Validation of a novel semi-automated ECG quantification tool, applied to a cardio-oncology setting Semi-Automated ECG Tool applied to cardio-oncology

Validation of a novel semi-automated ECG quantification tool, applied to a cardio-oncology setting Semi-Automated ECG Tool applied to cardio-oncology | 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 Validation of a novel semi-automated ECG quantification tool, applied to a cardio-oncology setting Semi-Automated ECG Tool applied to cardio-oncology Samuel D. COHEN, Maxime ROBERT-HALABI, Adrien PROCUREUR, Mathieu JAMELOT, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7204236/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 19 Dec, 2025 Read the published version in Cardio-Oncology → Version 1 posted 7 You are reading this latest preprint version Abstract Background. Electrocardiogram (ECG) analysis is crucial to detect cardiotoxicity. Manual methods are time-consuming and limited by inter-reader variability, highlighting the need for precise, reproducible and rapid semi-automated digital tools in clinical practice. Objective. This study evaluates the triplicate concatenation method (TCM) using a semi-automated ECG software (CalECG-4.2, AMPS ®) by assessing intra-and inter-reader variabilities in two distinct cardio-oncology populations: breast cancer patients receiving ribociclib (a QT-prolonging drug) and patients admitted with severe immune checkpoint inhibitors (ICI)-myocarditis, a condition marked by QRS alterations. Methods. A total of 420 ECG from 31 patients (21 ribociclib, and 10 ICI-myocarditis) were independently analyzed by two readers. Variability was assessed using Bland-Altman analyses and intraclass correlation coefficients (ICC). Nonlinear mixed-effects modelling quantified time-dependent changes in heart rate (HR), PR, QTc (Friderica’s HR correction), QRS duration and voltage (Sokolow-Lyon) accounting for inter-reader variability. Results. Intra and inter-reader reproducibility was excellent (ICC>0.99, including Sokolow-Lyon voltage; standard-deviation<4ms across all time-derived parameters). In ribociclib-treated patients (cycles of 21/28 days on drug), QTc peaked at day 14 (16±1ms, p<0.001) before decreasing by day 28 (-6±1ms, p<0.001) compared to baseline. In ICI-myocarditis, QRS duration increased at day 5 before returning to baseline starting day 28, while Sokolow-Lyon voltages increased progressively on immunosuppressive treatments, peaking at day 28 (458±49µV, p<0.001) and remaining constant afterwards for the next month. Conclusion. TCM with CalECG-4.2 ensures a high reproducibility while monitoring key parameters like QTc duration and Sokolow-Lyon voltage, making it a reliable and time-saving alternative for the ECG surveillance of drug toxicities in cardio-oncology. Methods cardio-oncology QT interval ECG pharmacology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background The electrocardiogram (ECG) is an essential tool in clinical and research settings to diagnose and monitor cardiac diseases. Cardio-oncology has emerged as a new field specializing in the surveillance and treatment of cardiovascular toxicities caused by various cancer therapies.( 1 – 4 ) Such treatments include cyclin-dependent kinase inhibitors, like ribociclib, which may lead to life-threatening torsades de pointes by prolonging the heart rate corrected QT interval (QTc).( 5 – 7 ) On the other hand, immune checkpoint inhibitors (ICI) can induce severe myocarditis, a rare but potentially fatal condition.( 8 – 11 ) ECG monitoring of such patients revealed several conduction abnormalities, such as PR and QRS prolongation as well as voltage changes.( 12 , 13 ) The latter two have been associated with worse outcomes in ICI-myocarditis, as they likely indicate myocardial edema and/or damage from macrophage infiltration.( 12 , 13 ) A thorough monitoring of these ECG parameters is thus indicated but may be challenging in clinical practice. While manual measurement methods are widely used, they are time-consuming and prone to inter-reader, intra-reader and beat-to-beat variability, which limits their reliability.( 14 – 16 ) For instance, QTc measurements can vary due to differences in baseline selection and correction formulas applied, leading to inconsistencies that limit the diagnostic accuracy.( 15 , 16 ) Similarly, since the assessment of Sokolow voltage is highly dependent on the precision of manual tracing measurements, it is intrinsically associated to a high degree of variability and poor reliability.( 17 , 18 ) Software using automated methods relying on mathematical concepts, such as CalECG, offer a promising alternative by providing consistent and precise measurements, particularly in complex clinical settings ( 14 ). Notwithstanding these improvements, accurate automated ECG measurements may still be hindered by signal noise and artifacts. To address these issues and further enhance the signal-to-noise ratio, triplicate ECG of 10sec and superimposed median beats have been implemented ( 18 , 19 ). Semi-automated methods, such as the triplicate assessment method (TAM), are widely adopted. TAM consists of repeating measurements three times (individual analysis of 10sec ECG within a triplicate series) to ensure accuracy and reliability, which, although more precise, can be labor-intensive. The triplicate concatenation method (TCM) has been introduced as a novel and more efficient technique that build a 30-second ECG by concatenating three 10-second tracing while excluding artifacts at the concatenation points.( 14 , 20 ) Although promising, this novel method remains to be validated in large real-life patient populations, particularly for automated QRS voltage and Sokolow calculations, which were previously unavailable in readily available ECG analysis software. Building on prior research validating the CalECG 3.7 software® for QTc measurement ( 14 ), this study aims to evaluate the reproducibility of TCM using the updated CalECG 4.2 software®. This updated version integrates a more precise analysis of the QRS segments and its voltage, enabling a semi-automated calculation of the Sokolow-Lyon voltage, a feature previously unavailable. By assessing intra- and inter-reader variabilities as well as characterizing ECG changes over time, we seek to determine whether TCM offers a reliable alternative for ECG interval measurements in cardio-oncology patients. Specifically, we focus on patients treated with ribociclib or followed for immune checkpoint inhibitor (ICI)-induced myocarditis, since these patient populations exhibit greater ECG variability, thereby providing a robust test for this novel method. Methods Study Population We recruited thirty-one patients followed in a French quaternary care University-affiliated cardio-oncology center (Pitié-Salpêtrière Hospital, Paris) and enrolled in NEOCARDIO ( NCT03882580 , approved by the Ethics Committee of Sorbonne University on March 12th, 2021), an observational, ambispective cohort. They were first seen between December 2022 and July 2024 and then evaluated prospectively. On inclusion, each patient had 3 resting 12-lead 10-sec ECG, each separated by 2-minute intervals. Patients were placed in supine position and asked to remain immobile during the ECG acquisition to minimize physiological variability and artefacts. Included patients were required to have high-quality non-electro stimulated ECG in sinus rhythm. Included patients were divided into two groups: Breast cancer patients : Twenty-one patients undergoing their first cycle of ribociclib. Each patient is followed for three visits: at day 0 (baseline), at days 14 ± 3 and 28 ± 3 (post-treatment). Ribociclib was administered on a 28-day cycle, with 21 days on-treatment followed by 7 days off-treatment, as part of the standard treatment regimen. Our main focus was to evaluate the treatment-induced effect on the QTc interval, as ribociclib is known to prolong it ( 4 , 21 ). ICI myocarditis patients : Ten patients with severe ICI-induced myocarditis were enrolled. Each patient had 8 planned visits within the first two months of their initial admission. The immunosuppressive regimen consisted of abatacept administered by intravenous injection with concomitant oral ruxolitinib and corticosteroids. Their individual dosing and administration regimen have previously been published ( 22 , 23 ). The main focus was to study the evolution of the QRS voltage and duration while on treatment, as these parameters have been shown useful to assess myocardial injury and inflammation and may represent a surrogate marker for disease severity and treatment response ( 12 , 24 ). In this ECG software validation study, six key ECG parameters were analyzed: heart rate (HR), PR interval, QRS duration, corrected QT interval for HR using Fridericia's formula (QTc) ( 25 ), and Sokolow-Lyon voltage, calculated using the formula: (S wave maximal voltage in leadV1) + (R wave maximal voltage in lead V5 or lead V6).( 26 ) Electrocardiogram Recording and Analysis ECG were recorded using a standardized digital electrocardiograph (ELI 280, V2.4.1.8; Mortara Instrument, Inc., Milwaukee, WI, USA) by trained nurses. The device has a sampling rate of 1000 Hz and a filter of 150 Hz, which were selected to capture detailed signal characteristics for subsequent analysis. In total, 420 ECG (140 triplicate sets) were analyzed ( Fig. 1 for detailed flow-chart ) . Quantitative ECG analysis was performed using the TAM and the TCM, both being semi-automated approaches. Methods variability were assessed by comparing inter-method variability (TAM vs TCM#1 reader B) and intra-method variability (TCM#1 vs TCM#2 reader B). TCM#1 corresponds to the first analysis of the triplicate, while TCM#2 refers to the second analysis. Inter-reader variability for TCM was determined by comparing results between reader A and reader B (Fig. 1 ). Reader A (JES), a Cardiologist, has over 15 years of experience in ECG analysis, while reader B (SDC), a PharmD, has more than 3 years of experience in ECG analysis. Both readers independently reviewed and analyzed the ECG. They remained blinded to each other's measurements and the study hypotheses throughout the process. CalECG 4.2 Software® Overview All ECG measurements were conducted using the latest version of the CalECG 4.2 software®, which automatically generates a representative beat for each of the 12 leads derived from detected sinus rhythm beats. This process involves aligning sinus rhythm ECG beats on the R-wave’s peak for each lead and creating a median beat by computing a median value for each ECG sample, producing a single representative signal, 1.2s long, per lead. As a result, the representative beat is not an actual ECG recorded beat but rather an averaged signal, synthesizing the characteristics of all recorded sinus rhythm beats in each lead. To build this representative beat, the software automatically eliminates all premature ventricular contractions within a 10-second ECG recording. The final output includes 12 representative beats, one for each lead. By superimposing these single-lead representative beats, a superimposed median beat is generated. This superimposed median beat is optimally represented by a vector of magnitude, calculated as the square root of the sum of the squared representative beats. The vector of magnitude facilitates automated measurements of PR, QT and QRS intervals and voltages using a threshold-based method. If automatic PR/QT/QRS fiducial marks are misplaced, users retain the ability to manually adjust the fiducial markers for critical points, including the onset of the P-wave, the onset and offset of the QRS complex or the offset of the T-wave. Additional ECG parameters of interest, particularly QRS voltages can be measured using the same process, as illustrated in Fig. 2 . Compared with the previous version (CalECG 3.7 software®) ( 14 ), the main improvement of the 4.2 version is a comprehensive revision of the AMPS Bravo algorithm. This update specifically targets waveforms within the QRS complex (Q, R, S, R’, S’), which were previously excluded from automated analysis and could only be measured manually in the 3.7 version. The addition of automated QRS voltage analysis, integrating Sokolow-Lyon and microvoltage assessments, enhances measurement precision and reliability. Statistical Analyses Bland-Altman plots and intra-class correlation coefficients (ICC) were used to evaluate the measurements’ reliability between the two readers/readings.