Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Epidemiology

Development of radiomics-based models on mammograms with mass lesions to predict prognostically relevant characteristics of invasive breast cancer in a screening cohort

British Journal of Cancer volume 132, pages 1040–1049 (2025)Cite this article

Subjects

Abstract

Background

Optimizing breast-screening performance involves minimizing overdiagnosis of prognostically favorable invasive breast cancer (IBC) that does not need immediate recall and underdiagnosis of prognostically unfavorable IBC that is not recalled timely. We investigated whether mammographic features of masses predict prognostically relevant IBC characteristics.

Methods

In a screening cohort, we obtained pathological information of 1587 IBCs presenting as a mass through the nationwide cancer registry and pathology databank. We developed models based on mammographic tumor appearance to predict whether IBC was prognostically favorable (T1N0M0 luminal A-like) or unfavorable. Models were based on 1095 positive screening mammograms (possible overdiagnosis), or on 603 last negative mammograms with in retrospect visible masses (possible underdiagnosis). We calculated performance metrics using cross-validation.

Results

23.5% of masses were prognostically favorable IBC. Using 1095 positive mammograms, the model’s predictions to have prognostically favorable IBC (10th–90th percentile range 8.7–47.0%) yielded AUC 0.75 (SD across repeats 0.01), slope 1.16 (SD 0.07). Performance in 603 last negative screening mammograms with masses was poor: AUC 0.60 (SD 0.02), slope 0.85 (SD 0.28).

Conclusions

Mammography-based models from masses representing IBC at time of recall (possible overdiagnosis) predict prognostically relevant characteristics of IBC. Models based on in retrospect visible masses (possible underdiagnosis) performed poorly.

Access through your institution
Buy or subscribe

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Flowchart showing the source population, study population and masses included for model development and evaluation.
Fig. 2: Schematic overview of candidate predictor sets, mammogram sets used for model development and evaluation, outcome definitions and relations to aims.
Fig. 3: Calibration plot based on repeated cross-validation for a combined model (radiomics, volume and SGR) to predict whether IBC is prognostically favorable.
Fig. 4: Calibration plots based on repeated cross-validation for a combined model (radiomics, volume and SGR) to predict the PREDICT-risk of breast cancer death.

Similar content being viewed by others

Data availability

The data that supports the findings of this study are available from third parties (Foundation of Population Screening Mid-West, the Netherlands Comprehensive Cancer Organisation (IKNL), the Ducth Nationwide Pathology Databank (Palga), Antonius Ziekenhuis (Nieuwegein, The Netherlands), Diakonessenhuis (Utrecht, The Netherlands) and UMC Utrecht (Utrecht, The Netherlands)). Restrictions apply to the availability of these data, which were used under license for this study. Researchers may contact the corresponding author upon reasonable request to access data with the permission of the third parties involved.

References

  1. Marmot MG, Altman D, Cameron D, Dewar J, Thompson S, Wilcox M. The benefits and harms of breast cancer screening: an independent review. Br J cancer. 2013;108:2205–40.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Ding L, Poelhekken K, Greuter MJ, Truyen I, De Schutter H, Goossens M, et al. Overdiagnosis of invasive breast cancer in population-based breast cancer screening: A short-and long-term perspective. Eur J Cancer. 2022;173:1–9.

    Article  PubMed  Google Scholar 

  3. Broeders MJ, Onland-Moret NC, Rijken HJ, Hendriks JH, Verbeek AL, Holland R. Use of previous screening mammograms to identify features indicating cases that would have a possible gain in prognosis following earlier detection. Eur J Cancer. 2003;39:1770–5.

    Article  CAS  PubMed  Google Scholar 

  4. Prat A, Chaudhury A, Solovieff N, Paré L, Martinez D, Chic N, et al. Correlative biomarker analysis of intrinsic subtypes and efficacy across the MONALEESA phase III studies. J Clin Oncol. 2021;39:1458.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Weiss A, Chavez-MacGregor M, Lichtensztajn DY, Yi M, Tadros A, Hortobagyi GN, et al. Validation study of the American Joint Committee on Cancer eighth edition prognostic stage compared with the anatomic stage in breast cancer. JAMA Oncol. 2018;4:203–9.

