Artificial intelligence methods to estimate overall mortality and non-relapse mortality following allogeneic HCT in the modern era: an EBMT-TCWP study.
B. Rius-Sansalvador
(1)
,
V. Moreno
(1)
,
C. Peczynski
(2)
,
E. Polge
(2)
,
J. E. Galimard
,
N. Kröger
(3)
,
D. Blaise
(4)
,
R. Peffault de Latour
(5, 6)
,
A. Kulagin
(7)
,
A. Mousavi
(8)
,
M. Stelljes
(9)
,
R. M. Hamladji
,
J. M. Middeke
(10)
,
U. Salmenniemi
,
H. Sengeloev
(11)
,
E. Forcade
(12)
,
U. Platzbecker
(13)
,
P. Reményi
(14)
,
E. Angelucci
,
P. Chevallier
(15)
,
Ibrahim Yakoub-Agha
(16)
,
C. Craddock
(17)
,
F. Ciceri
(18)
,
T. Schroeder
(19)
,
M. Aljurf
(20)
,
K. Ch
(21)
,
I. Moiseev
(7)
,
O. Penack
(22)
,
H. Schoemans
(23)
,
M. Mohty
(2)
,
B. Glass
(24)
,
A. Sureda
(1)
,
G. Basak
(25)
,
Z. Peric
(26)
1
University of Barcelona
2 CHU Saint-Antoine [AP-HP]
3 UKE - Universitaetsklinikum Hamburg-Eppendorf = University Medical Center Hamburg-Eppendorf [Hamburg]
4 IPC - Institut Paoli-Calmettes
5 IUH - Institut Universitaire d'Hématologie
6 AP-HP - Hopital Saint-Louis [AP-HP]
7 Pavlov First Saint Petersburg State Medical University [St. Petersburg]
8 Shariati Hospital [Tehran, Islamic Republic of Iran]
9 UKM - University Hospital Münster - Universitaetsklinikum Muenster [Germany]
10 University Hospital Carl Gustav Carus [Dresden, Germany]
11 Rigshospitalet [Copenhagen]
12 CHU Bordeaux
13 University Hospital Leipzig = Universitätsklinikum Leipzig
14 Dél-pesti Centrumkórház - Országos Hematológiai és Infektológiai Intézet [Budapest, Hungary]
15 CHU Nantes - Centre Hospitalier Universitaire de Nantes = Nantes University Hospital
16 LIRIC - Lille Inflammation Research International Center - U 995
17 Queens Elizabeth Hospital [Birmingham]
18 Ospedale San Raffaele
19 AöR - University Hospital Essen
20 KFSHRC - King Faisal Specialist Hospital and Resarch Centre [Riyadh, Saudi Arabia]
21 MHH - Medizinische Hochschule Hannover = Hannover Medical School
22 Charité - UniversitätsMedizin = Charité - University Hospital [Berlin]
23 University Hospitals Leuven [Leuven]
24 Helios Klinikum [Erfurt]
25 Medical University of Warsaw - Poland
26 University of Zagreb
2 CHU Saint-Antoine [AP-HP]
3 UKE - Universitaetsklinikum Hamburg-Eppendorf = University Medical Center Hamburg-Eppendorf [Hamburg]
4 IPC - Institut Paoli-Calmettes
5 IUH - Institut Universitaire d'Hématologie
6 AP-HP - Hopital Saint-Louis [AP-HP]
7 Pavlov First Saint Petersburg State Medical University [St. Petersburg]
8 Shariati Hospital [Tehran, Islamic Republic of Iran]
9 UKM - University Hospital Münster - Universitaetsklinikum Muenster [Germany]
10 University Hospital Carl Gustav Carus [Dresden, Germany]
11 Rigshospitalet [Copenhagen]
12 CHU Bordeaux
13 University Hospital Leipzig = Universitätsklinikum Leipzig
14 Dél-pesti Centrumkórház - Országos Hematológiai és Infektológiai Intézet [Budapest, Hungary]
15 CHU Nantes - Centre Hospitalier Universitaire de Nantes = Nantes University Hospital
16 LIRIC - Lille Inflammation Research International Center - U 995
17 Queens Elizabeth Hospital [Birmingham]
18 Ospedale San Raffaele
19 AöR - University Hospital Essen
20 KFSHRC - King Faisal Specialist Hospital and Resarch Centre [Riyadh, Saudi Arabia]
21 MHH - Medizinische Hochschule Hannover = Hannover Medical School
22 Charité - UniversitätsMedizin = Charité - University Hospital [Berlin]
23 University Hospitals Leuven [Leuven]
24 Helios Klinikum [Erfurt]
25 Medical University of Warsaw - Poland
26 University of Zagreb
J. E. Galimard
- Fonction : Auteur
R. M. Hamladji
- Fonction : Auteur
U. Salmenniemi
- Fonction : Auteur
E. Angelucci
- Fonction : Auteur
Résumé
Allogeneic haematopoietic cell transplantation (alloHCT) has curative potential counterbalanced by its toxicity. Prognostic scores fail to include current era patients and alternative donors. We examined adult patients from the EBMT registry who underwent alloHCT between 2010 and 2019 for oncohaematological disease. Our primary objective was to develop a new prognostic score for overall mortality (OM), with a secondary objective of predicting non-relapse mortality (NRM) using the OM score. AI techniques were employed. The model for OM was trained, optimized, and validated using 70%, 15%, and 15% of the data set, respectively. The top models, “gradient boosting” for OM (AUC = 0.64) and “elasticnet” for NRM (AUC = 0.62), were selected. The analysis included 33,927 patients. In the final prognostic model, patients with the lowest score had a 2-year OM and NRM of 18 and 13%, respectively, while those with the highest score had a 2-year OM and NRM of 82 and 93%, respectively. The results were consistent in the subset of the haploidentical cohort (n = 4386). Our score effectively stratifies the risk of OM and NRM in the current era but do not significantly improve mortality prediction. Future prognostic scores can benefit from identifying biological or dynamic markers post alloHCT.