Original Title: Multigene-expression tests in early breast cancer

Authors' objectives: BACKGROUND Breast cancer is the second most common cancer in Switzerland, with approximately 6,300 women and 50 men being diagnosed annually. The disease arises from genetic mutations leading to uncontrolled cell growth in breast tissue. Diagnosis involves physical examination, imaging, and histopathological analysis, with staging based on tumour characteristics and biomarker expression. Treatment includes surgery, radiotherapy, endocrine therapy, and chemotherapy, guided by prognostic factors and patient-specific considerations. While earlier stages have better prognosis, recurrence remains a risk. Adjuvant chemotherapy (chemotherapy following surgery) reduces recurrence risk but can lead to adverse events, necessitating a decision-making approach based on individual patient profiles. Multigene-expression tests, which assess genes related to proliferation and the oestrogen receptor pathway, have been developed and are primarily used to assist with recurrence risk prediction and, in some cases, to evaluate the potential benefit of adjuvant chemotherapy. In Switzerland, multigene-expression tests are temporarily covered by the mandatory health insurance since 2015 for patients with ER+/HER2- breast cancer with up to 3 affected lymph nodes, for whom the results of conventional testing alone do not allow a clear decision to be made regarding adjuvant chemotherapy. The 4 multigene-expression tests that are currently covered are: Oncotype DX, MammaPrint, EndoPredict, and Prosigna. OBJECTIVE This health technology assessment (HTA) report assesses the efficacy, effectiveness, safety, cost-effectiveness and budget impact as well as ethical, legal, social, and organisational benefits and harms of multigene-expression tests when applied as described in the reimbursement texts in Switzerland.

