An artificial intelligence model to predict hepatocellular carcinoma risk in Korean and Caucasian patients with chronic hepatitis B

Hwi Young Kim; Pietro Lampertico; Joon Yeul Nam; Hyung Chul Lee; Seung Up Kim; Dong Hyun Sinn; Yeon Seok Seo; Han Ah Lee; Soo Young Park; Young Suk Lim; Eun Sun Jang; Eileen L. Yoon; Hyoung Su Kim; Sung Eun Kim; Sang Bong Ahn; Jae Jun Shim; Soung Won Jeong; Yong Jin Jung; Joo Hyun Sohn; Yong Kyun Cho; Dae Won Jun; George N. Dalekos; Ramazan Idilman; Vana Sypsa; Thomas Berg; Maria Buti; Jose Luis Calleja; John Goulis; Spilios Manolakopoulos; Harry L.A. Janssen; Myoung jin Jang; Yun Bin Lee; Yoon Jun Kim; Jung Hwan Yoon; George V. Papatheodoridis; Jeong Hoon Lee

doi:10.1016/j.jhep.2021.09.025

An artificial intelligence model to predict hepatocellular carcinoma risk in Korean and Caucasian patients with chronic hepatitis B

Hwi Young Kim, Pietro Lampertico, Joon Yeul Nam, Hyung Chul Lee, Seung Up Kim, Dong Hyun Sinn, Yeon Seok Seo, Han Ah Lee, Soo Young Park, Young Suk Lim, Eun Sun Jang, Eileen L. Yoon, Hyoung Su Kim, Sung Eun Kim, Sang Bong Ahn, Jae Jun Shim, Soung Won Jeong, Yong Jin Jung, Joo Hyun Sohn, Yong Kyun ChoDae Won Jun, George N. Dalekos, Ramazan Idilman, Vana Sypsa, Thomas Berg, Maria Buti, Jose Luis Calleja, John Goulis, Spilios Manolakopoulos, Harry L.A. Janssen, Myoung jin Jang, Yun Bin Lee, Yoon Jun Kim, Jung Hwan Yoon, George V. Papatheodoridis, Jeong Hoon Lee

Department of Internal Medicine

Research output: Contribution to journal › Article › peer-review

83 Scopus citations

Abstract

Background & Aims: Several models have recently been developed to predict risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB). Our aims were to develop and validate an artificial intelligence-assisted prediction model of HCC risk. Methods: Using a gradient-boosting machine (GBM) algorithm, a model was developed using 6,051 patients with CHB who received entecavir or tenofovir therapy from 4 hospitals in Korea. Two external validation cohorts were independently established: Korean (5,817 patients from 14 Korean centers) and Caucasian (1,640 from 11 Western centers) PAGE-B cohorts. The primary outcome was HCC development. Results: In the derivation cohort and the 2 validation cohorts, cirrhosis was present in 26.9%–50.2% of patients at baseline. A model using 10 parameters at baseline was derived and showed good predictive performance (c-index 0.79). This model showed significantly better discrimination than previous models (PAGE-B, modified PAGE-B, REACH-B, and CU-HCC) in both the Korean (c-index 0.79 vs. 0.64–0.74; all p <0.001) and Caucasian validation cohorts (c-index 0.81 vs. 0.57–0.79; all p <0.05 except modified PAGE-B, p = 0.42). A calibration plot showed a satisfactory calibration function. When the patients were grouped into 4 risk groups, the minimal-risk group (11.2% of the Korean cohort and 8.8% of the Caucasian cohort) had a less than 0.5% risk of HCC during 8 years of follow-up. Conclusions: This GBM-based model provides the best predictive power for HCC risk in Korean and Caucasian patients with CHB treated with entecavir or tenofovir. Lay summary: Risk scores have been developed to predict the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B. We developed and validated a new risk prediction model using machine learning algorithms in 13,508 antiviral-treated patients with chronic hepatitis B. Our new model, based on 10 common baseline characteristics, demonstrated superior performance in risk stratification compared with previous risk scores. This model also identified a group of patients at minimal risk of developing HCC, who could be indicated for less intensive HCC surveillance.

