PO.BCS01.06 · 生物信息与计算
基于H&E组织切片深度学习预测晚期EGFR突变型非小细胞肺癌对EGFR抑制剂的无进展生存期
Deep-learning prediction of progression-free survival to EGFR inhibitors from H&E tissue slides in advanced EGFR -mutated non-small cell lung cancer
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
背景：奥希替尼（osimertinib）等第三代酪氨酸激酶抑制剂（TKI）是伴有常见EGFR突变（19号外显子缺失/L858R）的未经治疗晚期非小细胞肺癌（aNSCLC）的标准治疗。与化疗、双特异性抗体或抗血管生成药物联合可延长生存期，但会增加毒性。目前尚无生物标志物能够识别从联合治疗中获益（相比单用奥希替尼）的患者。
方法：我们训练了一个多示例模型，从苏木精-伊红（H&E）组织切片预测一线EGFR-TKI的无进展生存期（PFS），并根据TKI类型进行校正。将伴有常见EGFR突变的aNSCLC原发肿瘤的诊断切片在三家独立机构以20×进行数字化，并回顾性纳入临床数据。采用自助法（bootstrapping），以连续PFS的C指数和1年PFS的AUC评估性能。
结果：共纳入141例患者（n=73奥希替尼；n=68厄洛替尼/吉非替尼）。我们在Gustave Roussy（GR，法国）上训练，并在Campus Bio-Medico（CBM，意大利）和乌迪内大学医院（U，意大利）上进行推断。在GR、CBM、U队列中，患者大多为女性（65%、62%、72%），无吸烟习惯（63%、59%、55%），为腺癌（92%、94%、100%），伴19号外显子缺失（56%、56%、55%）。GR、CBM、U的中位（IQR）年龄分别为65（56-74）、71（62-78）、68（63-73）岁。真实中位PFS和模型性能见表。
结论：这是首个从H&E切片预测EGFR-TKI的PFS的研究。尽管样本量有限，这些发现支持进一步探究数字病理学用于EGFR突变型NSCLC的PFS预测。将在大会上呈现更多队列的结果、模型与临床变量的结合以及可解释性分析。
模型在各队列中的性能 队列 亚组 患者数 真实PFS中位数（95%CI） 连续PFS的C指数 均值（95%CI） 1年PFS的AUC 均值（95%CI） 所有TKI 78 11.9（9.5, 15.4） NA-训练队列 NA-训练队列 GR 奥希替尼 42 12.6（9.8, 16.9） NA-训练队列 NA-训练队列 厄洛替尼/吉非替尼 36 9.9（7.0, 15.4） NA-训练队列 NA-训练队列 所有TKI 34 9.5（6.4, 16.3） 0.63（0.53, 0.73） 0.73（0.57, 0.86） CBM 奥希替尼 16 13.6（5.5, 23.3） 0.63（0.49, 0.77） 0.62（0.40, 0.86） 厄洛替尼/吉非替尼 18 7.4（2.1, 9.5） 0.59（0.44, 0.74） 0.77（0.57, 0.94） 所有TKI 29 13.0（9.4, 17.9） 0.78（0.70, 0.84） 0.83（0.69, 0.95） U 奥希替尼 15 23.0（6.7, 50.1） 0.84（0.70, 0.94） 0.90（0.72, 1.00） 厄洛替尼/吉非替尼 14 12.7（8.0, 17.5） 0.74（0.61, 0.85） 0.73（0.46, 0.93）
查看英文原文 English abstract
Background: Third-generation tyrosine kinase inhibitors (TKIs) like osimertinib are standard for untreated advanced non-small cell lung cancer (aNSCLC) with common EGFR mutations (exon 19 deletion/L858R). Combination with chemotherapy, bispecific antibodies or anti-angiogenic agents prolong survival but raises toxicity. No biomarker identifies patients benefiting from combinations versus osimertinib alone, yet.
Methods: We trained a multiple-instance model to predict progression-free survival (PFS) to first line EGFR-TKIs from hematoxylin-eosin (H&E) tissue slides, adjusting to the TKI type. Diagnostic slides from primary tumors of aNSCLC with common EGFR mutations were digitized at 20× at three independent institutions, and retrospectively included with clinical data. Performance was evaluated with C-index for continuous PFS and AUC for 1-year PFS, using bootstrapping.