Authors' results and conclusions: 1. Clinical effectiveness In total, 25 studies on Oncotype DX (including 2 RCTs: RxPONDER and TAILORx), 9 studies on MammaPrint (including 1 RCT: MINDACT), 7 studies on EndoPredict, and 4 studies on Prosigna were included in the systematic review. An ongoing RCT was identified for Prosigna (OPTIMA). Of the 4 multigene-expression tests, only for MammaPrint evidence is available on the comparative clinical effectiveness of the multigene-expression test versus conventional testing for guiding adjuvant chemotherapy decisions in early breast cancer patients. The other studies on MammaPrint and the studies on Oncotype DX, EndoPredict and Prosigna reported intermediate measures on the predictive ability, prognostic ability, and impact on treatment management. The intermediate measures predictive and prognostic ability lack the comparison between the multigene-expression test versus conventional testing and in the impact on treatment management follow-up of clinical outcomes is not included. The predictive and prognostic ability of the multigene-expression tests were based on different survival outcomes. No data was reported on HRQoL. Except for a small number of studies on the impact on treatment management, none of the studies included the population for whom it was unclear based on conventional testing whether to prescribe adjuvant chemotherapy. Oncotype DX Comparative clinical effectiveness: No studies or ongoing RCTs identified. Intermediate measure – predictive ability/chemotherapy benefit: Early breast cancer patients classified by Oncotype DX in genomic low or intermediate risk may have little to no chemotherapy benefit on overall, breast cancer-specific, disease-free, recurrence-free, invasive disease-free, and distant metastasis-free survival (2 RCTs: moderate to low certainty evidence & 7 studies: very low certainty evidence). Early breast cancer patients classified by Oncotype DX in genomic high risk may have a chemotherapy benefit on overall, disease-free, recurrence-free, invasive disease-free, and distant metastasis-free survival, but the evidence is very uncertain (3 studies: very low certainty evidence). Intermediate measure – prognostic ability: Oncotype DX may be prognostic for invasive diseasefree survival in early breast cancer patients (1 RCT: low certainty evidence). Intermediate measure – impact on treatment management: Oncotype DX may change the adjuvant treatment recommendation in 31% of the early breast cancer patients, but the evidence is very uncertain (16 studies: very low certainty evidence). MammaPrint Comparative clinical effectiveness: The design of the MINDACT trial is aimed at assessing the clinical outcomes of treating early breast cancer patients with adjuvant chemotherapy based on their clinical risk (by Adjuvant!Online) or their genomic risk (by MammaPrint). Patients with discordant risks (i.e. C-high/G-low or C-low/G-high) were randomised to chemotherapy or no chemotherapy. In a prespecified secondary interim analysis not based on these discordant risk groups, the comparative clinical effectiveness of MammaPrint versus Adjuvant!Online was estimated by comparing distant metastasis-free survival resulting from treating according to genomic risk strategy or clinical risk strategy. Treatment based on MammaPrint may result in 46% fewer clinical high risk early breast cancer patients being treated with adjuvant chemotherapy and little difference in distant metastasis-free survival (1 RCT: low certainty evidence). The interim 8 years distant metastasis-free survival of adjuvant treatment based on MammaPrint compared to the clinical risk strategy was 90.9% versus 91.3%, and this slightly shorter survival becomes larger at 10 years follow-up. Whether adjuvant treatment based on MammaPrint is clinically beneficial depends on weighing the benefits (i.e. less adverse events due to adjuvant chemotherapy) and harms (i.e. possibly a slightly shorter distant metastasis-free survival) in a contextualised appraisal. This is not reported in the current MINDACT publications. Intermediate measure – predictive ability/chemotherapy benefit: Early breast cancer patients classified by Adjuvant!Online and MammaPrint in C-low/G-high may have little to no chemotherapy benefit on overall, disease-free and distant metastasis-free survival (1 RCT: low certainty evidence). Early breast cancer patients classified by Adjuvant!Online and MammaPrint in Chigh/G-low may have a chemotherapy benefit on distant metastasis-free survival and may have little to no chemotherapy benefit on overall and disease-free survival (1 RCT low certainty evidence). MammaPrint may not be predictive for chemotherapy benefit on breast cancer-specific survival in early breast cancer patients, but the evidence is very uncertain (1 study: very low certainty evidence). Intermediate measure – prognostic ability: MammaPrint may be prognostic for overall, diseasefree and distant metastasis-free survival in early breast cancer patients, but the evidence is very uncertain (1 RCT: very low certainty evidence). Intermediate measure – impact on treatment management: MammaPrint may change the adjuvant treatment recommendation in 35% of the early breast cancer patients, but the evidence is very uncertain (6 studies: very low certainty evidence). EndoPredict Comparative clinical effectiveness: No studies or ongoing RCTs identified. Intermediate measure – predictive ability/chemotherapy benefit: EndoPredict may be predictive for chemotherapy benefit on breast cancer and distant recurrence in early breast cancer patients, but the evidence is very uncertain (1 study: very low certainty evidence). Intermediate measure – prognostic ability: EndoPredict may be prognostic for breast cancer and distant recurrence in early breast cancer patients, but the evidence is very uncertain (3 studies: very low certainty evidence). Intermediate measure – impact on treatment management: EndoPredict may change the adjuvant treatment recommendation in 40% of the early breast cancer patients, but the evidence is very uncertain (4 studies: very low certainty evidence). Prosigna Comparative clinical effectiveness: No studies identified. An RCT is ongoing with an estimated completion date in December 2034 (5-year follow-up data is anticipated to be published mid2026). Intermediate measure – predictive ability/chemotherapy benefit: No studies identified. Intermediate measure – prognostic ability: Prosigna may be prognostic for breast cancer-specific and distant metastasis-free survival in early breast cancer patients, but the evidence is very uncertain (1 study: very low certainty evidence). Intermediate measure – impact on treatment management: Prosigna may change the adjuvant treatment recommendation in 17% of the early breast cancer patients, but the evidence is very uncertain (3 studies: very low certainty evidence). 2. Costs, cost-effectiveness and budget impact In the economic review 36 studies on multigene-expression tests for patients with HR+, HER2-, LN0-3 early breast cancer were included: 29 evaluated Oncotype DX, 10 evaluated MammaPrint, 6 evaluated EndoPredict, and 5 evaluated Prosigna. In cost-effectiveness results of the studies, Oncotype DX and MammaPrint ranged from dominant to dominated, while EndoPredict and Prosigna ranged from dominant to positive ICERs. The results of the cost-effectiveness analysis are very uncertain. It cannot be determined whether adding one of the evaluated multigene-expression tests to conventional testing will result in a health gain or loss, nor in an increase or decrease in healthcare costs. Results strongly depend on the recurrence risk in the target population and the probability that adjuvant chemotherapy is used under either testing strategy. The evidence base is insufficient to make a robust estimate of the impact of multigene-expression tests on health outcomes, costs, and budget impact in Switzerland. 3. Ethical, legal, social and organisational issues Fourteen articles on ELSO domains were included. In the ethical domain, inadequate representation of ethnic minorities as well as the exclusion of men from validation trials were discussed. No legal issues were found from the searches. Identified social issues included emotional stress that multigene-expression testing can cause to patients along with barriers that prevent patients from being screened, such as out-of-pocket expenses and privacy concerns. In the organizational domain, several factors i.e. gradual learning, patient preferences, and adherence to guidelines are discussed, providing insight into how they inform oncologists' decisions to order multigeneexpression tests and variously shape testing utilization patterns. In addition, challenges related to integrating multigene-expression tests into clinical practice, positioning general practitioners as solely emotional support providers due to the lack of expertise in these tests, as well as the impact of sample contamination and extraction methods on test accuracy were also addressed. CONCLUSION Overall, the comparative evidence on multigene-expression tests is sparse. Of the 4 multigeneexpression tests, only for MammaPrint an interim analysis of one RCT is available on the comparative clinical effectiveness of the multigene-expression test versus conventional testing for guiding adjuvant chemotherapy decisions in early breast cancer patients. Treatment based on MammaPrint may result in 46% fewer clinical high risk early breast cancer patients being treated with adjuvant chemotherapy and little difference in or a slightly shorter distant metastasis-free survival compared to treatment based on conventional testing. All other evidence is based on the intermediate measures predictive ability/chemotherapy benefit and prognostic ability, which lack the comparison between the multigene-expression test and conventional testing, and the intermediate measure impact on treatment management, which lacks follow-up of clinical outcomes. The effect of adding multigene-expression tests to conventional testing on costs and health outcomes are uncertain due to a lack of evidence on the comparative clinical effectiveness of multigene-expression tests in general, and data for the Swiss setting in particular. Finally, the use of multigene is associated with several ethical, social and organisational issues.

