capacity.
Taken together, the findings show that access develops through several distinct stages:
Evidence → Coverage → Hospital adoption → Clinician ordering → Patient use → Treatment effect
A technology can succeed at the first two stages and still fail to reach much of its intended population. Reimbursement removes an important financial barrier; however, it does not automatically create a clinical pathway, an ordering habit, a multidisciplinary protocol or an audit system.
For that reason, the authors call for active implementation. They highlight standardised clinical pathways, uptake monitoring, investigation of unexplained regional variation and research into why eligible patients remain untested.
The people and organisations behind the study
Joyce Meijer and Nienke Hermanns led the research with Sabine Siesling and a broad Dutch clinical and policy network. The collaboration brought together the Netherlands Comprehensive Cancer Organisation (IKNL), the Technical Medical Centre at the University of Twente, UMC Utrecht, Erasmus School of Health Policy & Management, Amsterdam UMC, Erasmus MC, the Netherlands Cancer Institute, Borstkankervereniging Nederland, Isala, Ziekenhuisgroep Twente, Diakonessenhuis and the University of California, San Francisco.
The full author group comprised Joyce Meijer, Nienke Hermanns, Rhodé Bijlsma, Hedwig M. Blommestein, Desiree H. J. G. van den Bongard, Thijs van Dalen, Paul J. van Diest, Caroline A. Drukker, Cristina Guerrero Paez, Aafke H. Honkoop, Agnes Jager, Linetta B. Koppert, Sabine C. Linn, Marissa C. van Maaren, Ester J. Siemerink, Laura J. van ’t Veer, José H. Volders, Marie Jeanne Vrancken Peeters and Sabine Siesling.
IKNL maintains the Netherlands Cancer Registry, while the Dutch Nationwide Pathology Databank (Palga) supplies cancer notifications. Meanwhile, Zorginstituut Nederland made the reimbursement decisions analysed in the study. Agendia developed MammaPrint, and Exact Sciences provides Oncotype DX.
Why this paper matters
The paper does not argue that every eligible patient must undergo genomic testing. Nor does it claim that every course of chemotherapy given without a test was avoidable. Instead, it measures the gap between formal reimbursement and real-world utilisation. It then shows that the test result strongly influenced treatment choice.
For developers, payers and hospitals, the message is clear. A positive coverage decision marks an essential milestone, but it does not complete market access. Ultimately, precision medicine creates clinical and economic value only when clinicians identify the right patient, order the test, interpret the result and change treatment appropriately.
Reimbursement opens the door. Implementation determines who actually receives the benefit.
References
1. Meijer J, Hermanns N, Bijlsma R, et al. Impact of Reimbursement on the Utilisation of Gene Expression Profiles and Chemotherapy Decision-Making in Dutch Breast Cancer Patients: A Population-Based Study. International Journal of Cancer. 2026;1–13. doi:10.1002/ijc.70571. Source
2. Zorginstituut Nederland. Standpunt: MammaPrint en Oncotype DX vergoede zorg voor bepaalde groep vrouwen. 24 October 2023. Source
3. Zorginstituut Nederland. Standpunt Oncotype DX bij vroeg stadium borstkanker. 11 August 2021. Source
4. Cardoso F, van ’t Veer LJ, Bogaerts J, et al. 70-Gene Signature as an Aid to Treatment Decisions in Early-Stage Breast Cancer. New England Journal of Medicine. 2016;375(8):717–729. doi:10.1056/NEJMoa1602253. Source
5. Piccart M, van ’t Veer LJ, Poncet C, et al. 70-gene signature as an aid for treatment decisions in early breast cancer: updated results of the phase 3 randomised MINDACT trial with an exploratory analysis by age. Lancet Oncology. 2021;22(4):476–488. doi:10.1016/S1470-2045(21)00007-3. Source
6. Retèl VP, Byng D, Linn SC, et al. Cost-effectiveness analysis of the 70-gene signature compared with clinical assessment in breast cancer based on a randomised controlled trial. European Journal of Cancer. 2020;137:193–203. doi:10.1016/j.ejca.2020.07.002. Source
7. Simons MJ, Machielsen PM, Spoorendonk JA, et al. A cost-consequence model of using the 21-gene assay to identify patients with early-stage node-positive breast cancer who benefit from adjuvant chemotherapy in the Netherlands. Journal of Medical Economics. 2024;27(1):445–454. doi:10.1080/13696998.2024.2324612. Source
8. Hassan S, Younan R, Patocskai E, et al. Impact of the 21-gene recurrence score assay on treatment decisions and cost in patients with node-positive breast cancer: a multicenter study in Quebec. The Oncologist. 2022;27(10):822–831. doi:10.1093/oncolo/oyac123. Source
9. Agendia. MammaPrint: 70-gene breast cancer recurrence test. Source
10. Exact Sciences. Oncotype DX Breast Recurrence Score test. Source
Editorial note: This article interprets the cited population study. It does not suggest that every untested patient received unnecessary chemotherapy. On the contrary, the original authors state that researchers still need to investigate why clinicians did not order tests for many eligible patients.
Odelle Technology • August 2026 •