Clinical Interpretation Research Articles

Verbal paired associates learning lateralizes left hippocampal sclerosis in temporal lobe epilepsy.

Left hippocampal sclerosis (HS) is associated with verbal-specific memory impairment. This association is well established for word list learning tasks, and there is some evidence that this may also be relevant to verbal paired associates learning (PAL), though the evidence is limited. We aimed to evaluate the utility of verbal PAL as a marker for left HS, compare this with word list learning, and derive cutoff scores to facilitate clinical application. Retrospective analysis of Rey Auditory Verbal Learning Test (RAVLT) and PAL scores obtained from 116 surgically naïve temporal lobe epilepsy patients with HS (14 bilateral, 57 left, 45 right; range of N across test indices: 77-110). Relative to right HS, left HS and bilateral HS were associated with poorer performance on PAL Hard Pairs (p < 0.001) and all RAVLT indices (p < 0.001), with performance comparable in left HS and bilateral HS (p > 0.05). PAL Hard Pairs and all RAVLT indices displayed acceptable discriminatory ability (AUC > 0.70) in classifying left-sided HS (unilateral left or bilateral HS), with RAVLT delayed recall the strongest predictor (AUC = 0.87; PAL Hard Pairs Learning and Delay = 0.80 and 0.83, respectively). Optimal cutoff scores for left-sided HS classification were generated. Although the RAVLT delayed recall was the strongest predictor of left-sided HS, PAL Hard Pairs also demonstrated excellent discriminatory capacity, offering an additional cognitive marker of left hippocampal integrity to complement word list learning in clinical assessments. The results of the study show that the ability to learn unrelated pairs of words (e.g., "silver"-"run") is compromised in the setting of epilepsy with left-sided hippocampal sclerosis (HS), be it unilateral left HS or bilateral HS. The ability of unrelated word pair learning to discriminate left vs. right HS was comparable to word list learning, a task with demonstrated sensitivity to left HS. Our results suggest that the ability to learn unrelated pairs of words provides another useful marker of left-sided hippocampal compromise in epilepsy. We provide cutoff scores to facilitate clinical interpretation.

Impaction bone grafting for acetabular bone defects replacement in revision hip arthroplasty: biomechanical aspects

Background. The leading role in long-term survival of the prosthesis installed using impaction bone grafting is given to the mechanical properties of the graft. The aim of the study is to explore the mechanical properties of osteoplastic material and determine the potential impact of cyclic loads on dynamic changes in the position of the acetabular component after revision hip arthroplasty using impaction bone grafting. Methods. We conducted an experiment evaluating the impact of cyclic loads on the mechanical properties of osteoplastic material. At the first stage, a single-cycle tension-compression testing was carried out. Cyclic tests were carried out at the second stage of the experiment. Taking into account the presence of blood in the wound, we provided for a model with an aqueous solution of 45% glycerin. Clinical interpretation of biomechanics was carried out basing on the dynamic radiography data of two patients who underwent revision hip arthroplasty with the use of impaction bone grafting (IBG). The changes in the position of the rotation center and acetabular component were assessed. Results. During a single-cycle loading, we observed stress-strain dependences and instantaneous elastic moduli for each specimen. During cyclic tests, we obtained the increase of the instantaneous elastic modulus by 2.6 times for a “dry” specimen and from 3.9 to 4.7 times for the ones with liquid. X-rays of both patients showed the shift of the center rotation cranially and laterally: 2.4 and 1.5 mm in the first case and 14.9 and 9.5 mm in the second one, respectively. In the first case the change in the inclination was 18.7º, in the second case — 19.8º. The Hip Harris Score (HHS) was 97 points for the first patient, 53 points — for the second one. Conclusions. The material used for IBG is subject to deformation both in the intraoperative and postoperative period. Compression tests have suggested that the deformation of morselized impacted bone graft gradually tends to reach a plateau in the postoperative period, and with the completion of the deformation, migration of the acetabular component stops. The change in the position of the rotation center and acetabular component in the absence of a radiolucent line is not an absolute sign of loosening.