Original language	English
Pages (from-to)	311-318
Number of pages	8
Journal	Journal of Hepatology
Volume	76
Issue number	2
DOIs	https://doi.org/10.1016/j.jhep.2021.09.025
State	Published - Feb 2022

Bibliographical note

Publisher Copyright:
© 2021 European Association for the Study of the Liver

Keywords

HBV
HCC
antiviral treatment
chronic hepatitis B
deep neural networking
liver cancer

Access to Document

10.1016/j.jhep.2021.09.025

Cite this

Kim, H. Y., Lampertico, P., Nam, J. Y., Lee, H. C., Kim, S. U., Sinn, D. H., Seo, Y. S., Lee, H. A., Park, S. Y., Lim, Y. S., Jang, E. S., Yoon, E. L., Kim, H. S., Kim, S. E., Ahn, S. B., Shim, J. J., Jeong, S. W., Jung, Y. J., Sohn, J. H., ... Lee, J. H. (2022). An artificial intelligence model to predict hepatocellular carcinoma risk in Korean and Caucasian patients with chronic hepatitis B. Journal of Hepatology, 76(2), 311-318. https://doi.org/10.1016/j.jhep.2021.09.025

@article{4e33adc03922417888f824b21ac7fdd2,

title = "An artificial intelligence model to predict hepatocellular carcinoma risk in Korean and Caucasian patients with chronic hepatitis B",

abstract = "Background & Aims: Several models have recently been developed to predict risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB). Our aims were to develop and validate an artificial intelligence-assisted prediction model of HCC risk. Methods: Using a gradient-boosting machine (GBM) algorithm, a model was developed using 6,051 patients with CHB who received entecavir or tenofovir therapy from 4 hospitals in Korea. Two external validation cohorts were independently established: Korean (5,817 patients from 14 Korean centers) and Caucasian (1,640 from 11 Western centers) PAGE-B cohorts. The primary outcome was HCC development. Results: In the derivation cohort and the 2 validation cohorts, cirrhosis was present in 26.9%–50.2% of patients at baseline. A model using 10 parameters at baseline was derived and showed good predictive performance (c-index 0.79). This model showed significantly better discrimination than previous models (PAGE-B, modified PAGE-B, REACH-B, and CU-HCC) in both the Korean (c-index 0.79 vs. 0.64–0.74; all p <0.001) and Caucasian validation cohorts (c-index 0.81 vs. 0.57–0.79; all p <0.05 except modified PAGE-B, p = 0.42). A calibration plot showed a satisfactory calibration function. When the patients were grouped into 4 risk groups, the minimal-risk group (11.2% of the Korean cohort and 8.8% of the Caucasian cohort) had a less than 0.5% risk of HCC during 8 years of follow-up. Conclusions: This GBM-based model provides the best predictive power for HCC risk in Korean and Caucasian patients with CHB treated with entecavir or tenofovir. Lay summary: Risk scores have been developed to predict the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B. We developed and validated a new risk prediction model using machine learning algorithms in 13,508 antiviral-treated patients with chronic hepatitis B. Our new model, based on 10 common baseline characteristics, demonstrated superior performance in risk stratification compared with previous risk scores. This model also identified a group of patients at minimal risk of developing HCC, who could be indicated for less intensive HCC surveillance.",

keywords = "HBV, HCC, antiviral treatment, chronic hepatitis B, deep neural networking, liver cancer",