    Article  PubMed  Google Scholar 

  6. Wishart GC, Azzato EM, Greenberg DC, Rashbass J, Kearins O, Lawrence G, et al. PREDICT: a new UK prognostic model that predicts survival following surgery for invasive breast cancer. Breast Cancer Res. 2010;12:1–10.

    Article  Google Scholar 

  7. Tagliafico AS, Bignotti B, Rossi F, Matos J, Calabrese M, Valdora F, et al. Breast cancer Ki-67 expression prediction by digital breast tomosynthesis radiomics features. Eur Radiol Exp. 2019;3:1–6.

    Article  Google Scholar 

  8. Moshina N, Backmann HA, Skaane P, Hofvind S. Mammographic features and risk of breast cancer death among women with invasive screen-detected cancer in BreastScreen Norway 1996–2020. Eur Radiol. 2024;34:3364–74.

    Article  PubMed  Google Scholar 

  9. Killelea BK, Chagpar AB, Bishop J, Horowitz NR, Christy C, Tsangaris T, et al. Is there a correlation between breast cancer molecular subtype using receptors as surrogates and mammographic appearance? Ann surgical Oncol. 2013;20:3247–53.

    Article  Google Scholar 

  10. Alexander MC, Yankaskas BC, Biesemier KW. Association of stellate mammographic pattern with survival in small invasive breast tumors. Am J Roentgenol. 2006;187:29–37.

    Article  Google Scholar 

  11. Peters J, van Dijck JAAM, Elias SG, Otten JDM, Broeders MJM, IMAGINE Consortium. The prognostic potential of mammographic growth rate of invasive breast cancer in the Nijmegen breast cancer screening cohort. J Med Screening. 2024;31:66–175.

  12. Sechopoulos I, Teuwen J, Mann R, editors. Artificial intelligence for breast cancer detection in mammography and digital breast tomosynthesis: State of the art. Seminars in Cancer Biology; 2021: Elsevier.

  13. Casparie M, Tiebosch A, Burger G, Blauwgeers H, Van de Pol A, Van Krieken J, et al. Pathology databanking and biobanking in The Netherlands, a central role for PALGA, the nationwide histopathology and cytopathology data network and archive. Anal Cell Pathol. 2007;29:19–24.

    Article  CAS  Google Scholar 

  14. Sturesdotter L, Sandsveden M, Johnson K, Larsson A-M, Zackrisson S, Sartor H. Mammographic tumour appearance is related to clinicopathological factors and surrogate molecular breast cancer subtype. Sci Rep. 2020;10:20814.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Lång K, Josefsson V, Larsson A-M, Larsson S, Högberg C, Sartor H, et al. Artificial intelligence-supported screen reading versus standard double reading in the Mammography Screening with Artificial Intelligence trial (MASAI): a clinical safety analysis of a randomised, controlled, non-inferiority, single-blinded, screening accuracy study. Lancet Oncol. 2023;24:936–44.

    Article  PubMed  Google Scholar 

  16. Ehinger A, Malmström P, Bendahl P-O, Elston CW, Falck A-K, Forsare C, et al. Histological grade provides significant prognostic information in addition to breast cancer subtypes defined according to St Gallen 2013. Acta oncologica. 2017;56:68–74.

    Article  CAS  PubMed  Google Scholar 

  17. Goldhirsch A, Winer EP, Coates AS, Gelber RD, Piccart-Gebhart M, Thurlimann B, et al. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol. 2013;24:2206–23.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. van Maaren MC, van Steenbeek CD, Pharoah PD, Witteveen A, Sonke GS, Strobbe LJ, et al. Validation of the online prediction tool PREDICT v. 2.0 in the Dutch breast cancer population. Eur J cancer. 2017;86:364–72.

    Article  PubMed  Google Scholar 

  19. Niu S, Jiang W, Zhao N, Jiang T, Dong Y, Luo Y, et al. Intra-and peritumoral radiomics on assessment of breast cancer molecular subtypes based on mammography and MRI. J Cancer Res Clin Oncol. 2022;148:7–106.

  20. Ding J, Chen S, Sosa MS, Cattell R, Lei L, Sun J, et al. Optimizing the peritumoral region size in radiomics analysis for sentinel lymph node status prediction in breast cancer. Academic Radiol. 2022;29:S223–S8.

