OpenLB
Open Library of Bioscience
Advanced Search
BMC medical research methodology
Dec 20, 2024;24 (1): 312.

Efficient analysis of drug interactions in liver injury: a retrospective study leveraging natural language processing and machine learning.

Junlong Ma, Heng Chen, Ji Sun, Juanjuan Huang, Gefei He, Guoping Yang
Author Information
  1. Junlong Ma: Center of Clinical Pharmacology, Third Xiangya Hospital, Central South University, Changsha, Hunan, China.
  2. Heng Chen: Department of Pharmacy, The First Hospital of Changsha, Central South University, Changsha, Hunan, China.
  3. Ji Sun: Department of Pharmacy, The First Hospital of Changsha, Central South University, Changsha, Hunan, China.
  4. Juanjuan Huang: Department of Pharmacy, The First Hospital of Changsha, Central South University, Changsha, Hunan, China.
  5. Gefei He: Department of Pharmacy, The First Hospital of Changsha, Central South University, Changsha, Hunan, China. 326366726@qq.com.
  6. Guoping Yang: Center of Clinical Pharmacology, Third Xiangya Hospital, Central South University, Changsha, Hunan, China. ygp9880@126.com. ORCID
PMID: 39707270 DOI: 10.1186/s12874-024-02443-8

Abstract

BACKGROUND: Liver injury from drug-drug interactions (DDIs), notably with anti-tuberculosis drugs such as isoniazid, poses a significant safety concern. Electronic medical records contain comprehensive clinical information and have gained increasing attention as a potential resource for DDI detection. However, a substantial portion of adverse drug reaction (ADR) information is hidden in unstructured narrative text, which has yet to be efficiently harnessed, thereby introducing bias into the research. There is a significant need for an efficient framework for the DDI assessment.
METHODS: Using a Chinese natural language processing (NLP) model, we extracted 25,130 adverse drug reaction (ADR) records, dividing them into sets for training an automated normalization model. The trained models, in conjunction with liver function laboratory tests, were used to thoroughly and efficiently identify liver injury cases. Ultimately, we applied a case-control study design to detect DDI signals increasing isoniazid's liver injury risk.
RESULTS: The Logistic Regression model demonstrated stable and superior performance in classification task. Based on laboratory criteria and NLP, we identified 128 liver injury cases among a cohort of 3,209 patients treated with isoniazid. Preliminary screening of 113 drug combinations with isoniazid highlighted 20 potential signal drugs, with antibacterials constituting 25%. Sensitivity analysis confirmed the robustness of signal drugs, especially in cardiac therapy and antibacterials.
CONCLUSION: Our NLP and machine learning approach effectively identifies isoniazid-related DDIs that increase the risk of liver injury, identifying 20 signal drugs, mainly antibacterials. Further research is required to validate these DDI signals.

Keywords

Drug interaction Liver injury Machine learning Natural language processing Retrospective study

