AI-augmented pharmacovigilance for cardio-oncology: FDR- controlled disproportionality and anomaly detection of antineoplastic-associated cardiac events in FAERS (2015 Q1–2025 Q4)

AI-augmented pharmacovigilance for cardio-oncology: FDR- controlled disproportionality and anomaly detection of antineoplastic-associated cardiac events in FAERS (2015 Q1–2025 Q4) | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article AI-augmented pharmacovigilance for cardio-oncology: FDR- controlled disproportionality and anomaly detection of antineoplastic-associated cardiac events in FAERS (2015 Q1–2025 Q4) hasan burak işleyen, zehra sucuoğlu, sevil tuğrul, sercan bulut, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8961206/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background Cardiac adverse events increasingly shape the real-world tolerability of systemic cancer therapies, and contemporary cardio-oncology practice underscores the need for structured surveillance—particularly for immune checkpoint inhibitor (ICI) myocarditis. Rare but high-impact toxicities are often detected post-marketing. We developed an oncology-focused pharmacovigilance framework that combines false discovery rate (FDR)-controlled disproportionality screening with AI-based detection of emerging signals in the U.S. FDA Adverse Event Reporting System (FAERS). Methods We analyzed FAERS quarterly data (2015 Q1–2025 Q4), de-duplicated reports by CASEID (retaining the most recent PRIMARYID), and restricted to primary-suspect exposure to 10 representative antineoplastic agents. Cardiac outcomes were pre-specified MedDRA preferred terms. Disproportionality was quantified using the reporting odds ratio (ROR) with Fisher’s exact test and sparse-cell correction (Haldane–Anscombe); p-values were adjusted across 90 comparisons using Benjamini–Hochberg FDR. For the top signal, monthly counts were evaluated with Isolation Forest to flag anomalous increases. We report in line with STROBE guidance for observational studies. Results After de-duplication, the final primary-suspect oncology cohort comprised 99,515 unique FAERS reports (377,615 drug–reaction records). Across 90 pre-specified drug–event pairs, 16 cardiac safety signals met FDR validation (q < 0.05). The strongest myocarditis signal was observed with nivolumab (ROR 7.76, 95% CI 6.39–9.42; a = 337), followed by pembrolizumab (ROR 3.85, 95% CI 2.73–5.43; a = 35). Established cardiotoxicities were recovered as internal controls, including doxorubicin-associated cardiac failure (ROR 2.84; a = 178) and trastuzumab-associated cardiac failure (ROR 2.57; a = 112). AI-based time-series anomaly detection highlighted discrete temporal surges for the top-ranked signal, supporting time-resolved prioritization of potential emerging toxicities. Conclusions In an oncology-focused FAERS cohort, FDR-controlled disproportionality analysis identified multiple cardiac safety signals, while AI-based anomaly detection provided complementary time-resolved prioritization. This reproducible framework can support cardio-oncology surveillance and guide hypothesis-driven follow-up studies. pharmacovigilance cardio-oncology FAERS reporting odds ratio false discovery rate immune checkpoint inhibitor myocarditis anomaly detection Figures Figure 1 Figure 2 Figure 3 Background Advances in systemic cancer therapy have improved survival but have also broadened the spectrum of cardiovascular toxicities encountered in routine practice. Immune checkpoint inhibitors (ICIs) can cause fulminant myocarditis, while classical cytotoxic agents and targeted therapies remain clinically relevant contributors to arrhythmias and myocardial dysfunction in heterogeneous real-world populations. [ 1 , 2 ] Post-marketing safety surveillance is therefore essential in cardio-oncology. The U.S. FDA Adverse Event Reporting System (FAERS) is a large spontaneous reporting database that supports hypothesis generation and signal detection for rare adverse events. [ 3 , 4 ] However, FAERS analyses must contend with reporting biases, sparse-event instability, and multiplicity when screening many drug–event combinations. [ 8 ] We designed an oncology-focused FAERS pipeline that prioritizes clinical interpretability and robustness: first, restriction to primary-suspect (PS) antineoplastic exposures to reduce polypharmacy confounding; second, disproportionality screening using reporting odds ratios (RORs) with false discovery rate (FDR) control; and third, AI-based anomaly detection to highlight time windows of abrupt reporting increases for high-priority signals. We report the performance of this framework across 10 representative antineoplastic agents and pre-specified cardiac outcomes. Methods Data source and study period: FAERS quarterly data files were analyzed for the period 2015 Q1 through 2025 Q4. [ 3 , 4 ] FAERS is a spontaneous reporting system that contains reports submitted by manufacturers, healthcare professionals, and consumers. [ 3 ] Because FAERS is de-identified and publicly available, this study did not involve patient contact (ethics review not applicable). The fully curated analysis dataset and reproducible code are publicly available at Zenodo (DOI: 10.5281/zenodo.18761110 ; version v1.0).[ 19 ] Case de-duplication and cohort construction: Reports were de-duplicated at the case level using CASEID, retaining the most recent version using PRIMARYID (FDA quarterly data file conventions). [ 4 ] We restricted the cohort to reports where one of 10 pre-specified antineoplastic agents was coded as the primary suspect (PS): nivolumab, pembrolizumab, trastuzumab, doxorubicin, paclitaxel, rituximab, cyclophosphamide, fluorouracil, cisplatin, and carboplatin. These agents were selected to represent major antineoplastic classes (ICIs, anti-HER2 therapy, taxanes, anthracyclines, and platinum/cytotoxic agents) and to provide internal positive controls for known cardiotoxicity patterns. [ 1 , 14 , 15 ] The analytic dataset was constructed as drug–reaction records by linking demographic, drug, and reaction information at the report level. Cardiac outcomes: Outcomes were defined a priori using MedDRA preferred terms (PTs): myocarditis, cardiac failure, heart failure, atrial fibrillation, myocardial infarction, pericarditis, cardiac arrest, tachycardia, and bradycardia. [ 5 ] In this PS-restricted analytic cohort, the PT “heart failure” did not occur (zero counts across drugs) and therefore did not contribute to validated signals; it was retained in the screening set for completeness and transparency (Table S1 ). These outcomes capture clinically salient myocarditis, arrhythmic, ischemic, and pump-failure phenotypes relevant to contemporary cardio-oncology practice. [ 1 , 2 ] Disproportionality analysis and AI-based anomaly detection: For each drug–event pair (10 drugs × 9 outcomes; 90 total), disproportionality was quantified using the reporting odds ratio (ROR) derived from a 2×2 contingency table, with Fisher’s exact test for significance. [ 6 – 8 ] To improve stability in sparse cells, we applied the Haldane–Anscombe correction (adding 0.5 to all cells when needed). [ 9 , 10 ] To address multiplicity, p-values were adjusted across all 90 comparisons using the Benjamini–Hochberg false discovery rate (FDR) procedure (q < 0.05). [ 11 ] For time-resolved prioritization, the Isolation Forest anomaly detector was applied to monthly report counts for the top-ranked signal (contamination 0.05; 100 estimators). [ 12 ] Sensitivity and subgroup analyses We performed three pre-specified robustness checks: (1) overlap exclusion at the case level (excluding reports that listed more than one of the target antineoplastic agents as PS), (2) a stricter minimum count threshold (a ≥ 5), and (3) sex-stratified screening (male vs female) restricted to reports with non-missing sex. Full screening results and subgroup tables are provided as supplementary files. AI-based emerging signal detection for an ICI–myocarditis signal (illustrative). Monthly report counts were derived from FDA_DT and analyzed using Isolation Forest. Months classified as anomalous indicate abrupt increases in reporting volume and are intended to prioritize time windows for review; this does not estimate incidence. To complement static disproportionality, we performed time-resolved surveillance for the top-ranked signal. Report dates (FDA_DT) from the demographic table were used to aggregate monthly counts, and an Isolation Forest anomaly detector was applied to identify months with