Clinically validated machine learning algorithm for detecting residual diseases with multicolor flow cytometry analysis in acute myeloid leukemia and myelodysplastic syndromeResearch in context
Bor-Sheng Ko,
Yu-Fen Wang,
Jeng-Lin Li,
Chi-Cheng Li,
Pei-Fang Weng,
Szu-Chun Hsu,
Hsin-An Hou,
Huai-Hsuan Huang,
Ming Yao,
Chien-Ting Lin,
Jia-Hau Liu,
Cheng-Hong Tsai,
Tai-Chung Huang,
Shang-Ju Wu,
Shang-Yi Huang,
Wen-Chien Chou,
Hwei-Fang Tien,
Chi-Chun Lee,
Jih-Luh Tang
Affiliations
Bor-Sheng Ko
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Yu-Fen Wang
Tai-Cheng Stem Cell Therapy Center, National Taiwan University, Taipei, Taiwan
Jeng-Lin Li
Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan
Chi-Cheng Li
Tai-Cheng Stem Cell Therapy Center, National Taiwan University, Taipei, Taiwan; Center of Stem Cell and Precision Medicine, Buddhist Tzu Chi General Hospital, Hualien, Taiwan
Pei-Fang Weng
Tai-Cheng Stem Cell Therapy Center, National Taiwan University, Taipei, Taiwan
Szu-Chun Hsu
Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan
Hsin-An Hou
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Huai-Hsuan Huang
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Ming Yao
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Chien-Ting Lin
Tai-Cheng Stem Cell Therapy Center, National Taiwan University, Taipei, Taiwan
Jia-Hau Liu
Tai-Cheng Stem Cell Therapy Center, National Taiwan University, Taipei, Taiwan
Cheng-Hong Tsai
Tai-Cheng Stem Cell Therapy Center, National Taiwan University, Taipei, Taiwan
Tai-Chung Huang
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Shang-Ju Wu
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Shang-Yi Huang
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Wen-Chien Chou
Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan
Hwei-Fang Tien
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Chi-Chun Lee
Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan; Joint Research Center for AI Technology and All Vista Healthcare, Ministry of Science and Technology, Taiwan; Correspondence to: C. C. Lee, No. 101, Section 2, Kuang Fu Road, Hsinchu 30013, Taiwan.
Jih-Luh Tang
Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan; Tai-Cheng Stem Cell Therapy Center, National Taiwan University, Taipei, Taiwan; Correspondence to: J. L. Tang, No. 7, Chung-Shan South Road, Taipei 10002, Taiwan.
Background: Multicolor flow cytometry (MFC) analysis is widely used to identify minimal residual disease (MRD) after treatment for acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). However, current manual interpretation suffers from drawbacks of time consuming and interpreter idiosyncrasy. Artificial intelligence (AI), with the expertise in assisting repetitive or complex analysis, represents a potential solution for these drawbacks. Methods: From 2009 to 2016, 5333 MFC data from 1742 AML or MDS patients were collected. The 287 MFC data at post-induction were selected as the outcome set for clinical outcome validation. The rest were 4:1 randomized into the training set (n = 4039) and the validation set (n = 1007). AI algorithm learned a multi-dimensional MFC phenotype from the training set and input it to support vector machine (SVM) classifier after Gaussian mixture model (GMM) modeling, and the performance was evaluated in The validation set. Findings: Promising accuracies (84·6% to 92·4%) and AUCs (0·921–0·950) were achieved by the developed algorithms. Interestingly, the algorithm from even one testing tube achieved similar performance. The clinical significance was validated in the outcome set, and normal MFC interpreted by the AI predicted better progression-free survival (10·9 vs 4·9 months, p < 0·0001) and overall survival (13·6 vs 6·5 months, p < 0·0001) for AML. Interpretation: Through large-scaled clinical validation, we showed that AI algorithms can produce efficient and clinically-relevant MFC analysis. This approach also possesses a great advantage of the ability to integrate other clinical tests. Fund: This work was supported by the Ministry of Science and Technology (107-2634-F-007-006 and 103–2314-B-002-185-MY2) of Taiwan. Keywords: Acute myeloid leukemia, Myelodysplastic syndrome, Multicolor flow cytometry, Minimal residual disease, Artificial intelligence