( 27 , 28 ) Bland-Altman plots graphically represent the agreement between the two readers/readings, plotting the mean difference (bias) and the limits of agreement (LOA) (mean difference ± 1.96 standard deviation [SD]) against the average of the two measurements. Additionally, nonlinear mixed-effects modelling (NLME) was applied to analyze the influence of random effects (i.e., patient’s variability) and fixed effects (e.g., measurement method, or reader) on ECG parameters. All statistical analyses were performed using R software (version 4.4.2). A p-value of < 0.05 was deemed significant. Results The ICC and Bland-Altman analysis (including LOA) for the method/readings/readers comparisons are summarized in Table 1 and Figure 3 . Measurements variability with CalECG4.2® Inter-reader variability for TCM measurements demonstrated excellent consistency across measurements for all ECG parameters (ICC>0.994, see Table 1 ). For QTc, the mean bias was 0.02±1.32ms (LOA: -2.57 to 2.61 ms). The PR interval exhibited a mean bias of 0.47±2.13ms (LOA:-3.70 to 4.65ms). The QRS duration showed a mean bias of 1.08±2.77ms (LOA: -4.35 to 6.51ms). The Sokolow-Lyon voltage using lead V5 demonstrated a mean bias of 0.40±4.78 μV (LOA:-8.96 to 9.76µV). Inter-method variability between TAM and TCM demonstrated high consistency across measurements for all ECG parameters (ICC >0.989, see Table 1 ). For QTc, the mean bias was -1.17±3.12ms (LOA: -7.28 to 4.93ms). The PR interval exhibited a mean bias of 0.53±3.02ms (LOA: -5.39 to 6.44ms). The QRS duration showed a mean bias of 0.24 ± 2.17ms (LOA: -4.01 to 4.50ms). The Sokolow-Lyon voltage using lead V5 demonstrated a mean bias of 10.35±24.10μV (LOA: -36.88 to 57.58µV). Intra-method variability within TCM demonstrated excellent repeatability across measurements for all ECG parameters (ICC > 0.991, see Table 1 ). For QTc, the mean bias was -0.31±1.78ms (LOA: -3.80 to 3.19ms). The QRS duration showed a mean bias of -1.17±3.23ms (LOA: -7.50 to 5.15ms). The Sokolow-Lyon voltage using lead V5 demonstrated a mean bias of -0.28±2.44μV (LOA: -5.08 to 4.51µV). Application in real life cohort: ribociclib-treated breast cancer patients and ICI myocarditis patients In ribociclib-treated breast cancer patients, ECG parameters exhibited significant time-dependent changes (Figure 4) . NLME models were adjusted for visit type (D0, D14±3, D28±3), methods (TCM vs. TAM), readers (A vs. B), readings (reading #1 vs. #2), and age. The variables "methods", "readers", and "readings" did not show significant effects (Table S1). By day 14 (vs. D0), HR decreased slightly (-3±1 bpm, p<0.001), QTc increased notably (16±1ms, p<0.001), QRS duration shortened significantly (-1.5±0.5 ms, p=0.001), and the Sokolow voltage using lead V6 rose significantly (40±19 µV, p<0.05), with similar trends using lead V5. By day 28 (vs. D0), QTc decreased (-6±1ms, p<0.001), returning below baseline values, while the PR interval showed a modest increase (3±1ms, p=0.006). Both HR, Sokolow voltage and QRS duration also returned to baseline levels (see Figure 4 and Table S1 ). In severe ICI myocarditis patients treated by abatacept, ruxolitinib and corticosteroids, ECG parameters exhibited distinct temporal patterns (Figure 5) . NLME models were adjusted for visit type (D0, and 7 other visits between D5±3 to D60±5), methods (TCM vs. TAM), readers (A vs. B), readings (reading #1 vs. #2), age, and sex. The variables "methods", "readers", and "readings" did not show significant effects (Table S2) . By day 5 (vs. D0), HR declined significantly (-11±2 bpm, p<0.001) and QRS duration increased (6±1ms, p<0.001). The HR reduction persisted through day 60 (-12±3 bpm, p<0.001). The PR interval showed modest reductions at day 14 (-6±2 ms, p<0.001) and day 50 (-5±2ms, p=0.001). From day 21 through day 60, QTc decreased significantly, with the largest reduction observed by day 28 (-26±3ms, p<0.001), and smaller yet significant decreases persisted through day 60 (-10±3 ms, p<0.001). From day 14 through day 60, QRS duration returned to baseline values, while the Sokolow-Lyon voltage using lead V5 increased significantly (442 ± 48 µV, p<0.001), with the maximum increase observed by day 28 (458 ± 49 µV, p<0.001). This elevation persisted throughout the follow-up period ( see Figure 5 and Table S2 ). Discussion This study evaluated the inter- and intra-reader variability of TCM for ECG measurements and compared it to the traditional TAM across critical ECG intervals in two distinct real-life standardized cardio-oncology patient populations. Our findings demonstrate that TCM provides high reproducibility for key ECG parameters, including HR, PR interval, QRS duration, QTc, and Sokolow voltages using V5 and V6, with minimal bias and narrow limits of agreement. These results underscore the robustness and reliability of TCM, as they align with previous studies reporting comparable LOA values for key ECG parameters, such as QTc measured in dedicated Thorough QT trials ( 19 , 20 , 28 – 30 ). In practice, CalECG4.2® is even more accurate, due to its semi-automated QRS onset analysis, which improved upon the previous version that required manual QRS positioning ( 14 ). Semi-automated techniques, such as the TCM, offer superior precision compared to traditional manual measurements, with a reported inter-reader variability for QTc measurements ranging in terms of SD from 6ms for semi-automated methods to 15ms for manual methods ( 31 , 32 ). In comparison, our study showed intra and inter-reader variability below 3ms, demonstrating a level of precision that aligns with the highest reported accuracies. Such accuracy level of QTc interval measurement is crucial to easily identify drug-induced QT prolongation in real-life setting, and moreover to decrease the numbers of required subjects needed to identify QT-prolongation induced by drugs in clinical trials.( 19 , 30 ) The CALECG 4.2® system, which relies on classical mathematical approaches and vector analysis, has shown to be more precise than deep learning models for this specific task ( 33 , 34 ). Deep learning models, while powerful, initially learn from human measurements that inherently carries variability. For instance, Bos et al. demonstrated that deep learning models can effectively identify concealed long QT syndrome from ECG, but their accuracy was inherently limited by the quality and consistency of the training data, with reported measurement errors up to 10 ms.( 33 ) Similarly, Giudicessi et al. highlighted the potential of AI-enabled ECG analysis but also reported variability in QTc measurements standard deviations ranging from 5 to 15 ms.( 34 ) In contrast, classical methods like those employed in CalECG 4.2 currently offer superior precision for straightforward QTc measurements since it relies on deterministic algorithms that are less susceptible to human measurement errors. Therefore, while deep learning holds promise for complex tasks,( 35 ) its applicability to QTc measurement may need further refinement to match the precision of classical methods. The integration of NLME modelling in this study provided a detailed assessment of time-dependent variations in ECG parameters in response to cardiotoxic treatments. Our results highlight a significant temporal effect of ribociclib on QTc, PR, QRS, and Sokolow-Lyon indices, underscoring the dynamic nature of treatment-induced cardiotoxicity. QTc significantly increased by day 14 and shortened back to baseline by day 28, reflecting the transient impact of ribociclib on ventricular repolarization. This pattern aligns with ribociclib's administration schedule, which involves a 21-day treatment cycle followed by a 7-day break.( 36 , 37 ) Given its plasma half-life ranging from 29.7 to 54.7 hours, higher circulating drug levels are expected during the treatment phase, contributing to the observed ECG changes. Additionally, the PR interval prolongation observed by day 28 indicates a possible atrioventricular conduction delay induced by ribociclib, independent of concentration, unlike its well-documented QT prolonging effect ( 30 , 38 ). PR interval prolongation may be less reversible and warrants further evaluation, particularly regarding its potential impact on clinical conduction disturbances. Our study uniquely expands the understanding of ribociclib’s electrophysiological effects by highlighting its possible novel influence on the PR interval, which has not been previously emphasized. However, this relationship requires further clinical evaluations to be fully understood. We also observed a significant increase in Sokolow-Lyon voltage by day 14 on ribociclib. By day 28, the voltage returned near its baseline level. The significance of these changes is unclear, but may suggest a broader electrophysiological impact of ribociclib beyond the known QTc prolongation. This highlights the need for comprehensive assessments in future studies to fully understand its effects on cardiac electrophysiology. Historically, Sokolow-Lyon voltages have been used to assess left ventricular hypertrophy, although their usefulness has been limited by variability in manual measurements.( 26 ) Nevertheless, when compared to cardiac magnetic resonance imagining or echocardiography, which provide more precise measurements of left ventricular hypertrophy, Sokolow-Lyon voltages often show some level of concordance in identifying significant hypertrophy.( 39 ) Recent studies suggest that Sokolow-Lyon voltages may also be a marker of proper left intraventricular conduction as it represent the summation of action potentials occurring simultaneously.( 24 ) In cases of myocardial toxicity, particularly those mediated by lymphocytes and macrophages inflammation, the conduction can become disorganized, leading to a lower widespread QRS voltage. This phenomenon is observed in ICI-myocarditis, where the initial disruption in conduction results in a low Sokolow-Lyon voltage.( 12 , 22 , 40 ) The identification of its improvement under immunosuppressive treatment in this work is novel and highlights its potential as a recovery marker. This perspective shifts the focus from hypertrophy to a broader assessment of electrophysiological integrity. Limitations This study has several limitations. Since our analysis focused on two specific patient cohorts, it may not fully capture the variability present in other clinical settings. Variability tends to be lower among healthy volunteers without underlying pathology compared to patients with cardiac conditions, as the latter group presents more complex and variable ECG patterns. This study specifically focuses on patients with cardiac conditions to validate the method's robustness in a highly variable context. Additionally, although ECG analysis is widely used in clinical practice, it is important to acknowledge that the variability in measurements can be influenced by the quality of the ECG waveform. The near-perfect QRS complexes and overall waveform obtained in this study may contribute to the narrow variability, which might not be replicated in all clinical care settings where non-standardized ECG may be performed by less trained workers. Additionally, this study primarily serves as a validation for our ECG measurement method, and we were unable to integrate comprehensive clinical data and concomitant treatments that could potentially alter ECG parameters beyond the few demographic variables considered in the pathological settings used herein. Future research should aim to incorporate these factors to provide a more complete evaluation. Conclusion TCM demonstrates high reproducibility and minimal variability, establishing itself as a reliable alternative to TAM for ECG interval measurements. Its ability to capture time-dependent changes in ECG parameters, such as QTc, PR, QRS, and Sokolow-Lyon voltages, highlights its potential for monitoring drug-induced cardiotoxicity in cardio-oncology. Abbreviations ECG Electrocardiogram HR Heart rate ICC Intraclass correlation coefficient ICI Immune checkpoint inhibitor LOA Limits of agreement NLME Nonlinear mixed-effects modelling PR PR interval QTc Corrected QT interval (using Fridericia’s formula) QRS QRS complex duration SD Standard deviation TAM (Triplicate assessment method) A standard method where three separate 10 seconds ECGs are recorded and averaged to reduce beat-to-beat variability TCM (Triplicate concatenation method) Semi-automated method where three 10 seconds ECG tracings are merged into a single waveform Declarations Conflict of interest statement : SC, M.R-H, AP, MJ, MV, FB and EP have nothing to disclose. J.