abrupt increases in reporting. This analysis is intended for prioritization rather than causal inference. Software Analyses were performed in Python (pandas, SciPy, scikit-learn, matplotlib). Processed transparency tables are provided as Supplementary Tables S1–S3, and the analysis pipeline is available at https://github.com/drman44/FAERS-CardioOncology-2015-2025 . Reporting follows STROBE guidance for observational studies. [ 13 ] Results Following case-level de-duplication and PS-only cohort construction, the final analysis cohort comprised 99,515 unique FAERS reports covering 2015 Q1–2025 Q4 (Fig. 1 ). The drug–reaction analytic dataset included 377,615 records, of which 4,731 were cardiac records representing 4,459 cardiac cases. Descriptive characteristics (age/sex) were summarized where available; however, missingness in spontaneous reports limits inference and these fields were not used for adjustment. After Benjamini–Hochberg FDR correction, 16 drug–event pairs remained significant (Table 2 ; full screening results in Table S1 ). The strongest myocarditis signal was observed for nivolumab (ROR 7.76, 95% CI 6.39–9.42; a = 337) followed by pembrolizumab (ROR 3.85, 95% CI 2.73–5.43; a = 35). Other highly significant signals included rituximab–pericarditis (ROR 5.33; a = 397) and paclitaxel–tachycardia (ROR 5.50; a = 139). Established cardiotoxicities were recovered as internal controls, including doxorubicin–cardiac failure (ROR 2.84; a = 178) and trastuzumab–cardiac failure (ROR 2.57; a = 112). [ 14 , 15 ] Table 1 Characteristics of the deduplicated FAERS oncology cohort (unique reports). Drug Cases (n) Age, years median (IQR) Female, n (%) Male, n (%) Sex missing, n (%) Overall 99515 63 (53–71) 34428 (34.6) 36211 (36.4) 28876 (29.0) CARBOPLATIN 10519 63 (55–71) 4764 (45.3) 3594 (34.2) 2161 (20.5) CISPLATIN 3460 59 (49–67) 809 (23.4) 1265 (36.6) 1386 (40.1) CYCLOPHOSPHAMIDE 3667 57 (46–68) 1783 (48.6) 1213 (33.1) 671 (18.3) DOXORUBICIN 10397 57 (42–67) 3664 (35.2) 3205 (30.8) 3528 (33.9) FLUOROURACIL 3874 63 (53–70) 1085 (28.0) 1262 (32.6) 1527 (39.4) NIVOLUMAB 30100 65 (56–72) 9624 (32.0) 18023 (59.9) 2453 (8.1) PACLITAXEL 4865 61 (51–69) 2608 (53.6) 869 (17.9) 1388 (28.5) PEMBROLIZUMAB 1999 65 (57–73) 933 (46.7) 1000 (50.0) 66 (3.3) RITUXIMAB 25578 63 (51–72) 6505 (25.4) 5463 (21.4) 13610 (53.2) TRASTUZUMAB 5056 57 (48–66) 2653 (52.5) 317 (6.3) 2086 (41.3) Table 2 FDR-validated cardiac safety signals (q < 0.05) from disproportionality screening (10 drugs × 9 events). Drug Event a ROR 95% CI p FDR q NIVOLUMAB MYOCARDITIS 337 7.76 6.39–9.42 1.93e-105 1.74e-103 PACLITAXEL TACHYCARDIA 139 5.50 4.57–6.61 3.97e-51 8.93e-50 RITUXIMAB PERICARDITIS 397 5.33 4.34–6.54 1.65e-70 7.43e-69 PEMBROLIZUMAB MYOCARDITIS 35 3.85 2.73–5.43 1.17e-10 1.05e-09 DOXORUBICIN CARDIAC FAILURE 178 2.84 2.41–3.35 6.45e-29 9.67e-28 TRASTUZUMAB CARDIAC FAILURE 112 2.57 2.11–3.14 8.86e-17 8.86e-16 CARBOPLATIN TACHYCARDIA 149 2.52 2.10–3.02 4.91e-20 5.52e-19 PACLITAXEL CARDIAC ARREST 43 2.48 1.81–3.39 4.95e-07 2.12e-06 CYCLOPHOSPHAMIDE TACHYCARDIA 60 2.39 1.84–3.12 6.80e-09 3.83e-08 CYCLOPHOSPHAMIDE BRADYCARDIA 14 2.12 1.23–3.66 1.05e-02 2.56e-02 FLUOROURACIL CARDIAC ARREST 26 1.94 1.31–2.89 2.29e-03 6.86e-03 FLUOROURACIL MYOCARDIAL INFARCTION 29 1.89 1.30–2.76 1.87e-03 5.81e-03 CISPLATIN TACHYCARDIA 37 1.65 1.18–2.30 5.38e-03 1.42e-02 FLUOROURACIL TACHYCARDIA 35 1.58 1.13–2.22 1.30e-02 3.09e-02 NIVOLUMAB ATRIAL FIBRILLATION 295 1.52 1.32–1.74 7.46e-09 3.95e-08 NIVOLUMAB CARDIAC FAILURE 241 1.25 1.08–1.45 3.52e-03 9.91e-03 Abbreviations: a, case count for the drug–event pair in the primary-suspect cohort; ROR, reporting odds ratio; FDR, false discovery rate. Sensitivity analyses supported robustness. When applying a stricter minimum-count threshold (a ≥ 5), the key signals remained directionally consistent and the highest-priority associations persisted. Sex-stratified screening (male vs female) showed broadly consistent patterns for ICI-associated myocarditis, with subgroup results provided in Tables S2–S3. AI-based time-series analysis for the top signal is shown in Fig. 3 . Isolation Forest highlighted discrete months with abrupt increases in reporting, providing time-resolved prioritization for downstream review. Discussion In this oncology-focused FAERS analysis, we applied FDR-controlled disproportionality screening and identified 16 cardiac safety signals across 10 representative antineoplastic agents. The strongest association, nivolumab–myocarditis, is concordant with the malignant clinical phenotype of ICI-associated myocarditis described in contemporary registries and with FAERS-based pharmacovigilance observations in the ICI era. [ 2 , 16 ] The temporal anomalies identified by the Isolation Forest (Fig. 3 ) for nivolumab-associated myocarditis likely capture periods of stimulated reporting—an expected feature of spontaneous reporting systems when clinical attention shifts. [ 12 , 17 ] This supports the concept that AI-augmented surveillance can help separate steady-state reporting from dynamic surges that merit manual review, without implying changes in incidence. [ 17 , 18 ] Beyond ICIs, we observed signals consistent with known or biologically plausible cardio-oncology toxicities. Doxorubicin and trastuzumab were associated with cardiac failure signals, reflecting the persistent relevance of anthracycline and anti-HER2 cardiotoxicity in contemporary practice. [ 14 , 15 ] Tachyarrhythmia signals with paclitaxel and platinum-based therapy may reflect direct electrophysiologic effects, infusion-related stressors, or comorbidity burden in treated populations and merit targeted follow-up. [ 1 ] A key design choice was restricting exposure to primary-suspect (PS) coding. In routine oncology care, polypharmacy is the rule rather than the exception; analyses that include concomitant and secondary suspect drugs risk extensive confounding and signal dilution. By using a within-cohort active comparator, we aimed to keep the comparison clinically coherent while reducing confounding by indication. Nonetheless, disproportionality remains a hypothesis-generating method rather than a causal estimator. Multiple testing is an underappreciated threat in systematic pharmacovigilance screening. Using Benjamini–Hochberg FDR control, we prioritized a manageable set of signals while maintaining sensitivity, and we provide the full screening table to mitigate selective reporting concerns. [ 11 ] Time-resolved anomaly detection via Isolation Forest offers a flexible, distribution-free approach to flag unusual months for manual review, but should be interpreted in clinical context because stimulated reporting can generate apparent surges. [ 12 , 17 ] Limitations are inherent to spontaneous reporting databases, including under-reporting, missing data, reporting bias, and limited clinical detail (e.g., dose, baseline cardiac status). [ 17 , 18 ] Denominator data are unavailable; thus, incidence cannot be estimated, and disproportionality remains hypothesis-generating rather than causal. [ 8 , 17 ] In addition, sex was missing for a substantial proportion of reports, limiting sex-stratified inference. Despite these limitations, the present framework provides a reproducible and clinically oriented approach for prioritizing cardio-oncology signals to guide targeted follow-up studies. Conclusions In an oncology-focused FAERS cohort (2015 Q1–2025 Q4), FDR-controlled disproportionality analysis identified 16 cardiac safety signals across representative antineoplastic agents. AI-based anomaly detection provided complementary time-resolved prioritization for the top signal. This framework can support cardio-oncology surveillance and inform hypothesis-driven follow-up studies. Declarations The curated dataset and analysis scripts supporting the conclusions of this study are openly available in the Zenodo repository (DOI: 10.5281/zenodo.18761110 ; version v1.0). FAERS raw data are publicly accessible via the U.S. Food and Drug Administration website ( https://www.fda.gov ). Ethics approval and consent to participate Not applicable. This study used publicly available, de-identified FAERS data; ethics approval and informed consent were not required. