The MAMMACHECK project, funded by the National Association of German Breast Cancer Patients (NABF) under grant 03VP08290, ran from 1 October 2020 to 30 September 2023 at the University Hospital Freiburg’s Department of Gynecology. Its aim was to develop a non‑invasive, micro‑RNA‑based diagnostic test for breast cancer using urine samples. The study was led by Prof. Dr. med. Thalia Erbes and involved three core working groups within the university: the Molecular Oncology Laboratory at the Women’s Clinic, the Clinical Trials Center (ZKS), and the Institute of Microsystems Engineering at the Chair of Applied Development. An additional advisory role was provided by the Methods of Systems Biomedicine Group at the Institute of Medical Biometry and Statistics (IMBI). To increase patient recruitment, a second entry point was established at the St. Vincentius Clinics in Karlsruhe in spring 2021, which raised the weekly enrollment of breast‑cancer patients to five or six and met the milestone requirement for validation sample size.
By the end of the reporting period, 1 985 urine samples had been collected, comprising 1 565 from healthy controls and 420 from breast‑cancer patients. The samples were processed on a newly acquired Roche Lightcycler 480 Real‑Time PCR system, specifically chosen for its suitability to urine‑derived RNA. Initial analyses in early 2022 revealed only a weak separation between cancer and control groups. Consequently, extensive optimisation steps were undertaken: the new qPCR platform was benchmarked against the previous system; filter columns with different chemistries (SiO₂ versus SiC) were compared to identify the optimal purification method; and the composition of RNA‑isolation buffers and elution solutions was refined and standardised. These adjustments improved the sensitivity of the assay, but a subsequent interim analysis of 123 samples in May 2022 showed that the statistical separation deteriorated as the sample size increased, indicating that the existing biomarker panel required further refinement.
In late 2022, Next‑Generation Sequencing (NGS) studies were commissioned from Starseq and GenXPro, each analysing 48 samples (24 breast‑cancer and 24 healthy). The sequencing data led to the expansion of the biomarker panel from an initial set to 17 candidate micro‑RNAs. A pre‑validation phase in early 2023 tested these 17 miRNAs on 181 urine samples (84 cancer, 97 healthy) using RT‑qPCR. Logistic regression modelling of the full panel yielded an area under the receiver‑operator characteristic curve (AUC) of 0.80, while the top four miRNAs achieved an AUC of 0.72. Despite the slightly lower performance of the reduced panel, the decision was made to evaluate all 17 miRNAs across the full cohort of 1 687 valid urine samples, thereby maximizing the potential for robust biomarker discovery.
The collaborative framework of the project integrated clinical expertise from the Women’s Clinic, methodological support from the Clinical Trials Center, and engineering contributions from the Microsystems Institute, with statistical guidance from IMBI. The partnership with the St. Vincentius Clinics expanded the patient base and ensured a balanced representation of breast‑cancer and healthy individuals. Throughout the three‑year period, the project maintained rigorous ethical oversight, extending the original ethics approval to accommodate the additional recruitment site. The combined efforts of these multidisciplinary teams culminated in a comprehensive dataset and a refined micro‑RNA panel that advances the feasibility of a urine‑based, non‑invasive breast‑cancer screening tool.