Results: In total, 141 patients were included (n=73 osimertinib; n=68 erlotinib/gefitinib). We trained on Gustave Roussy (GR, France) and inferred on Campus Bio-Medico (CBM, Italy) and University Hospital of Udine (U, Italy). Patients were mostly female (65%, 62%, 72%), with no smoking habit (63%, 59%, 55%), had adenocarcinoma (92%, 94%, 100%) and exon 19 deletions (56%, 56%, 55%) across GR, CBM, U cohorts. Median (IQR) age was 65 (56-74), 71 (62-78), 68 (63-73) years in GR, CBM, U, respectively. Median true PFS and model performances are presented in the table.
Conclusion: This is the first study to predict PFS to EGFR-TKIs from H&E slides. Despite limited sample size, these findings support further investigation of digital pathology for PFS prediction in EGFR -mutated NSCLC. Results on additional cohorts, combination of the model with clinical variables, and interpretability will be presented at the congress.
Model's performance across cohorts Cohort Subset #Patients True PFS
median (95%CI) C-index continous PFS
Mean (95%CI) AUC 1-year PFS
Mean (95%CI) All TKIs 78 11.9 (9.5, 15.4) NA-Training cohort NA-Training cohort GR Osimertinib 42 12.6 (9.8, 16.9) NA-Training cohort NA-Training cohort Erlotinib/gefitinib 36 9.9 (7.0, 15.4) NA-Training cohort NA-Training cohort All TKIs 34 9.5 (6.4, 16.3) 0.63 (0.53, 0.73) 0.73 (0.57, 0.86) CBM Osimertinib 16 13.6 (5.5, 23.3) 0.63 (0.49, 0.77) 0.62 (0.40, 0.86) Erlotinib/gefitinib 18 7.4 (2.1, 9.5) 0.59 (0.44, 0.74) 0.77 (0.57, 0.94) All TKIs 29 13.0 (9.4, 17.9) 0.78 (0.70, 0.84) 0.83 (0.69, 0.95) U Osimertinib 15 23.0 (6.7, 50.1) 0.84 (0.70, 0.94) 0.90 (0.72, 1.00) Erlotinib/gefitinib 14 12.7 (8.0, 17.5) 0.74 (0.61, 0.85) 0.73 (0.46, 0.93)
利益披露 Disclosure
L. Zullo, None.
E. Samuelsson,
Owkin Employment.
F. Citarella, None.
M. Farag,
Owkin Employment.
F. Cortiula, None..
A. De Giglio, None..
F. Aboubakar, None.
K. von Loga,
owkin Employment.
E. Hatton,
Owkin Employment.
L. Nibid, None..
G. De Maglio, None..
F. Ambrosi, None..
B. Ramella Pollone, None..
M. Janson, None..
A. Russo, None..
A. Whittum, None..
I. Garberis, None.
V. Aubert,
Owkin Employment.
H. Deslandes,
Owkin Employment.
D. Planchard,
Abbvie Other, Consulting.
AstraZeneca Travel, Other, Consulting.
Boehringer Ingelheim Other, Consulting.
Bristol Myers Squibb Other, Consulting.
Celgene Other, Consulting.
Daiichi Sankyo Other, Consulting.
Eli Lilly Other, Consulting.
Janssen Other, Consulting.
Merck Other, Consulting.
Peer CME Other, Consulting.
Novartis Travel, Other, Consulting.
Pfizer Travel, Other, Consulting.
prIME Oncology Travel, Other, Consulting.
Roche Travel, Other, Consulting.
F. Andre,
Roche Other.
AstraZeneca Other.
Daiichi Sankyo Other.
Pfizer Other.
Novartis Other.
Eli Lilly Other.
C. Badoual, None.
B. Besse,
Abbvie Other, Consulting, Steering Committee.
Biontech SE Other, Consulting.
Beijing Avistone Biotechnology Other, Consulting.
Bristol Myers Squibb Other, consulting.
CureVac AG Other, consulting, steering committee.
PharmaMar Other, consulting, steering committee.
Regeneron Other, Consulting.
Sanofi Aventis Conseil Other, Consulting, Steering Committee.
Eli Lilly Other, Consulting.
Ellipses Pharma Ltd Other, Consulting.
F. Hoffmann-La Roche Ltd Other, Consulting.
Foghorn Therapeutics Inc Other, Consulting.
Takeda Other, Speaker.
Genmab Other, Consulting, steering committee.
Immunocore Other, consulting.
Owkin Other, Consulting.
Beigene Other, steering committee.
Janssen Other, Steering Committee.
Merck Sharp & Dohme Other, Steering Committee.
Ose Immunotherapeutics Other, Steering Committee.
M. Ghigna, None.
M. Aldea,
Amgen ).
Sandoz ).
Owkin ).
AstraZeneca ).