author = "Kim, {Hwi Young} and Pietro Lampertico and Nam, {Joon Yeul} and Lee, {Hyung Chul} and Kim, {Seung Up} and Sinn, {Dong Hyun} and Seo, {Yeon Seok} and Lee, {Han Ah} and Park, {Soo Young} and Lim, {Young Suk} and Jang, {Eun Sun} and Yoon, {Eileen L.} and Kim, {Hyoung Su} and Kim, {Sung Eun} and Ahn, {Sang Bong} and Shim, {Jae Jun} and Jeong, {Soung Won} and Jung, {Yong Jin} and Sohn, {Joo Hyun} and Cho, {Yong Kyun} and Jun, {Dae Won} and Dalekos, {George N.} and Ramazan Idilman and Vana Sypsa and Thomas Berg and Maria Buti and Calleja, {Jose Luis} and John Goulis and Spilios Manolakopoulos and Janssen, {Harry L.A.} and Jang, {Myoung jin} and Lee, {Yun Bin} and Kim, {Yoon Jun} and Yoon, {Jung Hwan} and Papatheodoridis, {George V.} and Lee, {Jeong Hoon}",

note = "Publisher Copyright: {\textcopyright} 2021 European Association for the Study of the Liver",

year = "2022",

month = feb,

doi = "10.1016/j.jhep.2021.09.025",

language = "English",

volume = "76",

pages = "311--318",

journal = "Journal of Hepatology",

issn = "0168-8278",

publisher = "Elsevier B.V.",

number = "2",

}

Kim, HY, Lampertico, P, Nam, JY, Lee, HC, Kim, SU, Sinn, DH, Seo, YS, Lee, HA, Park, SY, Lim, YS, Jang, ES, Yoon, EL, Kim, HS, Kim, SE, Ahn, SB, Shim, JJ, Jeong, SW, Jung, YJ, Sohn, JH, Cho, YK, Jun, DW, Dalekos, GN, Idilman, R, Sypsa, V, Berg, T, Buti, M, Calleja, JL, Goulis, J, Manolakopoulos, S, Janssen, HLA, Jang, MJ, Lee, YB, Kim, YJ, Yoon, JH, Papatheodoridis, GV & Lee, JH 2022, 'An artificial intelligence model to predict hepatocellular carcinoma risk in Korean and Caucasian patients with chronic hepatitis B', Journal of Hepatology, vol. 76, no. 2, pp. 311-318. https://doi.org/10.1016/j.jhep.2021.09.025

TY - JOUR

T1 - An artificial intelligence model to predict hepatocellular carcinoma risk in Korean and Caucasian patients with chronic hepatitis B

AU - Kim, Hwi Young

AU - Lampertico, Pietro

AU - Nam, Joon Yeul

AU - Lee, Hyung Chul

AU - Kim, Seung Up

AU - Sinn, Dong Hyun

AU - Seo, Yeon Seok

AU - Lee, Han Ah

AU - Park, Soo Young

AU - Lim, Young Suk

AU - Jang, Eun Sun

AU - Yoon, Eileen L.

AU - Kim, Hyoung Su

AU - Kim, Sung Eun

AU - Ahn, Sang Bong

AU - Shim, Jae Jun

AU - Jeong, Soung Won

AU - Jung, Yong Jin

AU - Sohn, Joo Hyun

AU - Cho, Yong Kyun

AU - Jun, Dae Won

AU - Dalekos, George N.

AU - Idilman, Ramazan

AU - Sypsa, Vana

AU - Berg, Thomas

AU - Buti, Maria

AU - Calleja, Jose Luis

AU - Goulis, John

AU - Manolakopoulos, Spilios

AU - Janssen, Harry L.A.

AU - Jang, Myoung jin

AU - Lee, Yun Bin

AU - Kim, Yoon Jun

AU - Yoon, Jung Hwan

AU - Papatheodoridis, George V.

