Application of Natural Language Processing and Network Analysis Techniques to Post-market Reports for the Evaluation of Dose-related Anti-Thymocyte Globulin Safety Patterns

Summary

Objective: To evaluate the feasibility of automated dose and adverse event information retrieval in supporting the identification of safety patterns.

Methods: We extracted all rabbit Anti-Thymocyte Globulin (rATG) reports submitted to the United States Food and Drug Administration Adverse Event Reporting System (FAERS) from the product’s initial licensure in April 16, 1984 through February 8, 2016. We processed the narratives using the Medication Extraction (MedEx) and the Event-based Text-mining of Health Electronic Records (ETHER) systems and retrieved the appropriate medication, clinical, and temporal information. When necessary, the extracted information was manually curated. This process resulted in a high quality dataset that was analyzed with the Pattern-based and Advanced Network Analyzer for Clinical Evaluation and Assessment (PANACEA) to explore the association of rATG dosing with post-transplant lymphoproliferative disorder (PTLD).

Results: Although manual curation was necessary to improve the data quality, MedEx and ETHER supported the extraction of the appropriate information. We created a final dataset of 1,380 cases with complete information for rATG dosing and date of administration. Analysis in PANACEA found that PTLD was associated with cumulative doses of rATG >8 mg/kg, even in periods where most of the submissions to FAERS reported low doses of rATG.

Conclusion: We demonstrated the feasibility of investigating a dose-related safety pattern for a particular product in FAERS using a set of automated tools.

Citation: Botsis T, Foster M, Arya N, Kreimeyer K, Pandey A, Arya D. Application of natural language processing and network analysis techniques to post-market reports for the evaluation of dose-related anti-thymocyte globulin safety patterns. Appl Clin Inform 2017; 8: 396–411 https://doi.org/10.4338/ACI-2016-10-RA-0169

Keywords

Natural language processing - network analysis - postmarketing product surveillance - information retrieval - data visualization

Clinical Relevance Statement

• The automated retrieval of medication and other clinical information from the United States Food and Drug Administration Adverse Event Reporting System (FAERS) is critical for pharma-coepidemiological analysis.

• Natural language processing can be combined with other approaches, such as network analysis, to support the evaluation of safety patterns associated with medical product administration.

• The use of advanced techniques in the decision making process may assist medical experts and epidemiologists in performing their routine safety surveillance tasks.

Protection of Human and Animal Subjects

Human and/or animal subjects were not included in the project.

• References

• 1 Gaber AO, Monaco AP, Russell JA, Lebranchu Y, Mohty M. Rabbit antithymocyte globulin (thymoglobulin): 25 years and new frontiers in solid organ transplantation and haematology. Drugs 2010; 70 (06) 691-732.
• 2 Mohty M, Bacigalupo A, Saliba F, Zuckermann A, Morelon E, Lebranchu Y. New directions for rabbit anti-thymocyte globulin (Thymoglobulin®) in solid organ transplants, stem cell transplants and autoimmunity. Drugs 2014; 74 (14) 1605-1634.
• 3 3. THYMOGLOBULIN –Anti-thymocyte Globulin (rabbit) Injection, Powder, Lyophilized, for Solution. U.S. National Library of Medicine. National Institutes of Health. 2016 [cited December 12, 2016]. Available from https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=bbd8ab99–552e-4b81-aca4–6b0c7af8b9ae
• 4 Marks WH, Ilsley JN, Dharnidharka VR. Posttransplantation lymphoproliferative disorder in kidney and heart transplant recipients receiving thymoglobulin: a systematic review. Transplant Proc 2011; 43 (05) 1395-1404.
• 5 Gaber AO, Matas AJ, Henry ML, Brennan DC, Stevens RB, Kapur S, Ilsley JN, Kistler KD, Cosimi AB. Thymoglobulin Antibody Immunosuppression in Living Donor Recipients I. Antithymocyte globulin induction in living donor renal transplant recipients: final report of the TAILOR registry. Transplantation 2012; 94 (04) 331-337.
• 6 FDA. Questions and Answers on FDA‘s Adverse Event Reporting System (FAERS). 2016 Available from: http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Surveillance/AdverseDrugEffects/default.htm
• 7 Duggirala HJ. et al. Use of data mining at the Food and Drug Administration. J Am Med Inform Assoc 2016; 23 (02) 428-434.
• 8 Xu H, Stenner SP, Doan S, Johnson KB, Waitman LR, Denny JC. MedEx: a medication information extraction system for clinical narratives. J Am Med Inform Assoc 2010; 17 (01) 19-24.
• 9 Botsis T. et al. Decision Support Environment for Medical Product Safety Surveillance. J Biomed Inform 2016; 64: 354-362.
• 10 Wang W, Kreimeyer K, Woo EJ, Ball R, Foster M, Pandey A, Scott J, Botsis T. A new algorithmic approach for the extraction of temporal associations from clinical narratives with an application to medical product safety surveillance reports. J Biomed Inform 2016; 62: 78-89.
• 11 Hartigan JA, Wong MA. Algorithm AS 136: A k-means clustering algorithm. Journal of the Royal Statistical Society Series C (Applied Statistics) 1979; 28 (01) 100-108.
• 12 Team RC. R: A language and environment for statistical computing. R Foundation for Statistical Computing; Vienna: Austria URL https://wwwR-projectorg/ 2013
• 13 Csardi G, Nepusz T. The igraph software package for complex network research. InterJournal, Complex Systems 2006; 1695 (05) 1-9.
• 14 de Nooy W, Mrvar A, Batagelj V. Exploratory Social Network Analysis with Pajek. Second ed. Granovetter M. editor: Cambridge University Press; 2011
• 15 Newman M. Networks: an introduction. Oxford university press; 2010
• 16 Doan S, Bastarache L, Klimkowski S, Denny JC, Xu H. Integrating existing natural language processing tools for medication extraction from discharge summaries. J Am Med Inform Assoc 2010; 17 (05) 528-531.
• 17 Jiang M, Wu Y, Shah A, Priyanka P, Denny JC, Xu H. Extracting and standardizing medication information in clinical text –the MedEx-UIMA system. AMIA Jt Summits Transl Sci Proc. 2014 2014: 37-42
• 18 Botsis T, Buttolph T, Nguyen MD, Winiecki S, Woo EJ, Ball R. Vaccine adverse event text mining system for extracting features from vaccine safety reports. J Am Med Inform Assoc 2012; 19 (06) 1011-1018.
• 19 Ball R, Botsis T. Can network analysis improve pattern recognition among adverse events following immunization reported to VAERS?. Clin Pharmacol Ther 2011; 90 (02) 271-278.
• 20 Botsis T, Ball R. Network analysis of possible anaphylaxis cases reported to the US vaccine adverse event reporting system after H1N1 influenza vaccine. Stud Health Technol Inform 2011; 169: 564-568.
• 21 Botsis T, Scott J, Woo EJ, Ball R. Identifying Similar Cases in Document Networks Using Cross-Reference Structures. IEEE J Biomed Health Inform 2015; 19 (06) 1906-1917.