-E.S. has received financial support from Novartis, BeiGene, BMS and Banook Group; and holds patents related to the prognostication and treatment of ICI myocarditis. Funding This study was supported by the ANR-20-CE17-0022 DeepECG4U funding from the French National Research Agency (ANR), by CIC-1901.This work was supported by the Sorbonne Center for Artificial Intelligence (SCAI), an institute dedicated to artificial intelligence within the Sorbonne University Alliance, funded by the excellence initiative IDEX SUPER under the ANR reference: 11-IDEX-0004. Author Contribution Samuel CohenData curation-Equal, Formal analysis-Lead, Investigation-Equal, Software-Equal, Writing - original draft-EqualMaxime robert-halabiData curation-Supporting, Writing - original draft-SupportingAdrien ProcureurData curation-Equal, Investigation-Equal, Software-Equal, Writing - review & editing-EqualMathieu JAMELOTData curation-Supporting, Investigation-Supporting, Writing - review & editing-SupportingMartino VaglioProject administration-Supporting, Software-Lead, Validation-Supporting, Writing - review & editing-EqualFabio BadiliniCRediT contribution not specifiedEdi PriftiConceptualization-Supporting, Methodology-Equal, Supervision-Equal, Validation-Equal, Visualization-Lead, Writing - review & editing-EqualJoe-Elie SalemConceptualization-Lead, Formal analysis-Supporting, Funding acquisition-Lead, Investigation-Lead, Methodology-Lead, Project administration-Lead, Resources-Lead, Software-Lead, Supervision-Lead, Validation-Lead, Visualization-Equal, Writing - original draft-Lead Acknowledgement The authors and study sponsor are indebted to the patients and their families, as well as the investigators participating in this study. Data Availability The data supporting the findings of this study are available from the corresponding author upon reasonable request. References Morelli MB, Bongiovanni C, Da Pra S, Miano C, Sacchi F, Lauriola M, et al. Cardiotoxicity of Anticancer Drugs: Molecular Mechanisms and Strategies for Cardioprotection. Front Cardiovasc Med. 2022;9:847012. Nguyen LS, Dolladille C, Drici M-D, Fenioux C, Alexandre J, Mira J-P, et al. Cardiovascular Toxicities Associated With Hydroxychloroquine and Azithromycin: An Analysis of the World Health Organization Pharmacovigilance Database. Circulation. 2020;142(3):303-5. Salem J-E, Manouchehri A, Moey M, Lebrun-Vignes B, Bastarache L, Pariente A, et al. Cardiovascular toxicities associated with immune checkpoint inhibitors: an observational, retrospective, pharmacovigilance study. Lancet Oncol. 2018;19(12):1579-89. Salem J-E, Nguyen LS, Moslehi JJ, Ederhy S, Lebrun-Vignes B, Roden DM, et al. Anticancer drug-induced life-threatening ventricular arrhythmias: a World Health Organization pharmacovigilance study. European Heart Journal. 2021;42(38):3915-28. Alexandre J, Moslehi JJ, Bersell KR, Funck-Brentano C, Roden DM, Salem JE. Anticancer drug-induced cardiac rhythm disorders: Current knowledge and basic underlying mechanisms. Pharmacol Ther. 2018;189:89-103. Salem JE, Nguyen LS, Moslehi JJ, Ederhy S, Lebrun-Vignes B, Roden DM, et al. Anticancer drug-induced life-threatening ventricular arrhythmias: a World Health Organization pharmacovigilance study. Eur Heart J. 2021;42(38):3915-28. Postema PG, Wilde AAM. The Measurement of the QT Interval. Curr Cardiol Rev. 2014;10(3):287-94. Johnson DB, Balko JM, Compton ML, Chalkias S, Gorham J, Xu Y, et al. Fulminant Myocarditis with Combination Immune Checkpoint Blockade. N Engl J Med. 2016;375(18):1749-55. Fenioux C, Abbar B, Boussouar S, Bretagne M, Power JR, Moslehi JJ, et al. Thymus alterations and susceptibility to immune checkpoint inhibitor myocarditis. Nat Med. 2023;29(12):3100-10. Lehmann LH, Heckmann MB, Bailly G, Finke D, Procureur A, Power JR, et al. Cardiomuscular Biomarkers in the Diagnosis and Prognostication of Immune Checkpoint Inhibitor Myocarditis. Circulation. 2023;148(6):473-86. Salem JE, Bretagne M, Abbar B, Leonard-Louis S, Ederhy S, Redheuil A, et al. Abatacept/Ruxolitinib and Screening for Concomitant Respiratory Muscle Failure to Mitigate Fatality of Immune-Checkpoint Inhibitor Myocarditis. Cancer Discov. 2023;13(5):1100-15. Power JR, Alexandre J, Choudhary A, Ozbay B, Hayek S, Asnani A, et al. Electrocardiographic Manifestations of Immune Checkpoint Inhibitor Myocarditis. Circulation. 2021;144(18):1521-3. Power JR, Alexandre J, Choudhary A, Ozbay B, Hayek SS, Asnani A, et al. Association of early electrical changes with cardiovascular outcomes in immune checkpoint inhibitor myocarditis. Arch Cardiovasc Dis. 2022;115(5):315-30. Saqué V, Vaglio M, Funck-Brentano C, Kilani M, Bourron O, Hartemann A, et al. Fast, accurate and easy-to-teach QT interval assessment: The triplicate concatenation method. Archives of Cardiovascular Diseases. 2017;110(8-9):475-81. Malik M, Hnatkova K, Schmidt A, Smetana P. Accurately measured and properly heart-rate corrected QTc intervals show little daytime variability. Heart Rhythm. 2008;5(10):1424-31. Azie NE, Adams G, Darpo B, Francom SF, Polasek EC, Wisser JM, et al. Comparing methods of measurement for detecting drug-induced changes in the QT interval: implications for thoroughly conducted ECG studies. Annals of Noninvasive Electrocardiology: The Official Journal of the International Society for Holter and Noninvasive Electrocardiology, Inc. 2004;9(2):166-74. Farb A, Devereux RB, Kligfield P. Day-to-day variability of voltage measurements used in electrocardiographic criteria for left ventricular hypertrophy. J Am Coll Cardiol. 1990;15(3):618-23. Schijvenaars BJA, van Herpen G, Kors JA. Intraindividual variability in electrocardiograms. J Electrocardiol. 2008;41(3):190-6. Natekar M, Hingorani P, Gupta P, Karnad DR, Kothari S, de Vries M, et al. Effect of number of replicate electrocardiograms recorded at each time point in a thorough QT study on sample size and study cost. J Clin Pharmacol. 2011;51(6):908-14. Hingorani P, Karnad DR, Ramasamy A, Panicker GK, Salvi V, Bhoir H, et al. Semiautomated QT interval measurement in electrocardiograms from a thorough QT study: comparison of the grouped and ungrouped superimposed median beat methods. J Electrocardiol. 2012;45(3):225-30. Richardson DR, Parish PC, Tan X, Fabricio J, Andreini CL, Hicks CH, et al. Association of QTc Formula With the Clinical Management of Patients With Cancer. JAMA Oncol. 2022;8(11):1616-23. Salem J-E, Bretagne M, Abbar B, Leonard-Louis S, Ederhy S, Redheuil A, et al. Abatacept/Ruxolitinib and Screening for Concomitant Respiratory Muscle Failure to Mitigate Fatality of Immune-Checkpoint Inhibitor Myocarditis. Cancer Discov. 2023;13(5):1100-15. Salem J-E, Ederhy S, Belin L, Zahr N, Tubach F, Procureur A, et al. Abatacept dose-finding phase II triaL for immune checkpoint inhibitors myocarditis (ACHLYS) trial design. Archives of Cardiovascular Diseases. 2025;118(2):106-15. Power JR, Alexandre J, Choudhary A, Ozbay B, Hayek SS, Asnani A, et al. Association of early electrical changes with cardiovascular outcomes in immune checkpoint inhibitor myocarditis. Archives of Cardiovascular Diseases. 2022;115(5):315-30. Food, Drug Administration HHS. International Conference on Harmonisation; guidance on E14 Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic Drugs; availability. Notice. Fed Regist. 2005;70(202):61134-5. Sokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as obtained by unipolar precordial and limb leads. Am Heart J. 1949;37(2):161-86. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307-10. Koo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15(2):155-63. Sarapa N, Gussak I, Vajdic B, George S, Hadzievski L, Francom SF, et al. Comparison of QTinno, a fully automated electrocardiographic analysis program, to semiautomated electrocardiographic analysis methods in a drug safety study in healthy subjects. J Electrocardiol. 2009;42(4):358-66. Tyl B, Kabbaj M, Fassi B, De Jode P, Wheeler W. Comparison of semiautomated and fully automated methods for QT measurement during a thorough QT/QTc study: variability and sample size considerations. J Clin Pharmacol. 2009;49(8):905-15. Barbey JT, Connolly M, Beaty B, Krantz MJ. Man versus Machine: Comparison of Automated and Manual Methodologies for Measuring the QTc Interval: A Prospective Study. Annals of Noninvasive Electrocardiology : The Official Journal of the International Society for Holter and Noninvasive Electrocardiology, Inc. 2015;21(1):82-90. Darpo B, Agin M, Kazierad DJ, Layton G, Muirhead G, Gray P, et al. Man versus machine: is there an optimal method for QT measurements in thorough QT studies? J Clin Pharmacol. 2006;46(6):598-612. Bos JM, Attia ZI, Albert DE, Noseworthy PA, Friedman PA, Ackerman MJ. Use of Artificial Intelligence and Deep Neural Networks in Evaluation of Patients With Electrocardiographically Concealed Long QT Syndrome From the Surface 12-Lead Electrocardiogram. JAMA Cardiol. 2021;6(5):532. Giudicessi JR, Schram M, Bos JM, Galloway CD, Shreibati JB, Johnson PW, et al. Artificial Intelligence-Enabled Assessment of the Heart Rate Corrected QT Interval Using a Mobile Electrocardiogram Device. Circulation. 2021;143(13):1274-86. Prifti E, Fall A, Davogustto G, Pulini A, Denjoy I, Funck-Brentano C, et al. Deep learning analysis of electrocardiogram for risk prediction of drug-induced arrhythmias and diagnosis of long QT syndrome. Eur Heart J. 2021;42(38):3948-61. Tripathy D, Im SA, Colleoni M, Franke F, Bardia A, Harbeck N, et al. Ribociclib plus endocrine therapy for premenopausal women with hormone-receptor-positive, advanced breast cancer (MONALEESA-7): a randomised phase 3 trial. Lancet Oncol. 2018;19(7):904-15. Barber M, Nguyen LS, Wassermann J, Spano JP, Funck-Brentano C, Salem JE. Cardiac arrhythmia considerations of hormone cancer therapies. Cardiovasc Res. 2019;115(5):878-94. Metcalfe JZ, Economou T, Naufal F, Kucukosmanoglu M, Kleiman R, Phillips PPJ, et al. Validation of a Handheld 6-Lead Device for QT Interval Monitoring in Resource-Limited Settings. JAMA network open. 2024;7(6):e2415576. Antikainen RL, Grodzicki T, Palmer AJ, Beevers DG, Webster J, Bulpitt CJ, et al. Left ventricular hypertrophy determined by Sokolow-Lyon criteria: a different predictor in women than in men? J Hum Hypertens. 2006;20(6):451-9. Power JR, Dolladille C, Ozbay B, Procureur AM, Ederhy S, Palaskas NL, et al. Predictors and Risk Score for Immune Checkpoint-Inhibitor-Associated Myocarditis Severity. medRxiv. 