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Funding The authors received no specific funding for this work. Author Contribution Author 1: Conceptualization, Methodology, Software, Formal analysis, Data curation, Visualization, Writing – original draft, Supervision. Author 2: Investigation, Validation, Writing – review & editing. Author 3: Investigation, Data curation, Validation, Writing – review & editing. Author 4: Methodology, Formal analysis, Visualization, Writing – review & editing. Author 5: Data curation, Validation, Writing – review & editing. Author 6: Investigation, Resources, Writing – review & editing. Author 7: Resources, Validation, Writing – review & editing. Author 8: Investigation, Data curation, Writing – review & editing. Author 9: Software, Data curation, Visualization. Author 10: Methodology, Investigation, Writing – review & editing. Author 11: Supervision, Project administration, Writing – review & editing. All authors read and approved the final manuscript. Acknowledgements Not applicable. Data Availability FAERS data are publicly available from the U.S. Food and Drug Administration website. [3,4] Processed transparency tables (full 90-pair screening and subgroup outputs) are provided with this submission as Supplementary Tables S1–S3. Code to reproduce the main tables and figures from the processed outputs is available at https://github.com/drman44/FAERS-CardioOncology-2015-2025. References Lyon AR, Dent S, Stanway S, et al. 2022 ESC Guidelines on cardio-oncology developed in collaboration with the European Hematology Association (EHA), the European Society for Therapeutic Radiology and Oncology (ESTRO) and the International Cardio-Oncology Society (IC-OS). Eur Heart J. 2022;43(41):4229–361. 10.1093/eurheartj/ehac244 . Mahmood SS, Fradley MG, Cohen JV, et al. Myocarditis in patients treated with immune checkpoint inhibitors. J Am Coll Cardiol. 2018;71(16):1755–64. 10.1016/j.jacc.2018.02.037 . Food US, Administration D. FDA Adverse Event Reporting System (FAERS): Latest Quarterly Data Files. Available at: https://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/fda-adverse-event-reporting-system-faers-latest-quarterly-data-files . Accessed 24 Feb 2026. U.S. Food and Drug Administration. FAERS Quarterly Data Files Documentation. Available at: https://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/faers-quarterly-data-files-documentation . Accessed 24 Feb 2026. MedDRA MSSO. MedDRA Introductory Guide (Version 27.0). March 2024. Available at: https://alt.meddra.org/files_acrobat/intguide_27_0_English.pdf . Accessed 24 Feb 2026. van Puijenbroek EP, Bate A, Leufkens HGM, Lindquist M, Orre R, Egberts ACG. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol Drug Saf. 2002;11(1):3–10. 10.1002/pds.668 . Faillie JL. Case-non-case studies: principle, methods, bias and interpretation. Therapie. 2019;74(2):225–32. 10.1016/j.therap.2019.01.006 . Cutroneo PM, et al. Conducting and interpreting disproportionality analyses in pharmacovigilance: practical considerations. Front Drug Saf Regul. 2024. 10.3389/fdsfr.2023.1323057 . Haldane JBS. The estimation and significance of the logarithm of a ratio of frequencies. Ann Hum Genet. 1956;20(4):309–11. 10.1111/j.1469-1809.1955.tb01285.x . Anscombe FJ. On estimating binomial response relations. Biometrika. 1956;43(3–4):461–4. 10.1093/biomet/43.3-4.461 . Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B. 1995;57(1):289–300. Liu FT, Ting KM, Zhou ZH. Isolation Forest. In: 2008 Eighth IEEE International Conference on Data Mining (ICDM). 2008:413–422. 10.1109/ICDM.2008.17 von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–9. 10.1016/j.jclinepi.2007.11.008 . Belger C, Abrahams C, Imamdin A, Lecour S. Doxorubicin-induced cardiotoxicity and risk factors. IJC Heart Vasculature. 2023;50:101332. 10.1016/j.ijcha.2023.101332 . Nowsheen S, Viscuse PV, O’Sullivan CC, et al. Incidence, diagnosis, and treatment of cardiac toxicity from trastuzumab in patients with breast cancer. Curr Treat Options Cardiovasc Med. 2017;19:46. 10.1007/s12609-017-0249-4 . Reeves DJ, Leffers J, Rao K. Immune checkpoint inhibitor related myocarditis reported through the FDA adverse event reporting system: pharmacovigilance trends in reporting and outcomes. Front Oncol. 2025. 10.3389/fonc.2025.1498817 . McIntyre RS, Karpuram S, et al. Utility and limitations of the FDA adverse events reporting system public dashboard for safety analyses: a case study with vesicular monoamine transporter 2 inhibitors. Expert Opin Drug Saf. 2025. 10.1080/14740338.2025.2588634 . Crisafulli S, Bate A, Brown JS, et al. Interplay of spontaneous reporting and longitudinal healthcare databases for signal management: position statement from the International Society of Pharmacovigilance Real-World Evidence and Big Data Special Interest Group. Drug Saf. 2025;48(9):959–76. 10.1007/s40264-025-01548-3 . İşleyen HB, Sucuoglu ZFAERS, Cardio-Oncology. (2015–2025): PS-only disproportionality + AI emerging signal detection. Zenodo. 2026. 10.5281/zenodo.18761110 Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformation.docx AnalysisPipeline.py GraphicalAbstractmain.png graphabstract.png Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 30 Mar, 2026 Reviewers agreed at journal 24 Mar, 2026 Reviewers invited by journal 17 Mar, 2026 Editor invited by journal 26 Feb, 2026 Editor assigned by journal 25 Feb, 2026 Submission checks completed at journal 25 Feb, 2026 First submitted to journal 24 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8961206","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":614868389,"identity":"fcefb95d-6d3f-465d-9ab7-e1fb8603e10b","order_by":0,"name":"hasan burak işleyen","email":"data:image/png;base64,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","orcid":"","institution":"Nişantaşı University","correspondingAuthor":true,"prefix":"","firstName":"hasan","middleName":"burak","lastName":"işleyen","suffix":""},{"id":614868390,"identity":"87d70fc2-0421-4144-88e3-031753f052c9","order_by":1,"name":"zehra sucuoğlu","email":"","orcid":"","institution":"Cemil Taşçıoğlu City Hospital","correspondingAuthor":false,"prefix":"","firstName":"zehra","middleName":"","lastName":"sucuoğlu","suffix":""},{"id":614868391,"identity":"9b8552f8-c054-444a-b8c7-dd67b3d31988","order_by":2,"name":"sevil tuğrul","email":"","orcid":"","institution":"Başakşehir Çam and Sakura City Hospital","correspondingAuthor":false,"prefix":"","firstName":"sevil","middleName":"","lastName":"tuğrul","suffix":""},{"id":614868392,"identity":"e76f64b2-688b-49cf-b6af-ed61e874cb4d","order_by":3,"name":"sercan bulut","email":"","orcid":"","institution":"Bağcılar Eğitim ve Araştırma Hastanesi","correspondingAuthor":false,"prefix":"","firstName":"sercan","middleName":"","lastName":"bulut","suffix":""},{"id":614868393,"identity":"ea250a3a-ef66-44da-a7e6-e76603ed9833","order_by":4,"name":"Oğuzhan Abanoz","email":"","orcid":"","institution":"Bezmialem Foundation University Medical Faculty Hospital","correspondingAuthor":false,"prefix":"","firstName":"Oğuzhan","middleName":"","lastName":"Abanoz","suffix":""},{"id":614868394,"identity":"0bcbd044-aa12-47a6-8d0a-59406a05d331","order_by":5,"name":"Esra Danışman","email":"","orcid":"","institution":"Bezmialem Foundation University Medical Faculty Hospital","correspondingAuthor":false,"prefix":"","firstName":"Esra","middleName":"","lastName":"Danışman","suffix":""},{"id":614868395,"identity":"2f3bf1cc-f148-41be-b51a-187b31ad0899","order_by":6,"name":"Kamran Kerimzade","email":"","orcid":"","institution":"Bezmialem Foundation University Medical Faculty Hospital","correspondingAuthor":false,"prefix":"","firstName":"Kamran","middleName":"","lastName":"Kerimzade","suffix":""},{"id":614868396,"identity":"0b683409-a2dd-441f-a398-1a6ad904f79b","order_by":7,"name":"Lina Boukhemis","email":"","orcid":"","institution":"Bezmialem Foundation University Medical Faculty Hospital","correspondingAuthor":false,"prefix":"","firstName":"Lina","middleName":"","lastName":"Boukhemis","suffix":""},{"id":614868397,"identity":"a1bae993-3ddc-49af-b10f-d236c5556f3f","order_by":8,"name":"Fatih Kızkapan","email":"","orcid":"","institution":"Bağcılar Eğitim ve Araştırma Hastanesi","correspondingAuthor":false,"prefix":"","firstName":"Fatih","middleName":"","lastName":"Kızkapan","suffix":""},{"id":614868398,"identity":"e21fe0f2-ad85-4db7-988f-f42a66097098","order_by":9,"name":"Cevahir Alioğlu","email":"","orcid":"","institution":"Bezmialem Foundation University Medical Faculty Hospital","correspondingAuthor":false,"prefix":"","firstName":"Cevahir","middleName":"","lastName":"Alioğlu","suffix":""},{"id":614868399,"identity":"b62666c9-3635-4b99-a072-52ab7404735b","order_by":10,"name":"Şaban seçmeler","email":"","orcid":"","institution":"Medical Park Bahçelievler Hospital","correspondingAuthor":false,"prefix":"","firstName":"Şaban","middleName":"","lastName":"seçmeler","suffix":""}],"badges":[],"createdAt":"2026-02-24 21:39:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8961206/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8961206/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106116990,"identity":"a571d8f9-4ffd-4cf0-b666-fee0551ae57a","added_by":"auto","created_at":"2026-04-03 16:40:16","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":220761,"visible":true,"origin":"","legend":"\u003cp\u003eCohort construction and analytic dataset. FAERS quarterly data (2015 Q1–2025 Q4) were de-duplicated by CASEID and restricted to primary-suspect exposure to 10 pre-specified antineoplastic agents. The final analytic dataset comprised drug–reaction records and a pre-specified set of cardiac outcomes.