AU - Lee, Jeong Hoon

PY - 2022/2

Y1 - 2022/2

N2 - Background & Aims: Several models have recently been developed to predict risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB). Our aims were to develop and validate an artificial intelligence-assisted prediction model of HCC risk. Methods: Using a gradient-boosting machine (GBM) algorithm, a model was developed using 6,051 patients with CHB who received entecavir or tenofovir therapy from 4 hospitals in Korea. Two external validation cohorts were independently established: Korean (5,817 patients from 14 Korean centers) and Caucasian (1,640 from 11 Western centers) PAGE-B cohorts. The primary outcome was HCC development. Results: In the derivation cohort and the 2 validation cohorts, cirrhosis was present in 26.9%–50.2% of patients at baseline. A model using 10 parameters at baseline was derived and showed good predictive performance (c-index 0.79). This model showed significantly better discrimination than previous models (PAGE-B, modified PAGE-B, REACH-B, and CU-HCC) in both the Korean (c-index 0.79 vs. 0.64–0.74; all p <0.001) and Caucasian validation cohorts (c-index 0.81 vs. 0.57–0.79; all p <0.05 except modified PAGE-B, p = 0.42). A calibration plot showed a satisfactory calibration function. When the patients were grouped into 4 risk groups, the minimal-risk group (11.2% of the Korean cohort and 8.8% of the Caucasian cohort) had a less than 0.5% risk of HCC during 8 years of follow-up. Conclusions: This GBM-based model provides the best predictive power for HCC risk in Korean and Caucasian patients with CHB treated with entecavir or tenofovir. Lay summary: Risk scores have been developed to predict the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B. We developed and validated a new risk prediction model using machine learning algorithms in 13,508 antiviral-treated patients with chronic hepatitis B. Our new model, based on 10 common baseline characteristics, demonstrated superior performance in risk stratification compared with previous risk scores. This model also identified a group of patients at minimal risk of developing HCC, who could be indicated for less intensive HCC surveillance.

AB - Background & Aims: Several models have recently been developed to predict risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB). Our aims were to develop and validate an artificial intelligence-assisted prediction model of HCC risk. Methods: Using a gradient-boosting machine (GBM) algorithm, a model was developed using 6,051 patients with CHB who received entecavir or tenofovir therapy from 4 hospitals in Korea. Two external validation cohorts were independently established: Korean (5,817 patients from 14 Korean centers) and Caucasian (1,640 from 11 Western centers) PAGE-B cohorts. The primary outcome was HCC development. Results: In the derivation cohort and the 2 validation cohorts, cirrhosis was present in 26.9%–50.2% of patients at baseline. A model using 10 parameters at baseline was derived and showed good predictive performance (c-index 0.79). This model showed significantly better discrimination than previous models (PAGE-B, modified PAGE-B, REACH-B, and CU-HCC) in both the Korean (c-index 0.79 vs. 0.64–0.74; all p <0.001) and Caucasian validation cohorts (c-index 0.81 vs. 0.57–0.79; all p <0.05 except modified PAGE-B, p = 0.42). A calibration plot showed a satisfactory calibration function. When the patients were grouped into 4 risk groups, the minimal-risk group (11.2% of the Korean cohort and 8.8% of the Caucasian cohort) had a less than 0.5% risk of HCC during 8 years of follow-up. Conclusions: This GBM-based model provides the best predictive power for HCC risk in Korean and Caucasian patients with CHB treated with entecavir or tenofovir. Lay summary: Risk scores have been developed to predict the risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B. We developed and validated a new risk prediction model using machine learning algorithms in 13,508 antiviral-treated patients with chronic hepatitis B. Our new model, based on 10 common baseline characteristics, demonstrated superior performance in risk stratification compared with previous risk scores. This model also identified a group of patients at minimal risk of developing HCC, who could be indicated for less intensive HCC surveillance.

KW - HBV

KW - HCC

KW - antiviral treatment

KW - chronic hepatitis B

KW - deep neural networking

KW - liver cancer

UR - http://www.scopus.com/inward/record.url?scp=85120864891&partnerID=8YFLogxK

U2 - 10.1016/j.jhep.2021.09.025

DO - 10.1016/j.jhep.2021.09.025

M3 - Article

C2 - 34606915

AN - SCOPUS:85120864891

SN - 0168-8278

VL - 76

SP - 311

EP - 318

JO - Journal of Hepatology

JF - Journal of Hepatology

IS - 2

ER -

An artificial intelligence model to predict hepatocellular carcinoma risk in Korean and Caucasian patients with chronic hepatitis B

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this