2024:2024.06.02.24308336. Table 1 Table 1. Summary of ICC and Bland-Altman Analysis for Method Comparisons. This table displays ICC and Bland-Altman analysis results for three method comparisons assessing specific ECG parameters, including HR, PR, QRS, QTcF and Sokoloff indices (V5 and V6). Parameters ICC* Mean±SD Bias Upper LOA (95% CI) Lower LOA (95% CI) TCM 1 reader A vs. TCM 1 reader B (n=140) HR (bpm) 1 0±0 0 0 PR (ms) 0.995 0.47±2.13 4.65 -3.70 QRS (ms) 0.994 1.08±2.77 6.51 -4.35 QTcF (ms) 0.999 0.02±1.32 2.61 -2.57 Sokoloff V5 (µV) 0.999 0.40±4.78 9.76 -8.96 Sokoloff V6 (µV) 0.999 0.42±5.12 10.45 -9.60 TAM 1 reader B vs. TCM 1 reader B (n=140) HR (bpm) 0.999 -0.08±0.55 1.00 -1.16 PR (ms) 0.989 0.53±3.02 6.44 -5.39 QRS (ms) 0.997 0.24±2.17 4.50 -4.01 QTcF (ms) 0.992 -1.17±3.12 4.93 -7.28 Sokoloff V5 (µV) 0.999 10.35±24.10 57.58 -36.88 Sokoloff V6 (µV) 0.999 10.11±23.00 55.20 -34.98 TCM 1 reader B vs. TCM 2 reader B (n=140) HR (bpm) 1 0±0 0 0 PR (ms) 0.993 -0.61±2.32 3.93 -5.16 QRS (ms) 0.991 -1.17±3.23 5.15 -7.50 QTcF (ms) 0.998 -0.31±1.78 3.19 -3.80 Sokoloff V5 (µV) 1 -0.28±2.44 4.51 -5.08 Sokoloff V6 (µV) 1 -0.19±2.60 4.90 -5.28 bpm: beats per minute; CI: Confidence Interval; HR: heart rate; ICC: intraclass correlation coefficients, LOA: limits of agreement; SD: standard deviation. *All ICC are significant (p<0.001). Additional Declarations No competing interests reported. Supplementary Files TableS1.docx TableS2.docx Cite Share Download PDF Status: Published Journal Publication published 19 Dec, 2025 Read the published version in Cardio-Oncology → Version 1 posted Editorial decision: Revision requested 10 Sep, 2025 Reviews received at journal 16 Aug, 2025 Reviewers agreed at journal 16 Aug, 2025 Reviewers invited by journal 11 Aug, 2025 Editor assigned by journal 06 Aug, 2025 Submission checks completed at journal 01 Aug, 2025 First submitted to journal 24 Jul, 2025 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-7204236","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":501319843,"identity":"f477bd19-6192-4958-84f5-020bde2cd4af","order_by":0,"name":"Samuel D. COHEN","email":"data:image/png;base64,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","orcid":"","institution":"Sorbonne University, INSERM CIC-1901, Pitié-Salpêtrière Hospital, APHP","correspondingAuthor":true,"prefix":"","firstName":"Samuel","middleName":"D.","lastName":"COHEN","suffix":""},{"id":501319844,"identity":"12c47e67-97a3-4a54-afa3-e23890a5d005","order_by":1,"name":"Maxime ROBERT-HALABI","email":"","orcid":"","institution":"Sorbonne University, INSERM CIC-1901, Pitié-Salpêtrière Hospital, APHP","correspondingAuthor":false,"prefix":"","firstName":"Maxime","middleName":"","lastName":"ROBERT-HALABI","suffix":""},{"id":501319845,"identity":"65ccb2c8-da6a-4964-bbad-a2487c065b74","order_by":2,"name":"Adrien PROCUREUR","email":"","orcid":"","institution":"Sorbonne University, INSERM CIC-1901, Pitié-Salpêtrière Hospital, APHP","correspondingAuthor":false,"prefix":"","firstName":"Adrien","middleName":"","lastName":"PROCUREUR","suffix":""},{"id":501319846,"identity":"1d304d31-d172-4a04-a197-20adac988f51","order_by":3,"name":"Mathieu JAMELOT","email":"","orcid":"","institution":"Institut Universitaire de Cancérologie, Sorbonne University, Tenon Hospital","correspondingAuthor":false,"prefix":"","firstName":"Mathieu","middleName":"","lastName":"JAMELOT","suffix":""},{"id":501319847,"identity":"c34d6335-8bb1-4a82-a6a5-56d680e1c93b","order_by":4,"name":"Martino VAGLIO","email":"","orcid":"","institution":"AMPS LLC","correspondingAuthor":false,"prefix":"","firstName":"Martino","middleName":"","lastName":"VAGLIO","suffix":""},{"id":501319848,"identity":"841da419-771a-45b5-8439-2438e7f1a8c5","order_by":5,"name":"Fabio BADILINI","email":"","orcid":"","institution":"AMPS LLC","correspondingAuthor":false,"prefix":"","firstName":"Fabio","middleName":"","lastName":"BADILINI","suffix":""},{"id":501319849,"identity":"8c3ad3bf-37ba-4baa-8053-2d4fcb779c7f","order_by":6,"name":"Edi PRIFTI","email":"","orcid":"","institution":"IRD, Sorbonne University, UMMISCO, INSERM, NutriOmics","correspondingAuthor":false,"prefix":"","firstName":"Edi","middleName":"","lastName":"PRIFTI","suffix":""},{"id":501319850,"identity":"4400d2fc-dc99-42f9-a99e-b3364a2f3d19","order_by":7,"name":"Joe-Elie SALEM","email":"","orcid":"","institution":"Sorbonne University, INSERM CIC-1901, Pitié-Salpêtrière Hospital, APHP","correspondingAuthor":false,"prefix":"","firstName":"Joe-Elie","middleName":"","lastName":"SALEM","suffix":""}],"badges":[],"createdAt":"2025-07-24 09:53:41","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7204236/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7204236/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s40959-025-00405-7","type":"published","date":"2025-12-19T15:58:07+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89385616,"identity":"42ec89d3-e265-45f9-b797-acf58deec82e","added_by":"auto","created_at":"2025-08-19 12:31:28","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":578199,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlow-chart of the study.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/e723044736b6a1b89bbdfa8a.jpg"},{"id":89385620,"identity":"12b76f57-c619-4aa8-9563-6c2178f42487","added_by":"auto","created_at":"2025-08-19 12:31:28","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1076942,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExample of an ECG Analysis in a severe ICI myocarditis patient using TCM with CalECG 4.2®. \u003c/strong\u003e\u003cem\u003eSuperimposed median beat from 29 QRS with vector magnitude (green). Automatic caliper placements (PR, QRS, QT onset, offset and peaks) with manual editing option. QTc (Fridericia’s heart rate correction) was derived from the averaged RR intervals. The Sokolow-Lyon indices are calculated by measuring the S wave voltage in V1 + the R wave voltage in V5 (Sokolow V5) or V6 (Sokolow V6). The pathological QRS duration (\u0026gt;120 ms) observed herein, reflects the severity of the ICI-myocarditis.\u003c/em\u003e \u003cem\u003eLow QRS voltage is coded as \"YES\" when all frontal leads have a voltage ≤ 500 µV or all precordial leads have a voltage ≤ 1000 µV. Conversely, it is coded as \"NO\" if any frontal lead exceeds 500 µV or any precordial lead exceeds 1000 µV. Soko for Sokolow-Lyon voltage.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/3be77f5ccc78803da860e066.jpg"},{"id":89385627,"identity":"f4705afd-2081-4bb3-ab9f-1d3566632b84","added_by":"auto","created_at":"2025-08-19 12:31:28","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1332949,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBland-Altman Manual Plot for all analyses by cohort. \u003c/strong\u003e\u003cem\u003eThe dashed blue lines represent the limits of agreement (LOA), and the dashed black line indicates the mean bias for the different measurements. For the cohorts, the orange points represent breast cancer patients on ribociclib, while the green points represent immune checkpoint inhibitor (ICI) myocarditis patients.\u003c/em\u003e\u003cstrong\u003e \u003c/strong\u003e\u003cem\u003eSoko for Sokolow-Lyon voltage.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/74ff1dc60fff248b135bd7d4.jpg"},{"id":89389322,"identity":"cd5a73ed-cf70-4fd2-b5ce-56f099147b34","added_by":"auto","created_at":"2025-08-19 12:47:28","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1024919,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eElectrocardiographic changes over time in ribociclib-treated breast cancer patients.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/821e41e97ae937e954cad139.jpg"},{"id":89385618,"identity":"b2aa339f-c1a1-49fb-bf63-e28376a2bd51","added_by":"auto","created_at":"2025-08-19 12:31:28","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1937132,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eElectrocardiographic changes over time in ICI myocarditis patients. \u003c/strong\u003e\u003cem\u003eSoko for Sokolow-Lyon voltage.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"fig5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/90f7d43fdae818e4952aeb53.jpg"},{"id":98814144,"identity":"77380ca3-0510-4d30-98cb-fbce70626249","added_by":"auto","created_at":"2025-12-22 16:11:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6769290,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/e705989f-c96c-463f-b5c7-3f7fcb22326b.pdf"},{"id":89385615,"identity":"5fcf9371-cbc4-408c-b903-82459b1ff8df","added_by":"auto","created_at":"2025-08-19 12:31:28","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":19431,"visible":true,"origin":"","legend":"","description":"","filename":"TableS1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/118aa4e48474c57e8fe30789.docx"},{"id":89385617,"identity":"2666114e-e0c7-4af5-a7c1-cee8a14991ee","added_by":"auto","created_at":"2025-08-19 12:31:28","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":16298,"visible":true,"origin":"","legend":"","description":"","filename":"TableS2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7204236/v1/ce3efa18a2d06c9b8f05e1b0.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Validation of a novel semi-automated ECG quantification tool, applied to a cardio-oncology setting Semi-Automated ECG Tool applied to cardio-oncology","fulltext":[{"header":"Background","content":"\u003cp\u003eThe electrocardiogram (ECG) is an essential tool in clinical and research settings to diagnose and monitor cardiac diseases. Cardio-oncology has emerged as a new field specializing in the surveillance and treatment of cardiovascular toxicities caused by various cancer therapies.(\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e–\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) Such treatments include cyclin-dependent kinase inhibitors, like ribociclib, which may lead to life-threatening \u003cem\u003etorsades de pointes\u003c/em\u003e by prolonging the heart rate corrected QT interval (QTc).(\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e–\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e) On the other hand, immune checkpoint inhibitors (ICI) can induce severe myocarditis, a rare but potentially fatal condition.(\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e–\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e) ECG monitoring of such patients revealed several conduction abnormalities, such as PR and QRS prolongation as well as voltage changes.(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e) The latter two have been associated with worse outcomes in ICI-myocarditis, as they likely indicate myocardial edema and/or damage from macrophage infiltration.(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eA thorough monitoring of these ECG parameters is thus indicated but may be challenging in clinical practice. While manual measurement methods are widely used, they are time-consuming and prone to inter-reader, intra-reader and beat-to-beat variability, which limits their reliability.(\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e–\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e) For instance, QTc measurements can vary due to differences in baseline selection and correction formulas applied, leading to inconsistencies that limit the diagnostic accuracy.(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e) Similarly, since the assessment of Sokolow voltage is highly dependent on the precision of manual tracing measurements, it is intrinsically associated to a high degree of variability and poor reliability.(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eSoftware using automated methods relying on mathematical concepts, such as CalECG, offer a promising alternative by providing consistent and precise measurements, particularly in complex clinical settings (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Notwithstanding these improvements, accurate automated ECG measurements may still be hindered by signal noise and artifacts. To address these issues and further enhance the signal-to-noise ratio, triplicate ECG of 10sec and superimposed median beats have been implemented (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). Semi-automated methods, such as the triplicate assessment method (TAM), are widely adopted. TAM consists of repeating measurements three times (individual analysis of 10sec ECG within a triplicate series) to ensure accuracy and reliability, which, although more precise, can be labor-intensive. The triplicate concatenation method (TCM) has been introduced as a novel and more efficient technique that build a 30-second ECG by concatenating three 10-second tracing while excluding artifacts at the concatenation points.