\u003c/p\u003e","description":"","filename":"Figure1FlowDiagram.png","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/eae197d1f0a830cb7f283f89.png"},{"id":106116989,"identity":"3bbb40dc-a1a5-40cf-bc1a-0789fdb7b0d4","added_by":"auto","created_at":"2026-04-03 16:40:15","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":203987,"visible":true,"origin":"","legend":"\u003cp\u003eVolcano plot of disproportionality screening across 90 drug–event pairs. The x-axis shows log2(ROR) and the y-axis shows −log10(p-value). Points meeting the pre-defined FDR-validated signal criteria are highlighted.\u003c/p\u003e","description":"","filename":"Figure2VolcanoPlot.png","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/ebfc936e56b2bcd500ab917e.png"},{"id":106116984,"identity":"7e2ce075-4ddc-4772-97d5-2244cf2f4149","added_by":"auto","created_at":"2026-04-03 16:40:14","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":870759,"visible":true,"origin":"","legend":"\u003cp\u003eAI-based emerging signal detection for the top-ranked drug–event pair. Monthly report counts were derived from FDA_DT and analyzed using Isolation Forest. Months classified as anomalous indicate abrupt increases in reporting and are intended to prioritize time windows for review.\u003c/p\u003e","description":"","filename":"Figure3AnomalyDetection.png","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/e46403e9082b2b5d2d7df531.png"},{"id":106117071,"identity":"cf3ed395-f381-4646-8bd3-899a6a582ba9","added_by":"auto","created_at":"2026-04-03 16:40:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1861357,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/2633dbbc-1ffa-40bf-b57b-71ae96303faa.pdf"},{"id":106116968,"identity":"a73f50e8-fd85-4b6c-9313-fffaf418d80f","added_by":"auto","created_at":"2026-04-03 16:40:09","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":37239,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/31ef4e3f76046b9179652f05.docx"},{"id":106116985,"identity":"b1c06203-e832-441f-9da5-9352e1d9c32c","added_by":"auto","created_at":"2026-04-03 16:40:14","extension":"py","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":20611,"visible":true,"origin":"","legend":"","description":"","filename":"AnalysisPipeline.py","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/0c5a0f2727f12513d9830446.py"},{"id":106116983,"identity":"e8aad662-663b-4331-869d-f3d2c4fa8811","added_by":"auto","created_at":"2026-04-03 16:40:14","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":312864,"visible":true,"origin":"","legend":"","description":"","filename":"GraphicalAbstractmain.png","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/695c45c478f2fa121edb07e7.png"},{"id":106116979,"identity":"45011a1d-c93f-48f8-9748-b2b8e65e5edc","added_by":"auto","created_at":"2026-04-03 16:40:12","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":2370278,"visible":true,"origin":"","legend":"","description":"","filename":"graphabstract.png","url":"https://assets-eu.researchsquare.com/files/rs-8961206/v1/4bc606b3ad8a883e71ffde77.png"}],"financialInterests":"No competing interests reported.","formattedTitle":"AI-augmented pharmacovigilance for cardio-oncology: FDR- controlled disproportionality and anomaly detection of antineoplastic-associated cardiac events in FAERS (2015 Q1–2025 Q4)","fulltext":[{"header":"Background","content":"\u003cp\u003eAdvances in systemic cancer therapy have improved survival but have also broadened the spectrum of cardiovascular toxicities encountered in routine practice. Immune checkpoint inhibitors (ICIs) can cause fulminant myocarditis, while classical cytotoxic agents and targeted therapies remain clinically relevant contributors to arrhythmias and myocardial dysfunction in heterogeneous real-world populations. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/p\u003e \u003cp\u003ePost-marketing safety surveillance is therefore essential in cardio-oncology. The U.S. FDA Adverse Event Reporting System (FAERS) is a large spontaneous reporting database that supports hypothesis generation and signal detection for rare adverse events. [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] However, FAERS analyses must contend with reporting biases, sparse-event instability, and multiplicity when screening many drug\u0026ndash;event combinations. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eWe designed an oncology-focused FAERS pipeline that prioritizes clinical interpretability and robustness: first, restriction to primary-suspect (PS) antineoplastic exposures to reduce polypharmacy confounding; second, disproportionality screening using reporting odds ratios (RORs) with false discovery rate (FDR) control; and third, AI-based anomaly detection to highlight time windows of abrupt reporting increases for high-priority signals. We report the performance of this framework across 10 representative antineoplastic agents and pre-specified cardiac outcomes.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eData source and study period: FAERS quarterly data files were analyzed for the period 2015 Q1 through 2025 Q4. [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] FAERS is a spontaneous reporting system that contains reports submitted by manufacturers, healthcare professionals, and consumers. [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] Because FAERS is de-identified and publicly available, this study did not involve patient contact (ethics review not applicable). The fully curated analysis dataset and reproducible code are publicly available at Zenodo (DOI: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5281/zenodo.18761110\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.18761110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e; version v1.0).[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eCase de-duplication and cohort construction: Reports were de-duplicated at the case level using CASEID, retaining the most recent version using PRIMARYID (FDA quarterly data file conventions). [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] We restricted the cohort to reports where one of 10 pre-specified antineoplastic agents was coded as the primary suspect (PS): nivolumab, pembrolizumab, trastuzumab, doxorubicin, paclitaxel, rituximab, cyclophosphamide, fluorouracil, cisplatin, and carboplatin. These agents were selected to represent major antineoplastic classes (ICIs, anti-HER2 therapy, taxanes, anthracyclines, and platinum/cytotoxic agents) and to provide internal positive controls for known cardiotoxicity patterns. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] The analytic dataset was constructed as drug\u0026ndash;reaction records by linking demographic, drug, and reaction information at the report level.\u003c/p\u003e \u003cp\u003eCardiac outcomes: Outcomes were defined a priori using MedDRA preferred terms (PTs): myocarditis, cardiac failure, heart failure, atrial fibrillation, myocardial infarction, pericarditis, cardiac arrest, tachycardia, and bradycardia. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] In this PS-restricted analytic cohort, the PT \u0026ldquo;heart failure\u0026rdquo; did not occur (zero counts across drugs) and therefore did not contribute to validated signals; it was retained in the screening set for completeness and transparency (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). These outcomes capture clinically salient myocarditis, arrhythmic, ischemic, and pump-failure phenotypes relevant to contemporary cardio-oncology practice. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eDisproportionality analysis and AI-based anomaly detection: For each drug\u0026ndash;event pair (10 drugs \u0026times; 9 outcomes; 90 total), disproportionality was quantified using the reporting odds ratio (ROR) derived from a 2\u0026times;2 contingency table, with Fisher\u0026rsquo;s exact test for significance. [\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e] To improve stability in sparse cells, we applied the Haldane\u0026ndash;Anscombe correction (adding 0.5 to all cells when needed). [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] To address multiplicity, p-values were adjusted across all 90 comparisons using the Benjamini\u0026ndash;Hochberg false discovery rate (FDR) procedure (q\u0026thinsp;\u0026lt;\u0026thinsp;0.05). [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] For time-resolved prioritization, the Isolation Forest anomaly detector was applied to monthly report counts for the top-ranked signal (contamination 0.05; 100 estimators). [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSensitivity and subgroup analyses\u003c/h2\u003e \u003cp\u003eWe performed three pre-specified robustness checks: (1) overlap exclusion at the case level (excluding reports that listed more than one of the target antineoplastic agents as PS), (2) a stricter minimum count threshold (a\u0026thinsp;\u0026ge;\u0026thinsp;5), and (3) sex-stratified screening (male vs female) restricted to reports with non-missing sex. Full screening results and subgroup tables are provided as supplementary files.\u003c/p\u003e \u003cp\u003eAI-based emerging signal detection for an ICI\u0026ndash;myocarditis signal (illustrative). Monthly report counts were derived from FDA_DT and analyzed using Isolation Forest. Months classified as anomalous indicate abrupt increases in reporting volume and are intended to prioritize time windows for review; this does not estimate incidence.\u003c/p\u003e \u003cp\u003eTo complement static disproportionality, we performed time-resolved surveillance for the top-ranked signal. Report dates (FDA_DT) from the demographic table were used to aggregate monthly counts, and an Isolation Forest anomaly detector was applied to identify months with abrupt increases in reporting. This analysis is intended for prioritization rather than causal inference.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSoftware\u003c/h3\u003e\n\u003cp\u003eAnalyses were performed in Python (pandas, SciPy, scikit-learn, matplotlib). Processed transparency tables are provided as Supplementary Tables S1\u0026ndash;S3, and the analysis pipeline is available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/drman44/FAERS-CardioOncology-2015-2025\u003c/span\u003e\u003cspan address=\"https://github.com/drman44/FAERS-CardioOncology-2015-2025\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Reporting follows STROBE guidance for observational studies. [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eFollowing case-level de-duplication and PS-only cohort construction, the final analysis cohort comprised 99,515 unique FAERS reports covering 2015 Q1\u0026ndash;2025 Q4 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The drug\u0026ndash;reaction analytic dataset included 377,615 records, of which 4,731 were cardiac records representing 4,459 cardiac cases.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eDescriptive characteristics (age/sex) were summarized where available; however, missingness in spontaneous reports limits inference and these fields were not used for adjustment.\u003c/p\u003e \u003cp\u003eAfter Benjamini\u0026ndash;Hochberg FDR correction, 16 drug\u0026ndash;event pairs remained significant (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; full screening results in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The strongest myocarditis signal was observed for nivolumab (ROR 7.76, 95% CI 6.39\u0026ndash;9.42; a\u0026thinsp;=\u0026thinsp;337) followed by pembrolizumab (ROR 3.85, 95% CI 2.73\u0026ndash;5.43; a\u0026thinsp;=\u0026thinsp;35). Other highly significant signals included rituximab\u0026ndash;pericarditis (ROR 5.33; a\u0026thinsp;=\u0026thinsp;397) and paclitaxel\u0026ndash;tachycardia (ROR 5.50; a\u0026thinsp;=\u0026thinsp;139). Established cardiotoxicities were recovered as internal controls, including doxorubicin\u0026ndash;cardiac failure (ROR 2.84; a\u0026thinsp;=\u0026thinsp;178) and trastuzumab\u0026ndash;cardiac failure (ROR 2.57; a\u0026thinsp;=\u0026thinsp;112). [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCharacteristics of the deduplicated FAERS oncology cohort (unique reports).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDrug\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCases (n)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAge, years median (IQR)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFemale, n (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMale, n (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSex missing, n (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOverall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e99515\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63 (53\u0026ndash;71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e34428 (34.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e36211 (36.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e28876 (29.0)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCARBOPLATIN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10519\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63 (55\u0026ndash;71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4764 (45.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3594 (34.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2161 (20.5)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCISPLATIN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3460\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e59 (49\u0026ndash;67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e809 (23.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1265 (36.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1386 (40.1)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCYCLOPHOSPHAMIDE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3667\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57 (46\u0026ndash;68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1783 (48.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1213 (33.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e671 (18.3)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDOXORUBICIN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10397\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57 (42\u0026ndash;67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3664 (35.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e3205 (30.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3528 (33.9)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFLUOROURACIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3874\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63 (53\u0026ndash;70)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1085 (28.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1262 (32.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1527 (39.4)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNIVOLUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e30100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65 (56\u0026ndash;72)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9624 (32.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e18023 (59.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2453 (8.1)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePACLITAXEL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4865\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e61 (51\u0026ndash;69)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2608 (53.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e869 (17.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1388 (28.5)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePEMBROLIZUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1999\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65 (57\u0026ndash;73)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e933 (46.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1000 (50.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e66 (3.3)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRITUXIMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e25578\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63 (51\u0026ndash;72)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6505 (25.