PMID- 36057739
OWN - NLM
STAT- MEDLINE
DCOM- 20221205
LR  - 20230210
IS  - 1530-0285 (Electronic)
IS  - 0893-3952 (Print)
IS  - 0893-3952 (Linking)
VI  - 35
IP  - 12
DP  - 2022 Dec
TI  - Direct identification of ALK and ROS1 fusions in non-small cell lung cancer from 
      hematoxylin and eosin-stained slides using deep learning algorithms.
PG  - 1882-1887
LID - 10.1038/s41379-022-01141-4 [doi]
AB  - Anaplastic lymphoma kinase (ALK) and ROS oncogene 1 (ROS1) gene fusions are 
      well-established key players in non-small cell lung cancer (NSCLC). Although 
      their frequency is relatively low, their detection is important for patient care 
      and guides therapeutic decisions. The accepted methods used for their detection 
      are immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) 
      assay, as well as DNA and RNA-based sequencing methodologies. These assays are 
      expensive, time-consuming, and require technical expertise and specialized 
      equipment as well as biological specimens that are not always available. Here we 
      present an alternative detection method using a computer vision deep learning 
      approach. An advanced convolutional neural network (CNN) was used to generate 
      classifier models to detect ALK and ROS1-fusions directly from scanned 
      hematoxylin and eosin (H&E) whole slide images prepared from NSCLC tumors of 
      patients. A two-step training approach was applied, with an initial unsupervised 
      training step performed on a pan-cancer sample cohort followed by a 
      semi-supervised fine-tuning step, which supported the development of a classifier 
      with performances equal to those accepted for diagnostic tests. Validation of the 
      ALK/ROS1 classifier on a cohort of 72 lung cancer cases who underwent ALK and 
      ROS1-fusion testing at the pathology department at Sheba Medical Center displayed 
      sensitivities of 100% for both genes (six ALK-positive and two ROS1-positive 
      cases) and specificities of 100% and 98.6% respectively for ALK and ROS1, with 
      only one false-positive result for ROS1-alteration. These results demonstrate the 
      potential advantages that machine learning solutions may have in the molecular 
      pathology domain, by allowing fast, standardized, accurate, and robust biomarker 
      detection overcoming many limitations encountered when using current techniques. 
      The integration of such novel solutions into the routine pathology workflow can 
      support and improve the current clinical pipeline.
CI  - (c) 2022. The Author(s).
FAU - Mayer, Chen
AU  - Mayer C
AUID- ORCID: 0000-0002-7871-6338
AD  - Institute of Pathology, Sheba Medical Center, Ramat Gan, Israel. 
      chen.mayer@sheba.health.gov.il.
FAU - Ofek, Efrat
AU  - Ofek E
AD  - Institute of Pathology, Sheba Medical Center, Ramat Gan, Israel.
FAU - Fridrich, Danielle Even
AU  - Fridrich DE
AD  - Institute of Pathology, Sheba Medical Center, Ramat Gan, Israel.
FAU - Molchanov, Yossef
AU  - Molchanov Y
AD  - Institute of Pathology, Sheba Medical Center, Ramat Gan, Israel.
FAU - Yacobi, Rinat
AU  - Yacobi R
AD  - Institute of Pathology, Sheba Medical Center, Ramat Gan, Israel.
FAU - Gazy, Inbal
AU  - Gazy I
AD  - Imagene AI, Tel Aviv, Israel.
FAU - Hayun, Ido
AU  - Hayun I
AD  - Imagene AI, Tel Aviv, Israel.
FAU - Zalach, Jonathan
AU  - Zalach J
AD  - Imagene AI, Tel Aviv, Israel.
FAU - Paz-Yaacov, Nurit
AU  - Paz-Yaacov N
AD  - Imagene AI, Tel Aviv, Israel.
FAU - Barshack, Iris
AU  - Barshack I
AD  - Institute of Pathology, Sheba Medical Center, Ramat Gan, Israel.
AD  - Sackler faculty of Medicine, Tel Aviv University, Tel Aviv, Israel.
LA  - eng
PT  - Journal Article
PT  - Research Support, Non-U.S. Gov't
DEP - 20220903
PL  - United States
TA  - Mod Pathol
JT  - Modern pathology : an official journal of the United States and Canadian Academy 
      of Pathology, Inc
JID - 8806605
RN  - EC 2.7.10.1 (Anaplastic Lymphoma Kinase)
RN  - TDQ283MPCW (Eosine Yellowish-(YS))
RN  - YKM8PY2Z55 (Hematoxylin)
RN  - EC 2.7.10.1 (Protein-Tyrosine Kinases)
RN  - 0 (Proto-Oncogene Proteins)
RN  - EC 2.7.10.1 (ROS1 protein, human)
RN  - EC 2.7.10.1 (ALK protein, human)
RN  - 0 (Oncogene Proteins, Fusion)
SB  - IM
MH  - Humans
MH  - Anaplastic Lymphoma Kinase/genetics
MH  - *Carcinoma, Non-Small-Cell Lung/genetics/diagnosis
MH  - *Deep Learning
MH  - Eosine Yellowish-(YS)
MH  - Gene Rearrangement
MH  - Hematoxylin
MH  - In Situ Hybridization, Fluorescence
MH  - *Lung Neoplasms/genetics/diagnosis
MH  - Protein-Tyrosine Kinases/genetics
MH  - Proto-Oncogene Proteins/genetics
MH  - Oncogene Proteins, Fusion
PMC - PMC9708557
COIS- IG, IH, JZ and NPY are employees of Imagene AI and have stock options. CM was 
      reimbursed by Imagene AI for attending a conference. EO, DEF, YM, RY and IB 
      declare no competing interests.
EDAT- 2022/09/04 06:00
MHDA- 2022/12/02 06:00
PMCR- 2022/09/03
CRDT- 2022/09/03 23:29
PHST- 2022/03/21 00:00 [received]
PHST- 2022/07/20 00:00 [accepted]
PHST- 2022/07/20 00:00 [revised]
PHST- 2022/09/04 06:00 [pubmed]
PHST- 2022/12/02 06:00 [medline]
PHST- 2022/09/03 23:29 [entrez]
PHST- 2022/09/03 00:00 [pmc-release]
AID - S0893-3952(22)05500-4 [pii]
AID - 1141 [pii]
AID - 10.1038/s41379-022-01141-4 [doi]
PST - ppublish
SO  - Mod Pathol. 2022 Dec;35(12):1882-1887. doi: 10.1038/s41379-022-01141-4. Epub 2022 
      Sep 3.