Scholarlink

<p>Precision oncology relies on access to high-quality data for increasingly smaller patient subgroups. The international atomCAT consortium investigates the potential of federated learning to support this, using anal cancer as a rare cancer exemplar. Here, we show that federated multivariable Cox models trained across 14 centres (1428 patients) and externally validated in two additional centres (277 patients) achieve consistent calibration and discrimination during leave-one-centre-out and external validation (c-indices 0.68-0.79). Lower T stage, absence of nodal involvement, smaller tumour volume, female sex, younger age, and mitomycin- or cisplatin-based chemotherapy are associated with improved overall survival. Lower T stage, smaller tumour volume, and female sex are associated with improved locoregional control, while absence of nodal involvement and smaller tumour volume are associated with better freedom from distant metastases. These findings demonstrate that federated learning enables robust, privacy-preserving prognostic modelling for rare cancers using real-world data, supporting international collaboration without data sharing.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An international multi-centre study to develop and validate federated learning-based prognostic models for anal cancer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月14日

查看全文
  • Stelios Theophanous,
  • Per-Ivar Lønne,
  • Ananya Choudhury,
  • Maaike Berbee,
  • Charlotte Deijen,
  • Andre Dekker,
  • Matthew Field,
  • Maria Antonietta Gambacorta,
  • Alexandra Gilbert,
  • Marianne Grønlie Guren,
  • Rashmi Jadon,
  • Rohit Kochhar,
  • Daniel Martin,
  • Ahmed Allam Mohamed,
  • Rebecca Muirhead,
  • Oriol Parés,
  • Łukasz Raszewski,
  • Rajarshi Roy,
  • Andrew Scarsbrook,
  • David Sebag-Montefiore,
  • Emiliano Spezi,
  • Vassilios Vassiliou,
  • Eirik Malinen,
  • Leonard Wee,
  • Ane Appelt,
  • Richard Adams,
  • Krystyna Adamska,
  • Muhammad Amin,
  • Nikola Dino Capocchiano,
  • Philip Chlap,
  • Peter Colley,
  • Andrea Damiani,
  • Viola De Luca,
  • Antri Demetriou,
  • Shrikant Deshpande,
  • Michael J. Eble,
  • Anthony Espinoza,
  • Emmanouil Fokas,
  • Loukia Georgiou,
  • Ann Henry,
  • Andrew Hoole,
  • Lois C. Holloway,
  • Thomas Jansen,
  • Tomas Janssen,
  • Alexandros Kritikopoulos,
  • Joanna Y. G. Lau,
  • Mark T. Lee,
  • John Lilley,
  • Gisela Lima,
  • Patricia Lopes,
  • Adam Loveday,
  • Stefania Manfrida,
  • Jenny Marsden,
  • Carlotta Masciocchi,
  • Joseph Mercer,
  • Elisavet Papageorgiou,
  • Gareth Price,
  • Thomas Rackley,
  • Claus Michael Rödel,
  • Mariachiara Savino,
  • Athina Sdrolia,
  • Joep Stroom,
  • Ioannis Stylianou,
  • David Thwaites,
  • Maciej Trojanowski,
  • Vincenzo Valentini,
  • Rens van Haveren,
  • Baukelien van Triest,
  • Amy Walker

摘要

Precision oncology relies on access to high-quality data for increasingly smaller patient subgroups. The international atomCAT consortium investigates the potential of federated learning to support this, using anal cancer as a rare cancer exemplar. Here, we show that federated multivariable Cox models trained across 14 centres (1428 patients) and externally validated in two additional centres (277 patients) achieve consistent calibration and discrimination during leave-one-centre-out and external validation (c-indices 0.68-0.79). Lower T stage, absence of nodal involvement, smaller tumour volume, female sex, younger age, and mitomycin- or cisplatin-based chemotherapy are associated with improved overall survival. Lower T stage, smaller tumour volume, and female sex are associated with improved locoregional control, while absence of nodal involvement and smaller tumour volume are associated with better freedom from distant metastases. These findings demonstrate that federated learning enables robust, privacy-preserving prognostic modelling for rare cancers using real-world data, supporting international collaboration without data sharing.