(\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e) Although promising, this novel method remains to be validated in large real-life patient populations, particularly for automated QRS voltage and Sokolow calculations, which were previously unavailable in readily available ECG analysis software.\u003c/p\u003e\u003cp\u003eBuilding on prior research validating the CalECG 3.7 software® for QTc measurement (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), this study aims to evaluate the reproducibility of TCM using the updated CalECG 4.2 software®. This updated version integrates a more precise analysis of the QRS segments and its voltage, enabling a semi-automated calculation of the Sokolow-Lyon voltage, a feature previously unavailable. By assessing intra- and inter-reader variabilities as well as characterizing ECG changes over time, we seek to determine whether TCM offers a reliable alternative for ECG interval measurements in cardio-oncology patients. Specifically, we focus on patients treated with ribociclib or followed for immune checkpoint inhibitor (ICI)-induced myocarditis, since these patient populations exhibit greater ECG variability, thereby providing a robust test for this novel method.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003eStudy Population\u003c/em\u003e\u003c/p\u003e\u003cp\u003eWe recruited thirty-one patients followed in a French quaternary care University-affiliated cardio-oncology center (Pitié-Salpêtrière Hospital, Paris) and enrolled in NEOCARDIO (\u003cem\u003eNCT03882580\u003c/em\u003e, approved by the Ethics Committee of Sorbonne University on March 12th, 2021), an observational, ambispective cohort. They were first seen between December 2022 and July 2024 and then evaluated prospectively. On inclusion, each patient had 3 resting 12-lead 10-sec ECG, each separated by 2-minute intervals. Patients were placed in supine position and asked to remain immobile during the ECG acquisition to minimize physiological variability and artefacts. Included patients were required to have high-quality non-electro stimulated ECG in sinus rhythm. Included patients were divided into two groups:\u003c/p\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eBreast cancer patients\u003c/b\u003e: Twenty-one patients undergoing their first cycle of ribociclib. Each patient is followed for three visits: at day 0 (baseline), at days 14 ± 3 and 28 ± 3 (post-treatment). Ribociclib was administered on a 28-day cycle, with 21 days on-treatment followed by 7 days off-treatment, as part of the standard treatment regimen. Our main focus was to evaluate the treatment-induced effect on the QTc interval, as ribociclib is known to prolong it (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cb\u003eICI myocarditis patients\u003c/b\u003e: Ten patients with severe ICI-induced myocarditis were enrolled. Each patient had 8 planned visits within the first two months of their initial admission. The immunosuppressive regimen consisted of abatacept administered by intravenous injection with concomitant oral ruxolitinib and corticosteroids. Their individual dosing and administration regimen have previously been published (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). The main focus was to study the evolution of the QRS voltage and duration while on treatment, as these parameters have been shown useful to assess myocardial injury and inflammation and may represent a surrogate marker for disease severity and treatment response (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e).\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003cp\u003eIn this ECG software validation study, six key ECG parameters were analyzed: heart rate (HR), PR interval, QRS duration, corrected QT interval for HR using Fridericia's formula (QTc) (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e), and Sokolow-Lyon voltage, calculated using the formula: (S wave maximal voltage in leadV1) + (R wave maximal voltage in lead V5 or lead V6).(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e)\u003c/p\u003e\u003cp\u003e\u003cem\u003eElectrocardiogram Recording and Analysis\u003c/em\u003e\u003c/p\u003e\u003cp\u003eECG were recorded using a standardized digital electrocardiograph (ELI 280, V2.4.1.8; Mortara Instrument, Inc., Milwaukee, WI, USA) by trained nurses. The device has a sampling rate of 1000 Hz and a filter of 150 Hz, which were selected to capture detailed signal characteristics for subsequent analysis. In total, 420 ECG (140 triplicate sets) were analyzed \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e for detailed flow-chart\u003cb\u003e)\u003c/b\u003e. Quantitative ECG analysis was performed using the TAM and the TCM, both being semi-automated approaches. Methods variability were assessed by comparing inter-method variability (TAM vs TCM#1 reader B) and intra-method variability (TCM#1 vs TCM#2 reader B). TCM#1 corresponds to the first analysis of the triplicate, while TCM#2 refers to the second analysis. Inter-reader variability for TCM was determined by comparing results between reader A and reader B (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Reader A (JES), a Cardiologist, has over 15 years of experience in ECG analysis, while reader B (SDC), a PharmD, has more than 3 years of experience in ECG analysis. Both readers independently reviewed and analyzed the ECG. They remained blinded to each other's measurements and the study hypotheses throughout the process.\u003c/p\u003e\u003cp\u003e\u003cem\u003eCalECG 4.2 Software® Overview\u003c/em\u003e\u003c/p\u003e\u003cp\u003eAll ECG measurements were conducted using the latest version of the CalECG 4.2 software®, which automatically generates a representative beat for each of the 12 leads derived from detected sinus rhythm beats. This process involves aligning sinus rhythm ECG beats on the R-wave’s peak for each lead and creating a median beat by computing a median value for each ECG sample, producing a single representative signal, 1.2s long, per lead. As a result, the representative beat is not an actual ECG recorded beat but rather an averaged signal, synthesizing the characteristics of all recorded sinus rhythm beats in each lead. To build this representative beat, the software automatically eliminates all premature ventricular contractions within a 10-second ECG recording. The final output includes 12 representative beats, one for each lead. By superimposing these single-lead representative beats, a superimposed median beat is generated. This superimposed median beat is optimally represented by a vector of magnitude, calculated as the square root of the sum of the squared representative beats. The vector of magnitude facilitates automated measurements of PR, QT and QRS intervals and voltages using a threshold-based method. If automatic PR/QT/QRS fiducial marks are misplaced, users retain the ability to manually adjust the fiducial markers for critical points, including the onset of the P-wave, the onset and offset of the QRS complex or the offset of the T-wave. Additional ECG parameters of interest, particularly QRS voltages can be measured using the same process, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Compared with the previous version (CalECG 3.7 software®) (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), the main improvement of the 4.2 version is a comprehensive revision of the AMPS Bravo algorithm. This update specifically targets waveforms within the QRS complex (Q, R, S, R’, S’), which were previously excluded from automated analysis and could only be measured manually in the 3.7 version. The addition of automated QRS voltage analysis, integrating Sokolow-Lyon and microvoltage assessments, enhances measurement precision and reliability.\u003c/p\u003e\u003cp\u003e\u003cem\u003eStatistical Analyses\u003c/em\u003e\u003c/p\u003e\u003cp\u003eBland-Altman plots and intra-class correlation coefficients (ICC) were used to evaluate the measurements’ reliability between the two readers/readings.(\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e) Bland-Altman plots graphically represent the agreement between the two readers/readings, plotting the mean difference (bias) and the limits of agreement (LOA) (mean difference ± 1.96 standard deviation [SD]) against the average of the two measurements. Additionally, nonlinear mixed-effects modelling (NLME) was applied to analyze the influence of random effects (i.e., patient’s variability) and fixed effects (e.g., measurement method, or reader) on ECG parameters. All statistical analyses were performed using R software (version 4.4.2). A p-value of \u0026lt; 0.05 was deemed significant.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe ICC and Bland-Altman analysis (including LOA) for the method/readings/readers comparisons are summarized in \u003cstrong\u003eTable 1\u0026nbsp;\u003c/strong\u003eand\u003cstrong\u003e\u0026nbsp;Figure 3\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eMeasurements variability with CalECG4.2\u0026reg; \u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eInter-reader variability for TCM measurements demonstrated excellent consistency across measurements for all ECG parameters (ICC\u0026gt;0.994, see \u003cstrong\u003eTable 1\u003c/strong\u003e). For QTc, the mean bias was 0.02\u0026plusmn;1.32ms (LOA: -2.57 to 2.61 ms). The PR interval exhibited a mean bias of 0.47\u0026plusmn;2.13ms (LOA:-3.70 to 4.65ms). The QRS duration showed a mean bias of 1.08\u0026plusmn;2.77ms (LOA: -4.35 to 6.51ms). The Sokolow-Lyon voltage using lead V5 demonstrated a mean bias of 0.40\u0026plusmn;4.78 \u0026mu;V (LOA:-8.96 to 9.76\u0026micro;V).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInter-method variability between TAM and TCM demonstrated high consistency across measurements for all ECG parameters (ICC \u0026gt;0.989, see \u003cstrong\u003eTable 1\u003c/strong\u003e). For QTc, the mean bias was -1.17\u0026plusmn;3.12ms (LOA: -7.28 to 4.93ms). The PR interval exhibited a mean bias of 0.53\u0026plusmn;3.02ms (LOA: -5.39 to 6.44ms). The QRS duration showed a mean bias of 0.24 \u0026plusmn; 2.17ms (LOA: -4.01 to 4.50ms). The Sokolow-Lyon voltage using lead V5 demonstrated a mean bias of 10.35\u0026plusmn;24.10\u0026mu;V (LOA: -36.88 to 57.58\u0026micro;V).\u003c/p\u003e\n\u003cp\u003eIntra-method variability within TCM demonstrated excellent repeatability across measurements for all ECG parameters (ICC \u0026gt; 0.991, see \u003cstrong\u003eTable 1\u003c/strong\u003e). For QTc, the mean bias was -0.31\u0026plusmn;1.78ms (LOA: -3.80 to 3.19ms). The QRS duration showed a mean bias of -1.17\u0026plusmn;3.23ms (LOA: -7.50 to 5.15ms). The Sokolow-Lyon voltage using lead V5 demonstrated a mean bias of -0.28\u0026plusmn;2.44\u0026mu;V (LOA: -5.08 to 4.51\u0026micro;V).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eApplication in real life cohort: ribociclib-treated breast cancer patients and ICI myocarditis patients\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn ribociclib-treated breast cancer patients, ECG parameters exhibited significant time-dependent changes \u003cstrong\u003e(Figure 4)\u003c/strong\u003e. NLME models were adjusted for visit type (D0, D14\u0026plusmn;3, D28\u0026plusmn;3), methods (TCM vs. TAM), readers (A vs. B), readings (reading #1 vs. #2), and age. The variables \u0026quot;methods\u0026quot;, \u0026quot;readers\u0026quot;, and \u0026quot;readings\u0026quot; did not show significant effects \u003cstrong\u003e(Table S1).