Institute of Pathology, Charité Universitätsmedizin, Berlin, Germany; BIFOLD, Berlin, Germany; Aignostics GmbH, Berlin, Germany.
Institute of Pathology, University Hospital Heidelberg, Heidelberg, Germany; Translational Lung Research Center Heidelberg (TLRC-H), Member of the German Center for Lung Research (DZL), 69120 Heidelberg, Germany.
Eur J Cancer. 2024 Nov;211:114292. doi: 10.1016/j.ejca.2024.114292. Epub 2024 Aug 23.
Molecular profiling of lung cancer is essential to identify genetic alterations that predict response to targeted therapy. While deep learning shows promise for predicting oncogenic mutations from whole tissue images, existing studies often face challenges such as limited sample sizes, a focus on earlier stage patients, and insufficient analysis of robustness and generalizability.
This retrospective study evaluates factors influencing mutation prediction accuracy using the large Heidelberg Lung Adenocarcinoma Cohort (HLCC), a cohort of 2356 late-stage FFPE samples. Validation is performed in the publicly available TCGA-LUAD cohort.
Models trained on the larger HLCC cohort generalized well to the TCGA dataset for mutations in EGFR (AUC 0.76), STK11 (AUC 0.71) and TP53 (AUC 0.75), in line with the hypothesis that larger cohort sizes improve model robustness. Variation in performance due to pre-processing and modeling choices, such as mutation variant calling, affected EGFR prediction accuracy by up to 7 %.
Model explanations suggest that acinar and papillary growth patterns are critical for the detection of EGFR mutations, whereas solid growth patterns and large nuclei are indicative of TP53 mutations. These findings highlight the importance of specific morphological features in mutation detection and the potential of deep learning models to improve mutation prediction accuracy.
Although deep learning models trained on larger cohorts show improved robustness and generalizability in predicting oncogenic mutations, they cannot replace comprehensive molecular profiling. However, they may support patient pre-selection for clinical trials and deepen the insight in genotype-phenotype relationships.
对肺癌进行分子谱分析对于确定预测靶向治疗反应的遗传改变至关重要。虽然深度学习在从全组织图像预测致癌突变方面显示出前景，但现有研究通常面临挑战，例如样本量有限、关注早期患者以及对稳健性和通用性的分析不足。
本回顾性研究使用大型海德堡肺腺癌队列（HLCC）评估了影响突变预测准确性的因素，该队列包含 2356 例晚期 FFPE 样本。在公开的 TCGA-LUAD 队列中进行验证。
在 TCGA 数据集上，在更大的 HLCC 队列上训练的模型对 EGFR（AUC 0.76）、STK11（AUC 0.71）和 TP53（AUC 0.75）的突变具有良好的通用性，这与更大的队列规模提高模型稳健性的假设一致。由于预处理和建模选择（如突变变体调用）的差异，性能的变化最多可影响 EGFR 预测准确性 7%。
模型解释表明，腺泡和乳头状生长模式对于检测 EGFR 突变至关重要，而实性生长模式和大核是 TP53 突变的指征。这些发现强调了特定形态特征在突变检测中的重要性以及深度学习模型在提高突变预测准确性方面的潜力。
尽管在预测致癌突变方面，经过更大队列训练的深度学习模型显示出了更好的稳健性和通用性，但它们不能替代全面的分子谱分析。然而，它们可以支持患者在临床试验中的预先选择，并加深对基因型-表型关系的了解。