References

  1. Int J Med Inform. 2018 Dec;120:92-100 [PMID: 30409350]
  2. Ther Innov Regul Sci. 2018 Jul;52(4):400-402 [PMID: 29714580]
  3. Liver Int. 2023 Jan;43(1):115-126 [PMID: 35899490]
  4. Antimicrob Agents Chemother. 2016 Aug 22;60(9):5285-93 [PMID: 27324775]
  5. Comput Struct Biotechnol J. 2022 Apr 19;20:2112-2123 [PMID: 35832629]
  6. J Am Med Inform Assoc. 2014 Mar-Apr;21(2):353-62 [PMID: 24158091]
  7. Biomed Pharmacother. 2023 Apr;160:114400 [PMID: 36805186]
  8. Drug Metab Dispos. 2017 Sep;45(9):1035-1043 [PMID: 28663285]
  9. Nat Protoc. 2019 Dec;14(12):3426-3444 [PMID: 31748751]
  10. Pharmacoepidemiol Drug Saf. 2016 Mar;25 Suppl 1:21-8 [PMID: 26147715]
  11. Hepatology. 2010 May;51(5):1665-74 [PMID: 20196116]
  12. Hepatology. 2010 Aug;52(2):798-9; author reply 799-800 [PMID: 20683973]
  13. Toxicol Appl Pharmacol. 2020 Sep 1;402:115134 [PMID: 32673658]
  14. CNS Drugs. 2019 Apr;33(4):383-395 [PMID: 30830574]
  15. Drug Saf. 2021 Nov;44(11):1125-1149 [PMID: 34533782]
  16. Int J Mol Sci. 2022 Nov 09;23(22): [PMID: 36430244]
  17. Korean J Chem Eng. 2023;40(2):276-285 [PMID: 36748027]
  18. J Chem Inf Model. 2024 Jan 8;64(1):96-109 [PMID: 38132638]
  19. BMC Gastroenterol. 2016 Oct 28;16(1):135 [PMID: 27793116]
  20. IEEE J Biomed Health Inform. 2024 Aug;28(8):4544-4552 [PMID: 38190664]
  21. Front Pharmacol. 2022 Oct 10;13:1037814 [PMID: 36299895]
  22. Antimicrob Agents Chemother. 2001 Feb;45(2):382-92 [PMID: 11158730]
  23. Clin Interv Aging. 2016 Mar 11;11:299-306 [PMID: 27042029]
  24. PLoS Comput Biol. 2021 Jul 6;17(7):e1009053 [PMID: 34228716]
  25. Br J Clin Pharmacol. 2016 Jun;81(6):1030-6 [PMID: 26773235]
  26. Clin Pharmacol Ther. 2007 Feb;81(2):298-304 [PMID: 17259955]
  27. Pharm Res. 2021 Sep;38(9):1485-1496 [PMID: 34518943]
  28. Clin Pharmacol Ther. 2019 Jun;105(6):1345-1361 [PMID: 30916389]
  29. J Appl Toxicol. 2022 Mar;42(3):516-528 [PMID: 34494278]
  30. Int J Mol Sci. 2023 May 29;24(11): [PMID: 37298396]
  31. Drug Discov Today. 2016 Apr;21(4):648-53 [PMID: 26948801]
  32. Clin Infect Dis. 2010 Mar 15;50(6):833-9 [PMID: 20156055]
  33. Comput Biol Med. 2021 Feb;129:104140 [PMID: 33278631]
  34. N Engl J Med. 2019 Jul 18;381(3):264-273 [PMID: 31314970]
  35. IEEE/ACM Trans Comput Biol Bioinform. 2023 Sep-Oct;20(5):3245-3256 [PMID: 37028367]
  36. Gastroenterology. 2019 Jun;156(8):2230-2241.e11 [PMID: 30742832]
  37. Allergy Asthma Immunol Res. 2020 May;12(3):430-442 [PMID: 32141257]
  38. Brief Bioinform. 2023 Jul 20;24(4): [PMID: 37344167]
  39. Am J Gastroenterol. 2007 Nov;102(11):2437-43 [PMID: 17662100]

Grants

  1. 2023JJ60513/Hunan Provincial Natural Science Foundation of China

MeSH Term

Humans
Natural Language Processing
Machine Learning
Chemical and Drug Induced Liver Injury
Drug Interactions
Retrospective Studies
Isoniazid
Antitubercular Agents
Female
Male
Middle Aged
Case-Control Studies
Electronic Health Records
Drug-Related Side Effects and Adverse Reactions
Adult
Logistic Models

Chemicals

Isoniazid
Antitubercular Agents
Journal Article
Article has an altmetric score of 1

Word Cloud

Created with Highcharts 10.0.0injuryliverdrugsDDIdrugisoniazidlanguageprocessingNLPmodelstudysignalantibacterialslearningLiverinteractionsDDIssignificantrecordsinformationincreasingpotentialadversereactionADRefficientlyresearchnaturallaboratorycasessignalsrisk20analysismachineBACKGROUND:drug-drugnotablyanti-tuberculosisposessafetyconcernElectronicmedicalcontaincomprehensiveclinicalgainedattentionresourcedetectionHoweversubstantialportionhiddenunstructurednarrativetextyetharnessedtherebyintroducingbiasneedefficientframeworkassessmentMETHODS:UsingChineseextracted25130dividingsetstrainingautomatednormalizationtrainedmodelsconjunctionfunctiontestsusedthoroughlyidentifyUltimatelyappliedcase-controldesigndetectisoniazid'sRESULTS:LogisticRegressiondemonstratedstablesuperiorperformanceclassificationtaskBasedcriteriaidentified128amongcohort3209patientstreatedPreliminaryscreening113combinationshighlightedconstituting25%SensitivityconfirmedrobustnessespeciallycardiactherapyCONCLUSION:approacheffectivelyidentifiesisoniazid-relatedincreaseidentifyingmainlyrequiredvalidateEfficientinjury:retrospectiveleveragingDruginteractionMachineNaturalRetrospective

Similar Articles

Cited By

No available data.