Scholarlink

Objective <p>Radiogenomics aims to non-invasively predict tumour genotypes from imaging, but most studies assume molecular homogeneity by assigning a single biopsy-derived label to all lesions within a patient. This approach risks substantial label noise given well-documented interlesional heterogeneity. We investigated whether anchoring training to biopsy-confirmed lesions improves radiogenomic model performance and generalisability.</p> Materials and methods <p>We retrospectively analysed 1646 patients (11473 segmented lesions) with contrast-enhanced CT and EGFR mutation status from next-generation sequencing at the Netherlands Cancer Institute, alongside an external NSCLC radiogenomics cohort (<i>n</i> = 158). All visible lesions were segmented, and the exact biopsy site was matched to its segmentation. Radiomic features were extracted, and machine learning models were trained with three lesion selection strategies: all lesions, non-biopsied lesions only, and biopsy-confirmed lesions only. To disentangle label quality from sample size, we created size-matched variants (one lesion per patient) for all-lesion and non-biopsied strategies.</p> Results <p>All models achieved significant discrimination of EGFR status on internal validation (AUC = 0.62–0.68). However, performance of the all-lesion and non-biopsied models declined on external validation (AUC = 0.55–0.63), while the biopsy-anchored model maintained stable performance (AUC = 0.62), despite having only 1/10th of the training sample size. When training sets were size-matched, the biopsy-anchored approach significantly outperformed a model trained on all available lesions on external validation (<i>p</i> = 0.037).</p> Conclusions <p>Radiogenomic models trained on biopsy-confirmed lesions outperform conventional all-lesion strategies in external validation, despite using an order of magnitude fewer samples. Prioritising lesion-level label fidelity can mitigate heterogeneity-driven noise, enhancing robustness and clinical translation of imaging-based genomic prediction.</p> Key Points <p><Emphasis Type="BoldItalic">Question</Emphasis> <i>Does assigning biopsy-derived molecular labels to all lesions introduce heterogeneity-driven label noise that reduces the generalisability of radiogenomic models?</i></p> <p><Emphasis Type="BoldItalic">Findings</Emphasis> <i>Models trained exclusively on biopsy-confirmed lesions demonstrated superior external generalisability compared with all-lesion approaches, despite being trained on substantially fewer samples.</i></p> <p><Emphasis Type="BoldItalic">Clinical relevance</Emphasis> <i>Biopsy-anchored radiogenomics improves the reliability of non-invasive mutation prediction by accounting for tumour heterogeneity, potentially supporting clinical decision-making when tissue sampling is limited or molecular results are discordant across lesions.</i></p> Graphical Abstract <p></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

Quality over quantity: biopsy-anchored CT radiogenomics models outperform all-lesion training in a multi-tumour cohort despite a smaller sample size

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月16日

查看全文
  • Diana Ivonne Rodríguez Sánchez,
  • Julian Middelkoop,
  • Thera Vanneste,
  • Olga Maxouri,
  • Stephan Ursprung,
  • Sajjad Rostami,
  • Nino Bogveradze,
  • Kalina Chupetlovska,
  • Francesca Castagnoli,
  • Federica Landolfi,
  • Eun Kyoung Hong,
  • Andrea Delli Pizzi,
  • Nicolo Gennaro,
  • Warissara Jutidamrongphan,
  • Liliana Petrychenko,
  • Petur Snaebjornsson,
  • Zuhir Bodalal,
  • Regina Beets-Tan

摘要

Objective

Radiogenomics aims to non-invasively predict tumour genotypes from imaging, but most studies assume molecular homogeneity by assigning a single biopsy-derived label to all lesions within a patient. This approach risks substantial label noise given well-documented interlesional heterogeneity. We investigated whether anchoring training to biopsy-confirmed lesions improves radiogenomic model performance and generalisability.

Materials and methods

We retrospectively analysed 1646 patients (11473 segmented lesions) with contrast-enhanced CT and EGFR mutation status from next-generation sequencing at the Netherlands Cancer Institute, alongside an external NSCLC radiogenomics cohort (n = 158). All visible lesions were segmented, and the exact biopsy site was matched to its segmentation. Radiomic features were extracted, and machine learning models were trained with three lesion selection strategies: all lesions, non-biopsied lesions only, and biopsy-confirmed lesions only. To disentangle label quality from sample size, we created size-matched variants (one lesion per patient) for all-lesion and non-biopsied strategies.

Results

All models achieved significant discrimination of EGFR status on internal validation (AUC = 0.62–0.68). However, performance of the all-lesion and non-biopsied models declined on external validation (AUC = 0.55–0.63), while the biopsy-anchored model maintained stable performance (AUC = 0.62), despite having only 1/10th of the training sample size. When training sets were size-matched, the biopsy-anchored approach significantly outperformed a model trained on all available lesions on external validation (p = 0.037).

Conclusions

Radiogenomic models trained on biopsy-confirmed lesions outperform conventional all-lesion strategies in external validation, despite using an order of magnitude fewer samples. Prioritising lesion-level label fidelity can mitigate heterogeneity-driven noise, enhancing robustness and clinical translation of imaging-based genomic prediction.

Key Points

Question Does assigning biopsy-derived molecular labels to all lesions introduce heterogeneity-driven label noise that reduces the generalisability of radiogenomic models?

Findings Models trained exclusively on biopsy-confirmed lesions demonstrated superior external generalisability compared with all-lesion approaches, despite being trained on substantially fewer samples.

Clinical relevance Biopsy-anchored radiogenomics improves the reliability of non-invasive mutation prediction by accounting for tumour heterogeneity, potentially supporting clinical decision-making when tissue sampling is limited or molecular results are discordant across lesions.

Graphical Abstract