Authors' methods: 1. Clinical effectiveness A systematic literature search was conducted in PubMed (MEDLINE), Embase.com and Cochrane Library on 6 May 2024. Studies were selected by applying pre-specified inclusion criteria, study design and type of evidence (i.e. comparative clinical effectivenessa and the intermediate measures predictive abilitya, prognostic abilitya, impact on treatment managementa). Outcomesa included survival outcomes (e.g. overall survival), health-related quality of life (HRQoL) and change in treatment management. The risk of bias of includes studies was critically appraised with RoB 2, adapted PROBAST or the adapted ROBINS-I tool. The overall certainty of the evidence was assessed with GRADE. The included studies were analysed separately for each multigene-expression test. For the outcome impact on treatment management pooled event rates were calculated by meta-analysis. 2. Costs, cost-effectiveness and budget impact For the economic review, a systematic literature search was conducted in PubMed (MEDLINE), Embase.com, Cochrane Library, Tufts Medical Centre Cost-Effectiveness Analysis (CEA) Registry, and National Health Service Economic Evaluation Database (NHS EED). Studies were selected by applying pre-specified inclusion criteria. The quality of the studies was assessed using the Philips checklist for health economic models. Based on this quality assessment, one health economic model was selected to be replicated and adapted to the Swiss setting by incorporating Swiss data on population characteristics, survival, and treatment costs. An economic analysis taking a healthcare payers’ perspective and lifetime horizon was conducted in a target population representing the Swiss HR+, HER2-, LN0-3 early breast cancer population, as well as an intermediate risk population which was used to represent the population as defined in the reimbursement text. 3. Ethical, legal, social and organisational issues Ethical, legal, social, and organisational (ELSO) issues were searched through the systematic literature searches of the clinical effectiveness and cost-effectiveness in PubMed (MEDLINE), Embase.com, Cochrane Library, Tufts Medical Centre Cost-Effectiveness Analysis (CEA) Registry, and National Health Service Economic Evaluation Database (NHS EED) and targeted nonsystematic searches. Findings were described narratively.

Project Status: Completed

URL for project: https://www.bag.admin.ch/en/multigene-expression-tests-for-breast-cancer

URL for protocol: https://www.bag.admin.ch/dam/en/sd-web/1k6zMv5dwnJv/h0067gemk-hta-protocol.pdf

Year Published: 2025

URL for published report: https://www.bag.admin.ch/dam/en/sd-web/uM14XdydJKBC/H0067%20HTA%20Report%20Multigene-expression%20tests%20in%20early%20breast%20cancer%2020250905.pdf

English language abstract: An English language summary is available

Publication Type: Full HTA

Country: Switzerland

Organisation Name: Swiss Federal Office of Public Health (FOPH)

Contact Address: Federal Office of Public Health, Schwarzenburgstrasse 157, CH-3003 Berne, Switzerland

Contact Name: Stephanie Vollenweider

Contact Email: hta@bag.admin.ch

Copyright: Swiss Federal Office of Public Health