    Article  Google Scholar 

  21. Mao N, Shi Y, Lian C, Wang Z, Zhang K, Xie H, et al. Intratumoral and peritumoral radiomics for preoperative prediction of neoadjuvant chemotherapy effect in breast cancer based on contrast-enhanced spectral mammography. Eur Radiol. 2022;32:3207–19.

    Article  CAS  PubMed  Google Scholar 

  22. Verboom SD, Caballo M, Peters J, Gommers J, van den Oever D, Broeders MJ, et al. Deep learning-based breast region segmentation in raw and processed digital mammograms: generalization across views and vendors. J Med Imaging. 2024;11:014001–014001.

    Google Scholar 

  23. Zheng Y, Keller BM, Ray S, Wang Y, Conant EF, Gee JC, et al. Parenchymal texture analysis in digital mammography: A fully automated pipeline for breast cancer risk assessment. Med Phys. 2015;42:4149–60.

    Article  PubMed  PubMed Central  Google Scholar 

  24. van Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational Radiomics System to Decode the Radiographic Phenotype. Cancer Res. 2017;77:e104–e7.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Zwanenburg A, Vallières M, Abdalah MA, Aerts HJ, Andrearczyk V, Apte A, et al. The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology. 2020;295:328–38.

    Article  PubMed  Google Scholar 

  26. Moriakov N, Peters J, Mann R, Karssemeijer N, van Dijck J, Broeders M, et al. Improving Lesion Volume Measurements on Digital Mammograms. Med Image Anal. 2024;97:103269.

    Article  PubMed  Google Scholar 

  27. Otten JD, van Schoor G, Peer PG, den Heeten GJ, Holland R, Broeders MJ, et al. Growth rate of invasive ductal carcinomas from a screened 50–74-year-old population. J Med Screen. 2018;25:40–6.

    Article  PubMed  Google Scholar 

  28. Chen T, Guestrin C, editors. Xgboost: A scalable tree boosting system. Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining; 2016.

  29. Krstajic D, Buturovic LJ, Leahy DE, Thomas S. Cross-validation pitfalls when selecting and assessing regression and classification models. J cheminformatics. 2014;6:1–15.

    Article  Google Scholar 

  30. Zhou J, Tan H, Bai Y, Li J, Lu Q, Chen R, et al. Evaluating the HER-2 status of breast cancer using mammography radiomics features. Eur J Radiol. 2019;121:108718.

    Article  PubMed  Google Scholar 

  31. Wang L, Yang W, Xie X, Liu W, Wang H, Shen J, et al. Application of digital mammography-based radiomics in the differentiation of benign and malignant round-like breast tumors and the prediction of molecular subtypes. Gland Surg. 2020;9:2005.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Ma W, Zhao Y, Ji Y, Guo X, Jian X, Liu P, et al. Breast cancer molecular subtype prediction by mammographic radiomic features. Academic Radiol. 2019;26:196–201.

    Article  Google Scholar 

  33. Zhang T, Tan T, Samperna R, Li Z, Gao Y, Wang X, et al. Radiomics and artificial intelligence in breast imaging: a survey. Artif Intell Rev. 2023;56:857–92.

    Article  Google Scholar 

  34. Ueda D, Yamamoto A, Takashima T, Onoda N, Noda S, Kashiwagi S, et al. Training, validation, and test of deep learning models for classification of receptor expressions in breast cancers from mammograms. JCO Precision Oncol. 2021;5:543–51.

  35. Panambur AB, Madhu P, Maier A. Classification of Luminal Subtypes in Full Mammogram Images Using Transfer Learning. arXiv preprint arXiv:230109282. 2023.

  36. Luo C, Zhao S, Peng C, Wang C, Hu K, Zhong X, et al. Mammography radiomics features at diagnosis and progression-free survival among patients with breast cancer. Br J Cancer. 2022;127:1886–92.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Neri A, Marrelli D, Megha T, Bettarini F, Tacchini D, De Franco L, et al. Clinical significance of multifocal and multicentric breast cancers and choice of surgical treatment: a retrospective study on a series of 1158 cases. BMC Surg. 2015;15:1–10.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Li H, Robinson K, Lan L, Baughan N, Chan C-W, Embury M, et al. Temporal Machine Learning Analysis of Prior Mammograms for Breast Cancer Risk Prediction. Cancers. 2023;15:2141.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Baeßler B, Weiss K, Dos Santos DP. Robustness and reproducibility of radiomics in magnetic resonance imaging: a phantom study. Investigative Radiol. 2019;54:221–8.

