Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images
Abstract Both histologic subtypes and tumor mutation burden (TMB) represent important biomarkers in lung cancer, with implications for patient prognosis and treatment decisions. Typically, TMB is evaluated by comprehensive genomic profiling but this requires use of finite tissue specimens and costly...
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/bb85166704e240079299ad3d736a4fbb |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:bb85166704e240079299ad3d736a4fbb |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:bb85166704e240079299ad3d736a4fbb2021-12-02T16:46:35ZComparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images10.1038/s41598-021-95747-42045-2322https://doaj.org/article/bb85166704e240079299ad3d736a4fbb2021-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-95747-4https://doaj.org/toc/2045-2322Abstract Both histologic subtypes and tumor mutation burden (TMB) represent important biomarkers in lung cancer, with implications for patient prognosis and treatment decisions. Typically, TMB is evaluated by comprehensive genomic profiling but this requires use of finite tissue specimens and costly, time-consuming laboratory processes. Histologic subtype classification represents an established component of lung adenocarcinoma histopathology, but can be challenging and is associated with substantial inter-pathologist variability. Here we developed a deep learning system to both classify histologic patterns in lung adenocarcinoma and predict TMB status using de-identified Hematoxylin and Eosin (H&E) stained whole slide images. We first trained a convolutional neural network to map histologic features across whole slide images of lung cancer resection specimens. On evaluation using an external data source, this model achieved patch-level area under the receiver operating characteristic curve (AUC) of 0.78–0.98 across nine histologic features. We then integrated the output of this model with clinico-demographic data to develop an interpretable model for TMB classification. The resulting end-to-end system was evaluated on 172 held out cases from TCGA, achieving an AUC of 0.71 (95% CI 0.63–0.80). The benefit of using histologic features in predicting TMB is highlighted by the significant improvement this approach offers over using the clinical features alone (AUC of 0.63 [95% CI 0.53–0.72], p = 0.002). Furthermore, we found that our histologic subtype-based approach achieved performance similar to that of a weakly supervised approach (AUC of 0.72 [95% CI 0.64–0.80]). Together these results underscore that incorporating histologic patterns in biomarker prediction for lung cancer provides informative signals, and that interpretable approaches utilizing these patterns perform comparably with less interpretable, weakly supervised approaches.Apaar SadhwaniHuang-Wei ChangAli BehroozTrissia BrownIsabelle Auvigne-FlamentHardik PatelRobert FindlaterVanessa VelezFraser TanKamilla TekielaEllery WulczynEunhee S. YiCraig H. MermelDebra HanksPo-Hsuan Cameron ChenKimary KuligCory BatenchukDavid F. SteinerPeter CimermancicNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
Medicine R Science Q |
| spellingShingle |
Medicine R Science Q Apaar Sadhwani Huang-Wei Chang Ali Behrooz Trissia Brown Isabelle Auvigne-Flament Hardik Patel Robert Findlater Vanessa Velez Fraser Tan Kamilla Tekiela Ellery Wulczyn Eunhee S. Yi Craig H. Mermel Debra Hanks Po-Hsuan Cameron Chen Kimary Kulig Cory Batenchuk David F. Steiner Peter Cimermancic Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images |
| description |
Abstract Both histologic subtypes and tumor mutation burden (TMB) represent important biomarkers in lung cancer, with implications for patient prognosis and treatment decisions. Typically, TMB is evaluated by comprehensive genomic profiling but this requires use of finite tissue specimens and costly, time-consuming laboratory processes. Histologic subtype classification represents an established component of lung adenocarcinoma histopathology, but can be challenging and is associated with substantial inter-pathologist variability. Here we developed a deep learning system to both classify histologic patterns in lung adenocarcinoma and predict TMB status using de-identified Hematoxylin and Eosin (H&E) stained whole slide images. We first trained a convolutional neural network to map histologic features across whole slide images of lung cancer resection specimens. On evaluation using an external data source, this model achieved patch-level area under the receiver operating characteristic curve (AUC) of 0.78–0.98 across nine histologic features. We then integrated the output of this model with clinico-demographic data to develop an interpretable model for TMB classification. The resulting end-to-end system was evaluated on 172 held out cases from TCGA, achieving an AUC of 0.71 (95% CI 0.63–0.80). The benefit of using histologic features in predicting TMB is highlighted by the significant improvement this approach offers over using the clinical features alone (AUC of 0.63 [95% CI 0.53–0.72], p = 0.002). Furthermore, we found that our histologic subtype-based approach achieved performance similar to that of a weakly supervised approach (AUC of 0.72 [95% CI 0.64–0.80]). Together these results underscore that incorporating histologic patterns in biomarker prediction for lung cancer provides informative signals, and that interpretable approaches utilizing these patterns perform comparably with less interpretable, weakly supervised approaches. |
| format |
article |
| author |
Apaar Sadhwani Huang-Wei Chang Ali Behrooz Trissia Brown Isabelle Auvigne-Flament Hardik Patel Robert Findlater Vanessa Velez Fraser Tan Kamilla Tekiela Ellery Wulczyn Eunhee S. Yi Craig H. Mermel Debra Hanks Po-Hsuan Cameron Chen Kimary Kulig Cory Batenchuk David F. Steiner Peter Cimermancic |
| author_facet |
Apaar Sadhwani Huang-Wei Chang Ali Behrooz Trissia Brown Isabelle Auvigne-Flament Hardik Patel Robert Findlater Vanessa Velez Fraser Tan Kamilla Tekiela Ellery Wulczyn Eunhee S. Yi Craig H. Mermel Debra Hanks Po-Hsuan Cameron Chen Kimary Kulig Cory Batenchuk David F. Steiner Peter Cimermancic |
| author_sort |
Apaar Sadhwani |
| title |
Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images |
| title_short |
Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images |
| title_full |
Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images |
| title_fullStr |
Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images |
| title_full_unstemmed |
Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images |
| title_sort |
comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/bb85166704e240079299ad3d736a4fbb |
| work_keys_str_mv |
AT apaarsadhwani comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT huangweichang comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT alibehrooz comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT trissiabrown comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT isabelleauvigneflament comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT hardikpatel comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT robertfindlater comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT vanessavelez comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT frasertan comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT kamillatekiela comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT ellerywulczyn comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT eunheesyi comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT craighmermel comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT debrahanks comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT pohsuancameronchen comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT kimarykulig comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT corybatenchuk comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT davidfsteiner comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages AT petercimermancic comparativeanalysisofmachinelearningapproachestoclassifytumormutationburdeninlungadenocarcinomausinghistopathologyimages |
| _version_ |
1718383417642450944 |