\u003c/strong\u003e By day 14 (vs. D0), HR decreased slightly (-3\u0026plusmn;1 bpm, p\u0026lt;0.001), QTc increased notably (16\u0026plusmn;1ms, p\u0026lt;0.001), QRS duration shortened significantly (-1.5\u0026plusmn;0.5 ms, p=0.001), and the Sokolow voltage using lead V6 rose significantly (40\u0026plusmn;19 \u0026micro;V, p\u0026lt;0.05), with similar trends using lead V5. By day 28 (vs. D0), QTc decreased (-6\u0026plusmn;1ms, p\u0026lt;0.001), returning below baseline values, while the PR interval showed a modest increase (3\u0026plusmn;1ms, p=0.006). Both HR, Sokolow voltage and QRS duration also returned to baseline levels (see\u003cstrong\u003e\u0026nbsp;Figure 4\u0026nbsp;\u003c/strong\u003eand\u003cstrong\u003e\u0026nbsp;Table S1\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eIn severe ICI myocarditis patients treated by abatacept, ruxolitinib and corticosteroids, ECG parameters exhibited distinct temporal patterns \u003cstrong\u003e(Figure 5)\u003c/strong\u003e. NLME models were adjusted for visit type (D0, and 7 other visits between D5\u0026plusmn;3 to D60\u0026plusmn;5), methods (TCM vs. TAM), readers (A vs. B), readings (reading #1 vs. #2), age, and sex. The variables \u0026quot;methods\u0026quot;, \u0026quot;readers\u0026quot;, and \u0026quot;readings\u0026quot; did not show significant effects \u003cstrong\u003e(Table S2)\u003c/strong\u003e. By day 5 (vs. D0), HR declined significantly (-11\u0026plusmn;2 bpm, p\u0026lt;0.001) and QRS duration increased (6\u0026plusmn;1ms, p\u0026lt;0.001). The HR reduction persisted through day 60 (-12\u0026plusmn;3 bpm, p\u0026lt;0.001). The PR interval showed modest reductions at day 14 (-6\u0026plusmn;2 ms, p\u0026lt;0.001) and day 50 (-5\u0026plusmn;2ms, p=0.001). From day 21 through day 60, QTc decreased significantly, with the largest reduction observed by day 28 (-26\u0026plusmn;3ms, p\u0026lt;0.001), and smaller yet significant decreases persisted through day 60 (-10\u0026plusmn;3 ms, p\u0026lt;0.001). From day 14 through day 60, QRS duration returned to baseline values, while the Sokolow-Lyon voltage using lead V5 increased significantly (442 \u0026plusmn; 48 \u0026micro;V, p\u0026lt;0.001), with the maximum increase observed by day 28 (458 \u0026plusmn; 49 \u0026micro;V, p\u0026lt;0.001). This elevation persisted throughout the follow-up period (\u003cstrong\u003esee Figure 5\u0026nbsp;\u003c/strong\u003eand\u003cstrong\u003e\u0026nbsp;Table S2\u003c/strong\u003e).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study evaluated the inter- and intra-reader variability of TCM for ECG measurements and compared it to the traditional TAM across critical ECG intervals in two distinct real-life standardized cardio-oncology patient populations. Our findings demonstrate that TCM provides high reproducibility for key ECG parameters, including HR, PR interval, QRS duration, QTc, and Sokolow voltages using V5 and V6, with minimal bias and narrow limits of agreement. These results underscore the robustness and reliability of TCM, as they align with previous studies reporting comparable LOA values for key ECG parameters, such as QTc measured in dedicated Thorough QT trials (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan additionalcitationids=\"CR29\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). In practice, CalECG4.2\u0026reg; is even more accurate, due to its semi-automated QRS onset analysis, which improved upon the previous version that required manual QRS positioning (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSemi-automated techniques, such as the TCM, offer superior precision compared to traditional manual measurements, with a reported inter-reader variability for QTc measurements ranging in terms of SD from 6ms for semi-automated methods to 15ms for manual methods (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). In comparison, our study showed intra and inter-reader variability below 3ms, demonstrating a level of precision that aligns with the highest reported accuracies. Such accuracy level of QTc interval measurement is crucial to easily identify drug-induced QT prolongation in real-life setting, and moreover to decrease the numbers of required subjects needed to identify QT-prolongation induced by drugs in clinical trials.(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eThe CALECG 4.2\u0026reg; system, which relies on classical mathematical approaches and vector analysis, has shown to be more precise than deep learning models for this specific task (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). Deep learning models, while powerful, initially learn from human measurements that inherently carries variability. For instance, Bos et al. demonstrated that deep learning models can effectively identify concealed long QT syndrome from ECG, but their accuracy was inherently limited by the quality and consistency of the training data, with reported measurement errors up to 10 ms.(\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e) Similarly, Giudicessi et al. highlighted the potential of AI-enabled ECG analysis but also reported variability in QTc measurements standard deviations ranging from 5 to 15 ms.(\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e) In contrast, classical methods like those employed in CalECG 4.2 currently offer superior precision for straightforward QTc measurements since it relies on deterministic algorithms that are less susceptible to human measurement errors. Therefore, while deep learning holds promise for complex tasks,(\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e) its applicability to QTc measurement may need further refinement to match the precision of classical methods.\u003c/p\u003e\u003cp\u003eThe integration of NLME modelling in this study provided a detailed assessment of time-dependent variations in ECG parameters in response to cardiotoxic treatments. Our results highlight a significant temporal effect of ribociclib on QTc, PR, QRS, and Sokolow-Lyon indices, underscoring the dynamic nature of treatment-induced cardiotoxicity. QTc significantly increased by day 14 and shortened back to baseline by day 28, reflecting the transient impact of ribociclib on ventricular repolarization. This pattern aligns with ribociclib's administration schedule, which involves a 21-day treatment cycle followed by a 7-day break.(\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e) Given its plasma half-life ranging from 29.7 to 54.7 hours, higher circulating drug levels are expected during the treatment phase, contributing to the observed ECG changes. Additionally, the PR interval prolongation observed by day 28 indicates a possible atrioventricular conduction delay induced by ribociclib, independent of concentration, unlike its well-documented QT prolonging effect (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e). PR interval prolongation may be less reversible and warrants further evaluation, particularly regarding its potential impact on clinical conduction disturbances. Our study uniquely expands the understanding of ribociclib\u0026rsquo;s electrophysiological effects by highlighting its possible novel influence on the PR interval, which has not been previously emphasized. However, this relationship requires further clinical evaluations to be fully understood. We also observed a significant increase in Sokolow-Lyon voltage by day 14 on ribociclib. By day 28, the voltage returned near its baseline level. The significance of these changes is unclear, but may suggest a broader electrophysiological impact of ribociclib beyond the known QTc prolongation. This highlights the need for comprehensive assessments in future studies to fully understand its effects on cardiac electrophysiology.\u003c/p\u003e\u003cp\u003eHistorically, Sokolow-Lyon voltages have been used to assess left ventricular hypertrophy, although their usefulness has been limited by variability in manual measurements.(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e) Nevertheless, when compared to cardiac magnetic resonance imagining or echocardiography, which provide more precise measurements of left ventricular hypertrophy, Sokolow-Lyon voltages often show some level of concordance in identifying significant hypertrophy.(\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e) Recent studies suggest that Sokolow-Lyon voltages may also be a marker of proper left intraventricular conduction as it represent the summation of action potentials occurring simultaneously.(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e) In cases of myocardial toxicity, particularly those mediated by lymphocytes and macrophages inflammation, the conduction can become disorganized, leading to a lower widespread QRS voltage. This phenomenon is observed in ICI-myocarditis, where the initial disruption in conduction results in a low Sokolow-Lyon voltage.(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e) The identification of its improvement under immunosuppressive treatment in this work is novel and highlights its potential as a recovery marker. This perspective shifts the focus from hypertrophy to a broader assessment of electrophysiological integrity.\u003c/p\u003e\u003cp\u003e\u003cem\u003eLimitations\u003c/em\u003e\u003c/p\u003e\u003cp\u003eThis study has several limitations. Since our analysis focused on two specific patient cohorts, it may not fully capture the variability present in other clinical settings. Variability tends to be lower among healthy volunteers without underlying pathology compared to patients with cardiac conditions, as the latter group presents more complex and variable ECG patterns. This study specifically focuses on patients with cardiac conditions to validate the method's robustness in a highly variable context. Additionally, although ECG analysis is widely used in clinical practice, it is important to acknowledge that the variability in measurements can be influenced by the quality of the ECG waveform. The near-perfect QRS complexes and overall waveform obtained in this study may contribute to the narrow variability, which might not be replicated in all clinical care settings where non-standardized ECG may be performed by less trained workers. Additionally, this study primarily serves as a validation for our ECG measurement method, and we were unable to integrate comprehensive clinical data and concomitant treatments that could potentially alter ECG parameters beyond the few demographic variables considered in the pathological settings used herein. Future research should aim to incorporate these factors to provide a more complete evaluation.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eTCM demonstrates high reproducibility and minimal variability, establishing itself as a reliable alternative to TAM for ECG interval measurements. Its ability to capture time-dependent changes in ECG parameters, such as QTc, PR, QRS, and Sokolow-Lyon voltages, highlights its potential for monitoring drug-induced cardiotoxicity in cardio-oncology.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eECG\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eElectrocardiogram\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eHR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eHeart rate\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eICC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eIntraclass correlation coefficient\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eICI\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eImmune checkpoint inhibitor\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLOA\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eLimits of agreement\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNLME\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNonlinear mixed-effects modelling\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ePR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003ePR interval\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eQTc\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eCorrected QT interval (using Fridericia\u0026rsquo;s formula)\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eQRS\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eQRS complex duration\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSD\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eStandard deviation\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eTAM (Triplicate assessment method)\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eA standard method where three separate 10 seconds ECGs are recorded and averaged to reduce beat-to-beat variability\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eTCM (Triplicate concatenation method)\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSemi-automated method where three 10 seconds ECG tracings are merged into a single waveform\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eConflict of interest statement\u003c/span\u003e:\u003c/h2\u003e\u003cp\u003eSC, M.R-H, AP, MJ, MV, FB and EP have nothing to disclose. J.