    Article  Google Scholar 

  40. Van Luijt P, Heijnsdijk E, Fracheboud J, Overbeek L, Broeders M, Wesseling J, et al. The distribution of ductal carcinoma in situ (DCIS) grade in 4232 women and its impact on overdiagnosis in breast cancer screening. Breast Cancer Res. 2016;18:1–10.

    Google Scholar 

  41. Libesman S, Zackrisson S, Hofvind S, Seidler AL, Bernardi D, Lång K, et al. An individual participant data meta-analysis of breast cancer detection and recall rates for digital breast tomosynthesis versus digital mammography population screening. Clin Breast Cancer. 2022;22:e647–e54.

    Article  CAS  PubMed  Google Scholar 

  42. Jiang T, Song J, Wang X, Niu S, Zhao N, Dong Y, et al. Intratumoral and peritumoral analysis of mammography, tomosynthesis, and multiparametric MRI for predicting Ki-67 level in breast cancer: a radiomics-based study. Molecular Imaging and Biology. 2022;24:550–9.

Download references

Acknowledgements

The authors thank the Foundation of Population Breast Cancer Screening Mid-West for supporting this study and providing mammograms, the registration team of the Netherlands Comprehensive Cancer Organisation (IKNL) for the collection of data for the Netherlands Cancer Registry as well as IKNL staff for scientific advice, the Dutch Nationwide Pathology Databank (Palga) for providing excerpts and intermediating to obtain mammograms and tumor blocks, participating hospitals Antonius Ziekenhuis, Diakonessenhuis and UMC Utrecht for providing clinical mammograms, radiologists E.J.M. Wolters-van de Ben, St. Antonius Ziekenhuis; L.M. Jongen, Diakonessenhuis; W. B. Veldhuis, UMC Utrecht; and pathologist H.J. van Slooten, St Antonius Ziekenhuis, for coordinating mammogram and tumor blocks collection at participating institutions, research assistants M. Eijgenberger and I. van de Kamp for lesion annotation of mammograms, J. Sanders for revisions of tumor blocks and the NKI Core Facility Molecular Pathology & Biobanking (CFMB) for additional ER and Her2 staining. All members of the IMAGINE consortium (C.H. van Gils, R.M. Pijnappel, M. van Oirsouw, E. Verschuur, J. Peters, M. van Leeuwen, N. Moriakov, J.A.A.M. van Dijck, R. M. Mann, J. Teuwen, EH. Lips, A. W. van den Belt-Dusebout, J. Wesseling, B.B.L. Penning de Vries, N. Karssemeijer, S. G. Elias & M.J.M. Broeders) have contributed to this work.

Funding

This IMAGINE study is financially supported by the Dutch Cancer Society (grant number 11835, to M. Broeders). This work was furthermore funded by the Dutch Research Council (ZonMW) for the Breast-CARE project (grant number 5550004201, to J. Wesseling). Research at the Netherlands Cancer Institute is supported by institutional grants from the Dutch Cancer Society and the Dutch Ministry of Health, Welfare and Sport.

Author information

Authors and Affiliations

  1. Department IQ Health, Radboud University Medical Center, Nijmegen, Netherlands

    Jim Peters, Jos A. A. M. van Dijck & Mireille J. M. Broeders

  2. Division of Molecular Pathology, Netherlands Cancer Institute (NKI), Amsterdam, Netherlands

    Merle M. van Leeuwen, Esther H. Lips, Alexandra W. van den Belt-Dusebout & Jelle Wesseling