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5463 (21.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e13610 (53.2)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTRASTUZUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57 (48\u0026ndash;66)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2653 (52.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e317 (6.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2086 (41.3)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFDR-validated cardiac safety signals (q\u0026thinsp;\u0026lt;\u0026thinsp;0.05) from disproportionality screening (10 drugs \u0026times; 9 events).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDrug\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEvent\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ea\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eROR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFDR q\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNIVOLUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMYOCARDITIS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e337\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6.39\u0026ndash;9.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.93e-105\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.74e-103\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePACLITAXEL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTACHYCARDIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e139\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.57\u0026ndash;6.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.97e-51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8.93e-50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRITUXIMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePERICARDITIS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e397\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.34\u0026ndash;6.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.65e-70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e7.43e-69\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePEMBROLIZUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMYOCARDITIS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.73\u0026ndash;5.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.17e-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.05e-09\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDOXORUBICIN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCARDIAC FAILURE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e178\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.41\u0026ndash;3.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.45e-29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9.67e-28\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTRASTUZUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCARDIAC FAILURE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e112\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.11\u0026ndash;3.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8.86e-17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8.86e-16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCARBOPLATIN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTACHYCARDIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e149\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2.10\u0026ndash;3.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4.91e-20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5.52e-19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePACLITAXEL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCARDIAC ARREST\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.81\u0026ndash;3.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4.95e-07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.12e-06\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCYCLOPHOSPHAMIDE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTACHYCARDIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.84\u0026ndash;3.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.80e-09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.83e-08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCYCLOPHOSPHAMIDE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBRADYCARDIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.23\u0026ndash;3.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.05e-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.56e-02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFLUOROURACIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCARDIAC ARREST\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.31\u0026ndash;2.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.29e-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e6.86e-03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFLUOROURACIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMYOCARDIAL INFARCTION\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.30\u0026ndash;2.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.87e-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5.81e-03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCISPLATIN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTACHYCARDIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.18\u0026ndash;2.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.38e-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.42e-02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFLUOROURACIL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTACHYCARDIA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.13\u0026ndash;2.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.30e-02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.09e-02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNIVOLUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eATRIAL FIBRILLATION\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e295\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.32\u0026ndash;1.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.46e-09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.95e-08\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNIVOLUMAB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCARDIAC FAILURE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e241\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.08\u0026ndash;1.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.52e-03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9.91e-03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eAbbreviations: a, case count for the drug\u0026ndash;event pair in the primary-suspect cohort; ROR, reporting odds ratio; FDR, false discovery rate.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eSensitivity analyses supported robustness. When applying a stricter minimum-count threshold (a\u0026thinsp;\u0026ge;\u0026thinsp;5), the key signals remained directionally consistent and the highest-priority associations persisted. Sex-stratified screening (male vs female) showed broadly consistent patterns for ICI-associated myocarditis, with subgroup results provided in Tables S2\u0026ndash;S3.\u003c/p\u003e \u003cp\u003eAI-based time-series analysis for the top signal is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Isolation Forest highlighted discrete months with abrupt increases in reporting, providing time-resolved prioritization for downstream review.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this oncology-focused FAERS analysis, we applied FDR-controlled disproportionality screening and identified 16 cardiac safety signals across 10 representative antineoplastic agents. The strongest association, nivolumab\u0026ndash;myocarditis, is concordant with the malignant clinical phenotype of ICI-associated myocarditis described in contemporary registries and with FAERS-based pharmacovigilance observations in the ICI era. [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eThe temporal anomalies identified by the Isolation Forest (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) for nivolumab-associated myocarditis likely capture periods of stimulated reporting\u0026mdash;an expected feature of spontaneous reporting systems when clinical attention shifts. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] This supports the concept that AI-augmented surveillance can help separate steady-state reporting from dynamic surges that merit manual review, without implying changes in incidence. [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eBeyond ICIs, we observed signals consistent with known or biologically plausible cardio-oncology toxicities. Doxorubicin and trastuzumab were associated with cardiac failure signals, reflecting the persistent relevance of anthracycline and anti-HER2 cardiotoxicity in contemporary practice. [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] Tachyarrhythmia signals with paclitaxel and platinum-based therapy may reflect direct electrophysiologic effects, infusion-related stressors, or comorbidity burden in treated populations and merit targeted follow-up. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eA key design choice was restricting exposure to primary-suspect (PS) coding. In routine oncology care, polypharmacy is the rule rather than the exception; analyses that include concomitant and secondary suspect drugs risk extensive confounding and signal dilution. By using a within-cohort active comparator, we aimed to keep the comparison clinically coherent while reducing confounding by indication. Nonetheless, disproportionality remains a hypothesis-generating method rather than a causal estimator.\u003c/p\u003e \u003cp\u003eMultiple testing is an underappreciated threat in systematic pharmacovigilance screening. Using Benjamini\u0026ndash;Hochberg FDR control, we prioritized a manageable set of signals while maintaining sensitivity, and we provide the full screening table to mitigate selective reporting concerns. [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] Time-resolved anomaly detection via Isolation Forest offers a flexible, distribution-free approach to flag unusual months for manual review, but should be interpreted in clinical context because stimulated reporting can generate apparent surges. [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eLimitations are inherent to spontaneous reporting databases, including under-reporting, missing data, reporting bias, and limited clinical detail (e.g., dose, baseline cardiac status). [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] Denominator data are unavailable; thus, incidence cannot be estimated, and disproportionality remains hypothesis-generating rather than causal. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] In addition, sex was missing for a substantial proportion of reports, limiting sex-stratified inference. Despite these limitations, the present framework provides a reproducible and clinically oriented approach for prioritizing cardio-oncology signals to guide targeted follow-up studies.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn an oncology-focused FAERS cohort (2015 Q1\u0026ndash;2025 Q4), FDR-controlled disproportionality analysis identified 16 cardiac safety signals across representative antineoplastic agents. AI-based anomaly detection provided complementary time-resolved prioritization for the top signal. This framework can support cardio-oncology surveillance and inform hypothesis-driven follow-up studies.\u003c/p\u003e"},{"header":"Declarations","content":" \u003cp\u003eThe curated dataset and analysis scripts supporting the conclusions of this study are openly available in the Zenodo repository (DOI: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5281/zenodo.18761110\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.18761110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e; version v1.0). FAERS raw data are publicly accessible via the U.S. Food and Drug Administration website (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.fda.gov\u003c/span\u003e\u003cspan address=\"https://www.fda.gov\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e \u003cp\u003eNot applicable. This study used publicly available, de-identified FAERS data; ethics approval and informed consent were not required.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot applicable.\u003c/p\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThe authors received no specific funding for this work.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAuthor 1: Conceptualization, Methodology, Software, Formal analysis, Data curation, Visualization, Writing \u0026ndash; original draft, Supervision. Author 2: Investigation, Validation, Writing \u0026ndash; review \u0026amp; editing. Author 3: Investigation, Data curation, Validation, Writing \u0026ndash; review \u0026amp; editing. Author 4: Methodology, Formal analysis, Visualization, Writing \u0026ndash; review \u0026amp; editing. Author 5: Data curation, Validation, Writing \u0026ndash; review \u0026amp; editing. Author 6: Investigation, Resources, Writing \u0026ndash; review \u0026amp; editing. Author 7: Resources, Validation, Writing \u0026ndash; review \u0026amp; editing. Author 8: Investigation, Data curation, Writing \u0026ndash; review \u0026amp; editing. Author 9: Software, Data curation, Visualization. Author 10: Methodology, Investigation, Writing \u0026ndash; review \u0026amp; editing. Author 11: Supervision, Project administration, Writing \u0026ndash; review \u0026amp; editing. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eNot applicable.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eFAERS data are publicly available from the U.S. Food and Drug Administration website. [3,4] Processed transparency tables (full 90-pair screening and subgroup outputs) are provided with this submission as Supplementary Tables S1\u0026ndash;S3. Code to reproduce the main tables and figures from the processed outputs is available at https://github.com/drman44/FAERS-CardioOncology-2015-2025.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLyon AR, Dent S, Stanway S, et al. 2022 ESC Guidelines on cardio-oncology developed in collaboration with the European Hematology Association (EHA), the European Society for Therapeutic Radiology and Oncology (ESTRO) and the International Cardio-Oncology Society (IC-OS). Eur Heart J. 2022;43(41):4229\u0026ndash;361. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/eurheartj/ehac244\u003c/span\u003e\u003cspan address=\"10.1093/eurheartj/ehac244\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMahmood SS, Fradley MG, Cohen JV, et al. Myocarditis in patients treated with immune checkpoint inhibitors. J Am Coll Cardiol. 2018;71(16):1755\u0026ndash;64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jacc.2018.02.037\u003c/span\u003e\u003cspan address=\"10.1016/j.jacc.2018.02.037\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFood US, Administration D. FDA Adverse Event Reporting System (FAERS): Latest Quarterly Data Files. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/fda-adverse-event-reporting-system-faers-latest-quarterly-data-files\u003c/span\u003e\u003cspan address=\"https://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/fda-adverse-event-reporting-system-faers-latest-quarterly-data-files\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 24 Feb 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eU.S. Food and Drug Administration. FAERS Quarterly Data Files Documentation. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/faers-quarterly-data-files-documentation\u003c/span\u003e\u003cspan address=\"https://www.fda.gov/drugs/fdas-adverse-event-reporting-system-faers/faers-quarterly-data-files-documentation\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 24 Feb 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMedDRA MSSO. MedDRA Introductory Guide (Version 27.0). March 2024. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://alt.meddra.org/files_acrobat/intguide_27_0_English.pdf\u003c/span\u003e\u003cspan address=\"https://alt.meddra.org/files_acrobat/intguide_27_0_English.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 24 Feb 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Puijenbroek EP, Bate A, Leufkens HGM, Lindquist M, Orre R, Egberts ACG. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol Drug Saf. 2002;11(1):3\u0026ndash;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/pds.668\u003c/span\u003e\u003cspan address=\"10.1002/pds.668\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaillie JL. Case-non-case studies: principle, methods, bias and interpretation. Therapie. 2019;74(2):225\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.therap.2019.01.006\u003c/span\u003e\u003cspan address=\"10.1016/j.therap.2019.01.