-E.S. has received financial support from Novartis, BeiGene, BMS and Banook Group; and holds patents related to the prognostication and treatment of ICI myocarditis.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis study was supported by the ANR-20-CE17-0022 DeepECG4U funding from the French National Research Agency (ANR), by CIC-1901.This work was supported by the Sorbonne Center for Artificial Intelligence (SCAI), an institute dedicated to artificial intelligence within the Sorbonne University Alliance, funded by the excellence initiative IDEX SUPER under the ANR reference: 11-IDEX-0004.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eSamuel CohenData curation-Equal, Formal analysis-Lead, Investigation-Equal, Software-Equal, Writing - original draft-EqualMaxime robert-halabiData curation-Supporting, Writing - original draft-SupportingAdrien ProcureurData curation-Equal, Investigation-Equal, Software-Equal, Writing - review \u0026amp; editing-EqualMathieu JAMELOTData curation-Supporting, Investigation-Supporting, Writing - review \u0026amp; editing-SupportingMartino VaglioProject administration-Supporting, Software-Lead, Validation-Supporting, Writing - review \u0026amp; editing-EqualFabio BadiliniCRediT contribution not specifiedEdi PriftiConceptualization-Supporting, Methodology-Equal, Supervision-Equal, Validation-Equal, Visualization-Lead, Writing - review \u0026amp; editing-EqualJoe-Elie SalemConceptualization-Lead, Formal analysis-Supporting, Funding acquisition-Lead, Investigation-Lead, Methodology-Lead, Project administration-Lead, Resources-Lead, Software-Lead, Supervision-Lead, Validation-Lead, Visualization-Equal, Writing - original draft-Lead\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors and study sponsor are indebted to the patients and their families, as well as the investigators participating in this study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data supporting the findings of this study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMorelli MB, Bongiovanni C, Da Pra S, Miano C, Sacchi F, Lauriola M, et al. Cardiotoxicity of Anticancer Drugs: Molecular Mechanisms and Strategies for Cardioprotection. Front Cardiovasc Med. 2022;9:847012.\u003c/li\u003e\n\u003cli\u003eNguyen LS, Dolladille C, Drici M-D, Fenioux C, Alexandre J, Mira J-P, et al. Cardiovascular Toxicities Associated With Hydroxychloroquine and Azithromycin: An Analysis of the World Health Organization Pharmacovigilance Database. Circulation. 2020;142(3):303-5.\u003c/li\u003e\n\u003cli\u003eSalem J-E, Manouchehri A, Moey M, Lebrun-Vignes B, Bastarache L, Pariente A, et al. Cardiovascular toxicities associated with immune checkpoint inhibitors: an observational, retrospective, pharmacovigilance study. Lancet Oncol. 2018;19(12):1579-89.\u003c/li\u003e\n\u003cli\u003eSalem J-E, Nguyen LS, Moslehi JJ, Ederhy S, Lebrun-Vignes B, Roden DM, et al. Anticancer drug-induced life-threatening ventricular arrhythmias: a World Health Organization pharmacovigilance study. European Heart Journal. 2021;42(38):3915-28.\u003c/li\u003e\n\u003cli\u003eAlexandre J, Moslehi JJ, Bersell KR, Funck-Brentano C, Roden DM, Salem JE. Anticancer drug-induced cardiac rhythm disorders: Current knowledge and basic underlying mechanisms. Pharmacol Ther. 2018;189:89-103.\u003c/li\u003e\n\u003cli\u003eSalem JE, Nguyen LS, Moslehi JJ, Ederhy S, Lebrun-Vignes B, Roden DM, et al. Anticancer drug-induced life-threatening ventricular arrhythmias: a World Health Organization pharmacovigilance study. Eur Heart J. 2021;42(38):3915-28.\u003c/li\u003e\n\u003cli\u003ePostema PG, Wilde AAM. The Measurement of the QT Interval. Curr Cardiol Rev. 2014;10(3):287-94.\u003c/li\u003e\n\u003cli\u003eJohnson DB, Balko JM, Compton ML, Chalkias S, Gorham J, Xu Y, et al. Fulminant Myocarditis with Combination Immune Checkpoint Blockade. N Engl J Med. 2016;375(18):1749-55.\u003c/li\u003e\n\u003cli\u003eFenioux C, Abbar B, Boussouar S, Bretagne M, Power JR, Moslehi JJ, et al. Thymus alterations and susceptibility to immune checkpoint inhibitor myocarditis. Nat Med. 2023;29(12):3100-10.\u003c/li\u003e\n\u003cli\u003eLehmann LH, Heckmann MB, Bailly G, Finke D, Procureur A, Power JR, et al. Cardiomuscular Biomarkers in the Diagnosis and Prognostication of Immune Checkpoint Inhibitor Myocarditis. Circulation. 2023;148(6):473-86.\u003c/li\u003e\n\u003cli\u003eSalem JE, Bretagne M, Abbar B, Leonard-Louis S, Ederhy S, Redheuil A, et al. Abatacept/Ruxolitinib and Screening for Concomitant Respiratory Muscle Failure to Mitigate Fatality of Immune-Checkpoint Inhibitor Myocarditis. Cancer Discov. 2023;13(5):1100-15.\u003c/li\u003e\n\u003cli\u003ePower JR, Alexandre J, Choudhary A, Ozbay B, Hayek S, Asnani A, et al. Electrocardiographic Manifestations of Immune Checkpoint Inhibitor Myocarditis. Circulation. 2021;144(18):1521-3.\u003c/li\u003e\n\u003cli\u003ePower JR, Alexandre J, Choudhary A, Ozbay B, Hayek SS, Asnani A, et al. Association of early electrical changes with cardiovascular outcomes in immune checkpoint inhibitor myocarditis. Arch Cardiovasc Dis. 2022;115(5):315-30.\u003c/li\u003e\n\u003cli\u003eSaqu\u0026eacute; V, Vaglio M, Funck-Brentano C, Kilani M, Bourron O, Hartemann A, et al. Fast, accurate and easy-to-teach QT interval assessment: The triplicate concatenation method. Archives of Cardiovascular Diseases. 2017;110(8-9):475-81.\u003c/li\u003e\n\u003cli\u003eMalik M, Hnatkova K, Schmidt A, Smetana P. Accurately measured and properly heart-rate corrected QTc intervals show little daytime variability. Heart Rhythm. 2008;5(10):1424-31.\u003c/li\u003e\n\u003cli\u003eAzie NE, Adams G, Darpo B, Francom SF, Polasek EC, Wisser JM, et al. Comparing methods of measurement for detecting drug-induced changes in the QT interval: implications for thoroughly conducted ECG studies. Annals of Noninvasive Electrocardiology: The Official Journal of the International Society for Holter and Noninvasive Electrocardiology, Inc. 2004;9(2):166-74.\u003c/li\u003e\n\u003cli\u003eFarb A, Devereux RB, Kligfield P. Day-to-day variability of voltage measurements used in electrocardiographic criteria for left ventricular hypertrophy. J Am Coll Cardiol. 1990;15(3):618-23.\u003c/li\u003e\n\u003cli\u003eSchijvenaars BJA, van Herpen G, Kors JA. Intraindividual variability in electrocardiograms. J Electrocardiol. 2008;41(3):190-6.\u003c/li\u003e\n\u003cli\u003eNatekar M, Hingorani P, Gupta P, Karnad DR, Kothari S, de Vries M, et al. Effect of number of replicate electrocardiograms recorded at each time point in a thorough QT study on sample size and study cost. J Clin Pharmacol. 2011;51(6):908-14.\u003c/li\u003e\n\u003cli\u003eHingorani P, Karnad DR, Ramasamy A, Panicker GK, Salvi V, Bhoir H, et al. Semiautomated QT interval measurement in electrocardiograms from a thorough QT study: comparison of the grouped and ungrouped superimposed median beat methods. J Electrocardiol. 2012;45(3):225-30.\u003c/li\u003e\n\u003cli\u003eRichardson DR, Parish PC, Tan X, Fabricio J, Andreini CL, Hicks CH, et al. Association of QTc Formula With the Clinical Management of Patients With Cancer. JAMA Oncol. 2022;8(11):1616-23.\u003c/li\u003e\n\u003cli\u003eSalem J-E, Bretagne M, Abbar B, Leonard-Louis S, Ederhy S, Redheuil A, et al. Abatacept/Ruxolitinib and Screening for Concomitant Respiratory Muscle Failure to Mitigate Fatality of Immune-Checkpoint Inhibitor Myocarditis. Cancer Discov. 2023;13(5):1100-15.\u003c/li\u003e\n\u003cli\u003eSalem J-E, Ederhy S, Belin L, Zahr N, Tubach F, Procureur A, et al. Abatacept dose-finding phase II triaL for immune checkpoint inhibitors myocarditis (ACHLYS) trial design. Archives of Cardiovascular Diseases. 2025;118(2):106-15.\u003c/li\u003e\n\u003cli\u003ePower JR, Alexandre J, Choudhary A, Ozbay B, Hayek SS, Asnani A, et al. Association of early electrical changes with cardiovascular outcomes in immune checkpoint inhibitor myocarditis. Archives of Cardiovascular Diseases. 2022;115(5):315-30.\u003c/li\u003e\n\u003cli\u003eFood, Drug Administration HHS. International Conference on Harmonisation; guidance on E14 Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic Drugs; availability. Notice. Fed Regist. 2005;70(202):61134-5.\u003c/li\u003e\n\u003cli\u003eSokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as obtained by unipolar precordial and limb leads. Am Heart J. 1949;37(2):161-86.\u003c/li\u003e\n\u003cli\u003eBland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307-10.\u003c/li\u003e\n\u003cli\u003eKoo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15(2):155-63.\u003c/li\u003e\n\u003cli\u003eSarapa N, Gussak I, Vajdic B, George S, Hadzievski L, Francom SF, et al. Comparison of QTinno, a fully automated electrocardiographic analysis program, to semiautomated electrocardiographic analysis methods in a drug safety study in healthy subjects. J Electrocardiol. 2009;42(4):358-66.\u003c/li\u003e\n\u003cli\u003eTyl B, Kabbaj M, Fassi B, De Jode P, Wheeler W. Comparison of semiautomated and fully automated methods for QT measurement during a thorough QT/QTc study: variability and sample size considerations. J Clin Pharmacol. 2009;49(8):905-15.\u003c/li\u003e\n\u003cli\u003eBarbey JT, Connolly M, Beaty B, Krantz MJ. Man versus Machine: Comparison of Automated and Manual Methodologies for Measuring the QTc Interval: A Prospective Study. Annals of Noninvasive Electrocardiology : The Official Journal of the International Society for Holter and Noninvasive Electrocardiology, Inc. 2015;21(1):82-90.\u003c/li\u003e\n\u003cli\u003eDarpo B, Agin M, Kazierad DJ, Layton G, Muirhead G, Gray P, et al. Man versus machine: is there an optimal method for QT measurements in thorough QT studies? J Clin Pharmacol. 2006;46(6):598-612.\u003c/li\u003e\n\u003cli\u003eBos JM, Attia ZI, Albert DE, Noseworthy PA, Friedman PA, Ackerman MJ. Use of Artificial Intelligence and Deep Neural Networks in Evaluation of Patients With Electrocardiographically Concealed Long QT Syndrome From the Surface 12-Lead Electrocardiogram. JAMA Cardiol. 2021;6(5):532.\u003c/li\u003e\n\u003cli\u003eGiudicessi JR, Schram M, Bos JM, Galloway CD, Shreibati JB, Johnson PW, et al. Artificial Intelligence-Enabled Assessment of the Heart Rate Corrected QT Interval Using a Mobile Electrocardiogram Device. Circulation. 2021;143(13):1274-86.\u003c/li\u003e\n\u003cli\u003ePrifti E, Fall A, Davogustto G, Pulini A, Denjoy I, Funck-Brentano C, et al. Deep learning analysis of electrocardiogram for risk prediction of drug-induced arrhythmias and diagnosis of long QT syndrome. Eur Heart J. 2021;42(38):3948-61.\u003c/li\u003e\n\u003cli\u003eTripathy D, Im SA, Colleoni M, Franke F, Bardia A, Harbeck N, et al. Ribociclib plus endocrine therapy for premenopausal women with hormone-receptor-positive, advanced breast cancer (MONALEESA-7): a randomised phase 3 trial. Lancet Oncol. 2018;19(7):904-15.\u003c/li\u003e\n\u003cli\u003eBarber M, Nguyen LS, Wassermann J, Spano JP, Funck-Brentano C, Salem JE. Cardiac arrhythmia considerations of hormone cancer therapies. Cardiovasc Res. 2019;115(5):878-94.