  3. Department of Radiation Oncology, Netherlands Cancer Institute (NKI), Amsterdam, Netherlands

    Nikita Moriakov & Jonas Teuwen

  4. Department of Radiology, Netherlands Cancer Institute (NKI), Amsterdam, Netherlands

    Nikita Moriakov & Ritse M. Mann

  5. Department of Medical Imaging, Radboud University Medical Center, Nijmegen, Netherlands

    Ritse M. Mann, Sarah Verboom & Nico Karssemeijer

  6. Department of Pathology, Netherlands Cancer Institute (NKI), Amsterdam, Netherlands

    Jelle Wesseling

  7. Department of Pathology, Leiden University Medical Center, Leiden, Netherlands

    Jelle Wesseling

  8. Julius Center for Health Sciences and Primary Care, UMC Utrecht, University Utrecht, Utrecht, Netherlands

    Bas B. L. Penning de Vries & Sjoerd G. Elias

  9. Dutch Expert Centre for Screening, Nijmegen, Netherlands

    Mireille J. M. Broeders

Authors
  1. Jim Peters
    View author publications

    Search author on:PubMed Google Scholar

  2. Merle M. van Leeuwen
    View author publications

    Search author on:PubMed Google Scholar

  3. Nikita Moriakov
    View author publications

    Search author on:PubMed Google Scholar

  4. Jos A. A. M. van Dijck
    View author publications

    Search author on:PubMed Google Scholar

  5. Ritse M. Mann
    View author publications

    Search author on:PubMed Google Scholar

  6. Jonas Teuwen
    View author publications

    Search author on:PubMed Google Scholar

  7. Esther H. Lips
    View author publications

    Search author on:PubMed Google Scholar

  8. Alexandra W. van den Belt-Dusebout
    View author publications

    Search author on:PubMed Google Scholar

  9. Jelle Wesseling
    View author publications

    Search author on:PubMed Google Scholar

  10. Bas B. L. Penning de Vries
    View author publications

    Search author on:PubMed Google Scholar

  11. Sarah Verboom
    View author publications

    Search author on:PubMed Google Scholar

  12. Nico Karssemeijer
    View author publications

    Search author on:PubMed Google Scholar

  13. Sjoerd G. Elias
    View author publications

    Search author on:PubMed Google Scholar

  14. Mireille J. M. Broeders
    View author publications

    Search author on:PubMed Google Scholar

Contributions

JD, RM, JT, EL, JW, SE, NK and MB conceptualized the study and acquired funding. JP, JD, JT, AB, NK and MB obtained screening and clinical mammograms. JP, JD and MB obtained clinical and pathological information from national registries. JP and ML curated pathological information and coordinated pathological review of tumor blocks, supervised by EL, AB and JW. JP, NM, RM, JT, SV and NK applied and/or developed methods to extract radiomics and other handcrafted features from mammograms. JP, JD, BP, SE and MB conceived and designed the statistical analysis plan. JP carried out the analysis and wrote code for prediction modeling, supervised by BP and SE. All authors were involved in interpretation of results. JP prepared the original draft. All authors reviewed and edited previous versions of the manuscript and approved the final version of the manuscript.

Corresponding author

Correspondence to Jim Peters.

Ethics declarations

Competing interests

The authors declare the following relationships: N.K. is board member and shareholder of ScreenPoint Medical and shareholder of Volpara Technology. RM. received research grants or equipment from Bayer, Siemens, ScreenPoint Medical, Koning, Becton Dickinson, PA Imaging and Lunit and provided consultancy for ScreenPoint Medical, Becton Dickinson, Bayer, Guerbet, Bracco and Siemens. All other authors declare no potential conflict of interest

Ethics approval and consent to participate

This study was conducted in accordance with the declaration of Helsinki. No explicit written or verbal consent was obtained, but consent was obtained through an opt-out procedure that exists in the Dutch screening program. When participating in the program, women are informed that their data can be used for evaluation of the program or scientific purposes to improve the program. If women chose to opt-out, their data was not used. The Radboud university medical center ethics committee declared that this study falls outside the scope of the Dutch Medical Research involving Human Subjects Act and could be carried out without approval of an Institutional Review Board.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Peters, J., van Leeuwen, M.M., Moriakov, N. et al. Development of radiomics-based models on mammograms with mass lesions to predict prognostically relevant characteristics of invasive breast cancer in a screening cohort. Br J Cancer 132, 1040–1049 (2025). https://doi.org/10.1038/s41416-025-02995-6

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41416-025-02995-6

This article is cited by

Search

Advanced search

Quick links