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCutroneo PM, et al. Conducting and interpreting disproportionality analyses in pharmacovigilance: practical considerations. Front Drug Saf Regul. 2024. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fdsfr.2023.1323057\u003c/span\u003e\u003cspan address=\"10.3389/fdsfr.2023.1323057\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaldane JBS. The estimation and significance of the logarithm of a ratio of frequencies. Ann Hum Genet. 1956;20(4):309\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1469-1809.1955.tb01285.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1469-1809.1955.tb01285.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnscombe FJ. On estimating binomial response relations. Biometrika. 1956;43(3\u0026ndash;4):461\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/biomet/43.3-4.461\u003c/span\u003e\u003cspan address=\"10.1093/biomet/43.3-4.461\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B. 1995;57(1):289\u0026ndash;300.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu FT, Ting KM, Zhou ZH. Isolation Forest. In: 2008 Eighth IEEE International Conference on Data Mining (ICDM). 2008:413\u0026ndash;422. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/ICDM.2008.17\u003c/span\u003e\u003cspan address=\"10.1109/ICDM.2008.17\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Elm E, Altman DG, Egger M, Pocock SJ, G\u0026oslash;tzsche PC, Vandenbroucke JP, STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jclinepi.2007.11.008\u003c/span\u003e\u003cspan address=\"10.1016/j.jclinepi.2007.11.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBelger C, Abrahams C, Imamdin A, Lecour S. Doxorubicin-induced cardiotoxicity and risk factors. IJC Heart Vasculature. 2023;50:101332. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ijcha.2023.101332\u003c/span\u003e\u003cspan address=\"10.1016/j.ijcha.2023.101332\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNowsheen S, Viscuse PV, O\u0026rsquo;Sullivan CC, et al. Incidence, diagnosis, and treatment of cardiac toxicity from trastuzumab in patients with breast cancer. Curr Treat Options Cardiovasc Med. 2017;19:46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12609-017-0249-4\u003c/span\u003e\u003cspan address=\"10.1007/s12609-017-0249-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReeves DJ, Leffers J, Rao K. Immune checkpoint inhibitor related myocarditis reported through the FDA adverse event reporting system: pharmacovigilance trends in reporting and outcomes. Front Oncol. 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fonc.2025.1498817\u003c/span\u003e\u003cspan address=\"10.3389/fonc.2025.1498817\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcIntyre RS, Karpuram S, et al. Utility and limitations of the FDA adverse events reporting system public dashboard for safety analyses: a case study with vesicular monoamine transporter 2 inhibitors. Expert Opin Drug Saf. 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/14740338.2025.2588634\u003c/span\u003e\u003cspan address=\"10.1080/14740338.2025.2588634\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrisafulli S, Bate A, Brown JS, et al. Interplay of spontaneous reporting and longitudinal healthcare databases for signal management: position statement from the International Society of Pharmacovigilance Real-World Evidence and Big Data Special Interest Group. Drug Saf. 2025;48(9):959\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s40264-025-01548-3\u003c/span\u003e\u003cspan address=\"10.1007/s40264-025-01548-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eİşleyen HB, Sucuoglu ZFAERS, Cardio-Oncology. (2015\u0026ndash;2025): PS-only disproportionality\u0026thinsp;+\u0026thinsp;AI emerging signal detection. Zenodo. 2026. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5281/zenodo.18761110\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.18761110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"pharmacovigilance, cardio-oncology, FAERS, reporting odds ratio, false discovery rate, immune checkpoint inhibitor, myocarditis, anomaly detection","lastPublishedDoi":"10.21203/rs.3.rs-8961206/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8961206/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eCardiac adverse events increasingly shape the real-world tolerability of systemic cancer therapies, and contemporary cardio-oncology practice underscores the need for structured surveillance\u0026mdash;particularly for immune checkpoint inhibitor (ICI) myocarditis. Rare but high-impact toxicities are often detected post-marketing. We developed an oncology-focused pharmacovigilance framework that combines false discovery rate (FDR)-controlled disproportionality screening with AI-based detection of emerging signals in the U.S. FDA Adverse Event Reporting System (FAERS).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe analyzed FAERS quarterly data (2015 Q1\u0026ndash;2025 Q4), de-duplicated reports by CASEID (retaining the most recent PRIMARYID), and restricted to primary-suspect exposure to 10 representative antineoplastic agents. Cardiac outcomes were pre-specified MedDRA preferred terms. Disproportionality was quantified using the reporting odds ratio (ROR) with Fisher\u0026rsquo;s exact test and sparse-cell correction (Haldane\u0026ndash;Anscombe); p-values were adjusted across 90 comparisons using Benjamini\u0026ndash;Hochberg FDR. For the top signal, monthly counts were evaluated with Isolation Forest to flag anomalous increases. We report in line with STROBE guidance for observational studies.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAfter de-duplication, the final primary-suspect oncology cohort comprised 99,515 unique FAERS reports (377,615 drug\u0026ndash;reaction records). Across 90 pre-specified drug\u0026ndash;event pairs, 16 cardiac safety signals met FDR validation (q\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The strongest myocarditis signal was observed with nivolumab (ROR 7.76, 95% CI 6.39\u0026ndash;9.42; a\u0026thinsp;=\u0026thinsp;337), followed by pembrolizumab (ROR 3.85, 95% CI 2.73\u0026ndash;5.43; a\u0026thinsp;=\u0026thinsp;35). Established cardiotoxicities were recovered as internal controls, including doxorubicin-associated cardiac failure (ROR 2.84; a\u0026thinsp;=\u0026thinsp;178) and trastuzumab-associated cardiac failure (ROR 2.57; a\u0026thinsp;=\u0026thinsp;112). AI-based time-series anomaly detection highlighted discrete temporal surges for the top-ranked signal, supporting time-resolved prioritization of potential emerging toxicities.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eIn an oncology-focused FAERS cohort, FDR-controlled disproportionality analysis identified multiple cardiac safety signals, while AI-based anomaly detection provided complementary time-resolved prioritization. This reproducible framework can support cardio-oncology surveillance and guide hypothesis-driven follow-up studies.\u003c/p\u003e","manuscriptTitle":"AI-augmented pharmacovigilance for cardio-oncology: FDR- controlled disproportionality and anomaly detection of antineoplastic-associated cardiac events in FAERS (2015 Q1–2025 Q4)","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-03 16:39:13","doi":"10.21203/rs.3.rs-8961206/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-03-30T22:19:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"260618281269581137962460703860552753425","date":"2026-03-24T11:20:20+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-17T16:43:49+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-26T14:49:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-26T00:48:49+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-26T00:47:42+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Cancer","date":"2026-02-24T21:28:39+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-cancer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bcan","sideBox":"Learn more about [BMC Cancer](http://bmccancer.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bcan/default.aspx","title":"BMC Cancer","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"84d8dc0c-edb7-4016-bf01-d082de110f0e","owner":[],"postedDate":"April 3rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-03T16:39:14+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-03 16:39:13","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8961206","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8961206","identity":"rs-8961206","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}