\u003c/li\u003e\n\u003cli\u003eMetcalfe JZ, Economou T, Naufal F, Kucukosmanoglu M, Kleiman R, Phillips PPJ, et al. Validation of a Handheld 6-Lead Device for QT Interval Monitoring in Resource-Limited Settings. JAMA network open. 2024;7(6):e2415576.\u003c/li\u003e\n\u003cli\u003eAntikainen RL, Grodzicki T, Palmer AJ, Beevers DG, Webster J, Bulpitt CJ, et al. Left ventricular hypertrophy determined by Sokolow-Lyon criteria: a different predictor in women than in men? J Hum Hypertens. 2006;20(6):451-9.\u003c/li\u003e\n\u003cli\u003ePower JR, Dolladille C, Ozbay B, Procureur AM, Ederhy S, Palaskas NL, et al. Predictors and Risk Score for Immune Checkpoint-Inhibitor-Associated Myocarditis Severity. medRxiv. 2024:2024.06.02.24308336.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table 1","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSummary of ICC and Bland-Altman Analysis for Method Comparisons.\u0026nbsp;\u003c/strong\u003e\u003cem\u003eThis table displays ICC and Bland-Altman analysis results for three method comparisons assessing specific ECG parameters, including HR, PR, QRS, QTcF and Sokoloff indices (V5 and V6).\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"444\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eParameters\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eICC*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean\u0026plusmn;SD Bias\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUpper LOA (95% CI)\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLower LOA (95% CI)\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 100%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTCM 1 reader A vs. TCM 1 reader B (n=140)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eHR (bpm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e\u0026nbsp;0\u0026plusmn;0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003ePR (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.995\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e0.47\u0026plusmn;2.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e4.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-3.70\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eQRS (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.994\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e1.08\u0026plusmn;2.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e6.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-4.35\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eQTcF (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e0.02\u0026plusmn;1.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e2.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-2.57\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eSokoloff V5 (\u0026micro;V)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e0.40\u0026plusmn;4.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e9.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-8.96\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eSokoloff V6 (\u0026micro;V)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e0.42\u0026plusmn;5.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e10.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-9.60\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 100%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTAM 1 reader B vs. TCM 1 reader B (n=140)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eHR (bpm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e-0.08\u0026plusmn;0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-1.16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003ePR (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.989\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e0.53\u0026plusmn;3.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e6.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-5.39\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eQRS (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.997\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e0.24\u0026plusmn;2.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e4.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-4.01\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eQTcF (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.992\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e-1.17\u0026plusmn;3.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e4.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-7.28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eSokoloff V5 (\u0026micro;V)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e10.35\u0026plusmn;24.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e57.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-36.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eSokoloff V6 (\u0026micro;V)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e10.11\u0026plusmn;23.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e55.20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-34.98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 100%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTCM 1 reader B vs. TCM 2 reader B (n=140)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eHR (bpm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e0\u0026plusmn;0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003ePR (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.993\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e-0.61\u0026plusmn;2.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e3.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-5.16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eQRS (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.991\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e-1.17\u0026plusmn;3.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e5.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-7.50\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eQTcF (ms)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e0.998\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e-0.31\u0026plusmn;1.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e3.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-3.80\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eSokoloff V5 (\u0026micro;V)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e-0.28\u0026plusmn;2.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e4.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-5.08\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 25.2252%;\"\u003e\n \u003cp\u003eSokoloff V6 (\u0026micro;V)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3604%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 18.4685%;\"\u003e\n \u003cp\u003e-0.19\u0026plusmn;2.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.8468%;\"\u003e\n \u003cp\u003e4.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 24.0991%;\"\u003e\n \u003cp\u003e-5.28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cem\u003ebpm: beats per minute; CI: Confidence Interval; HR: heart rate; ICC: intraclass correlation coefficients, LOA: limits of agreement; SD: standard deviation. *All ICC are significant (p\u0026lt;0.001).\u0026nbsp;\u003c/em\u003e\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cardio-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"caon","sideBox":"Learn more about [Cardio-Oncology](http://cardiooncologyjournal.biomedcentral.com)","snPcode":"40959","submissionUrl":"https://submission.nature.com/new-submission/40959/3","title":"Cardio-Oncology","twitterHandle":"@OncoBioMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Methods, cardio-oncology, QT interval, ECG, pharmacology","lastPublishedDoi":"10.21203/rs.3.rs-7204236/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7204236/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBackground. Electrocardiogram (ECG) analysis is crucial to detect cardiotoxicity. Manual methods are time-consuming and limited by inter-reader variability, highlighting the need for precise, reproducible and rapid semi-automated digital tools in clinical practice. Objective. This study evaluates the triplicate concatenation method (TCM) using a semi-automated ECG software (CalECG-4.2, AMPS ®) by assessing intra-and inter-reader variabilities in two distinct cardio-oncology populations: breast cancer patients receiving ribociclib (a QT-prolonging drug) and patients admitted with severe immune checkpoint inhibitors (ICI)-myocarditis, a condition marked by QRS alterations.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMethods. A total of 420 ECG from 31 patients (21 ribociclib, and 10 ICI-myocarditis) were independently analyzed by two readers. Variability was assessed using Bland-Altman analyses and intraclass correlation coefficients (ICC). Nonlinear mixed-effects modelling quantified time-dependent changes in heart rate (HR), PR, QTc (Friderica’s HR correction), QRS duration and voltage (Sokolow-Lyon) accounting for inter-reader variability.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eResults. Intra and inter-reader reproducibility was excellent (ICC\u0026gt;0.99, including Sokolow-Lyon voltage; standard-deviation\u0026lt;4ms across all time-derived parameters). In ribociclib-treated patients (cycles of 21/28 days on drug), QTc peaked at day 14 (16±1ms, p\u0026lt;0.001) before decreasing by day 28 (-6±1ms, p\u0026lt;0.001) compared to baseline. In ICI-myocarditis, QRS duration increased at day 5 before returning to baseline starting day 28, while Sokolow-Lyon voltages increased progressively on immunosuppressive treatments, peaking at day 28 (458±49µV, p\u0026lt;0.001) and remaining constant afterwards for the next month.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConclusion. TCM with CalECG-4.2 ensures a high reproducibility while monitoring key parameters like QTc duration and Sokolow-Lyon voltage, making it a reliable and time-saving alternative for the ECG surveillance of drug toxicities in cardio-oncology.\u003c/p\u003e","manuscriptTitle":"Validation of a novel semi-automated ECG quantification tool, applied to a cardio-oncology setting Semi-Automated ECG Tool applied to cardio-oncology","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-19 12:31:23","doi":"10.21203/rs.3.rs-7204236/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-10T12:06:58+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-16T16:30:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"276699704437865149981225789596997784715","date":"2025-08-16T16:07:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-11T15:09:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-06T16:54:18+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-01T09:39:02+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cardio-Oncology","date":"2025-07-24T09:50:24+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cardio-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"caon","sideBox":"Learn more about [Cardio-Oncology](http://cardiooncologyjournal.biomedcentral.com)","snPcode":"40959","submissionUrl":"https://submission.nature.com/new-submission/40959/3","title":"Cardio-Oncology","twitterHandle":"@OncoBioMed","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"95760c0a-42c3-4a27-a236-adb17d7b45cb","owner":[],"postedDate":"August 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-22T16:05:31+00:00","versionOfRecord":{"articleIdentity":"rs-7204236","link":"https://doi.org/10.1186/s40959-025-00405-7","journal":{"identity":"cardio-oncology","isVorOnly":false,"title":"Cardio-Oncology"},"publishedOn":"2025-12-19 15:58:07","publishedOnDateReadable":"December 19th, 2025"},"versionCreatedAt":"2025-08-19 12:31:23","video":"","vorDoi":"10.1186/s40959-025-00405-7","vorDoiUrl":"https://doi.org/10.1186/s40959-025-00405-7","workflowStages":[]},"version":"v1","identity":"rs-7204236","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7204236","identity":"rs-7204236","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}