PMID- 33446607
OWN - NLM
STAT- MEDLINE
DCOM- 20210906
LR  - 20210906
IS  - 1399-3003 (Electronic)
IS  - 0903-1936 (Linking)
VI  - 58
IP  - 2
DP  - 2021 Aug
TI  - Unsupervised phenotypic clustering for determining clinical status in children 
      with cystic fibrosis.
LID - 2002881 [pii]
LID - 10.1183/13993003.02881-2020 [doi]
AB  - BACKGROUND: Cystic fibrosis (CF) is a multisystem disease in which the assessment 
      of disease severity based on lung function alone may not be appropriate. The aim 
      of the study was to develop a comprehensive machine-learning algorithm to assess 
      clinical status independent of lung function in children. METHODS: A 
      comprehensive prospectively collected clinical database (Toronto, Canada) was 
      used to apply unsupervised cluster analysis. The defined clusters were then 
      compared by current and future lung function, risk of future hospitalisation, and 
      risk of future pulmonary exacerbation treated with oral antibiotics. A 
      k-nearest-neighbours (KNN) algorithm was used to prospectively assign clusters. 
      The methods were validated in a paediatric clinical CF dataset from Great Ormond 
      Street Hospital (GOSH). RESULTS: The optimal cluster model identified four (A-D) 
      phenotypic clusters based on 12 200 encounters from 530 individuals. Two clusters 
      (A and B) consistent with mild disease were identified with high forced 
      expiratory volume in 1 s (FEV(1)), and low risk of both hospitalisation and 
      pulmonary exacerbation treated with oral antibiotics. Two clusters (C and D) 
      consistent with severe disease were also identified with low FEV(1). Cluster D 
      had the shortest time to both hospitalisation and pulmonary exacerbation treated 
      with oral antibiotics. The outcomes were consistent in 3124 encounters from 171 
      children at GOSH. The KNN cluster allocation error rate was low, at 2.5% 
      (Toronto) and 3.5% (GOSH). CONCLUSION: Machine learning derived phenotypic 
      clusters can predict disease severity independent of lung function and could be 
      used in conjunction with functional measures to predict future disease 
      trajectories in CF patients.
CI  - Copyright (c)The authors 2021. For reproduction rights and permissions contact 
      permissions@ersnet.org.
FAU - Filipow, Nicole
AU  - Filipow N
AD  - UCL Great Ormond Street Institute of Child Health, London, UK.
AD  - Translational Medicine, SickKids Research Institute, Toronto, ON, Canada.
FAU - Davies, Gwyneth
AU  - Davies G
AUID- ORCID: 0000-0001-7937-2728
AD  - UCL Great Ormond Street Institute of Child Health, London, UK.
AD  - Great Ormond Street Hospital for Children and GOSH NIHR BRC, London, UK.
FAU - Main, Eleanor
AU  - Main E
AD  - UCL Great Ormond Street Institute of Child Health, London, UK.
FAU - Sebire, Neil J
AU  - Sebire NJ
AD  - UCL Great Ormond Street Institute of Child Health, London, UK.
AD  - Great Ormond Street Hospital for Children and GOSH NIHR BRC, London, UK.
FAU - Wallis, Colin
AU  - Wallis C
AD  - Great Ormond Street Hospital for Children and GOSH NIHR BRC, London, UK.
FAU - Ratjen, Felix
AU  - Ratjen F
AD  - Translational Medicine, SickKids Research Institute, Toronto, ON, Canada.
AD  - Division of Respiratory Medicine, Dept of Paediatrics, The Hospital for Sick 
      Children, Toronto, ON, Canada.
AD  - University of Toronto, Toronto, ON, Canada.
FAU - Stanojevic, Sanja
AU  - Stanojevic S
AD  - Translational Medicine, SickKids Research Institute, Toronto, ON, Canada.
AD  - Dept of Community Health and Epidemiology, Dalhousie University, Halifax, NS, 
      Canada.
LA  - eng
GR  - 204841/Z/16/Z/WT_/Wellcome Trust/United Kingdom
PT  - Journal Article
PT  - Research Support, Non-U.S. Gov't
DEP - 20210805
PL  - England
TA  - Eur Respir J
JT  - The European respiratory journal
JID - 8803460
SB  - IM
MH  - Child
MH  - Cluster Analysis
MH  - *Cystic Fibrosis/diagnosis
MH  - Forced Expiratory Volume
MH  - Humans
MH  - Lung
MH  - Respiratory Function Tests
COIS- Conflict of interest: N. Filipow has nothing to disclose. Conflict of interest: 
      G. Davies reports personal fees for lectures from Chiesi Limited, outside the 
      submitted work. Conflict of interest: E. Main has nothing to disclose. Conflict 
      of interest: N.J. Sebire has nothing to disclose. Conflict of interest: C. Wallis 
      has nothing to disclose. Conflict of interest: F. Ratjen reports grants and 
      personal fees for consultancy from Vertex, Calithera, Proteostasis, TranslateBio, 
      Genentech, Bayer and Boehringer Ingelheim, outside the submitted work. Conflict 
      of interest: S. Stanojevic reports grants from SickKids Foundation and European 
      Respiratory Society, during the conduct of the study.
EDAT- 2021/01/16 06:00
MHDA- 2021/09/07 06:00
CRDT- 2021/01/15 05:47
PHST- 2020/07/23 00:00 [received]
PHST- 2020/12/22 00:00 [accepted]
PHST- 2021/01/16 06:00 [pubmed]
PHST- 2021/09/07 06:00 [medline]
PHST- 2021/01/15 05:47 [entrez]
AID - 13993003.02881-2020 [pii]
AID - 10.1183/13993003.02881-2020 [doi]
PST - epublish
SO  - Eur Respir J. 2021 Aug 5;58(2):2002881. doi: 10.1183/13993003.02881-2020. Print 
      2021 Aug.