Efficient reinterpretation of rare disease cases using Exomiser

Whole genome sequencing has transformed rare disease research; however, 50–80% of rare disease patients remain undiagnosed after such testing. Regular reanalysis can identify new diagnoses, especially in newly discovered disease-gene associations, but efficient tools are required to support clinical interpretation. Exomiser, a phenotype-driven variant prioritisation tool, fulfils this role; within the 100,000 Genomes Project (100kGP), diagnoses were identified after reanalysis in 463 (2%) of 24,015 unsolved patients after previous analysis for variants in known disease genes. However, extensive manual interpretation was required. This led us to develop a reanalysis strategy to efficiently reveal candidates from recent disease gene discoveries or newly designated pathogenic/likely pathogenic variants. Optimal settings to highlight new candidates from Exomiser reanalysis were identified with high recall (82%) and precision (88%) when including Exomiser’s automated ACMG/AMP classifier, which correctly converted 92% of variants from unknown significance to pathogenic/likely pathogenic. In conclusion, Exomiser efficiently reinterprets previously unsolved cases.

“Visual thinking strategies” improves radiographic observational skills but not chart interpretation in third and fourth year veterinary students

The ability to observe and interpret images and clinical information is essential for veterinarians in clinical practice. The purpose of this study is to determine the utility of a novel teaching method in veterinary medicine, the incorporation of art interpretation using the Visual Thinking Strategies (VTS), on students’ observational and clinical interpretation skills when evaluating radiographs and patient charts. Students were asked to observe and interpret a set of radiographs and a patient chart, subsequently involved in art interpretation using VTS, and then asked to observe and interpret a different set of radiographs and a different patient chart. Qualitative and quantitative analysis was performed, including scoring of observations and interpretations by a radiologist and emergency and critical care resident. For radiographs, observation and interpretation scores increased significantly after VTS. There was no change in patient chart observation or interpretation scores after VTS. Broadly, VTS provided creative thinking and visual literacy exercises that students felt pushed students them to think more openly, notice subtleties, use evidential reasoning, identify thinking processes, and integrate details into a narrative. However, its impact on clinical reasoning, as assessed by chart observation and interpretation scores, was uncertain. Further studies are needed to determine the optimal way to incorporate art interpretation in the veterinary medical curriculum.

Functional analysis of G6PD variants associated with low G6PD activity in the All of Us Research Program.

The glucose-6-phosphate dehydrogenase (G6PD) enzyme protects red blood cells against oxidative damage. Individuals with G6PD-impairing polymorphisms are at risk of hemolytic anemia from oxidative stressors. Prevention of G6PD deficiency-related hemolytic anemia is achievable by identifying affected individuals through G6PD genetic testing. However, accurately predicting the clinical consequence of G6PD variants is limited by over 800 G6PD variants which remain of uncertain significance (VUS). There also remains inconsistency in which deficiency-causing variants are included in genetic testing arrays: many institutions only test c.202G > A, though dozens of other variants can cause G6PD deficiency. Here, we improve G6PD genotype interpretations using the All of Us Research Program data and a yeast functional assay. We confirm that G6PD coding variants are the main contributor to decreased G6PD activity and that 13% of individuals in the All of Us data with deficiency-causing variants would be missed by only genotyping for c.202G > A. We expand clinical interpretation for G6PD VUS, reporting that c.595A > G ("Dagua" or "Açores") and the novel variant c.430C > G reduce activity sufficiently to lead to G6PD deficiency. We also provide evidence that 5 missense VUS are unlikely to lead to G6PD deficiency, and we applied the new World Health Organization (WHO) guidelines to recommend classifying 2 synonymous variants as WHO Class C. In total, we provide new or updated clinical interpretations for 9 G6PD variants. We anticipate these results will improve the accuracy, and prompt increased use, of G6PD genetic tests through a more complete clinical interpretation of G6PD variants.

Socioeconomic and demographic inequalities in off-label prescription of glucagon-like peptide-1 (GLP-1) receptor agonists - a Swedish descriptive cohort study.

Introduction In Sweden, glucagon like peptide-1 (GLP-1) receptor agonists are subsidized for diabetes indication but not for obesity. Unregulated off-label prescription of GLP-1 receptor agonists for obesity treatment may raise concerns about potential inequalities for both patient groups. This study aims to describe socioeconomic and demographic characteristics of on- and off-label users of GLP-1 receptor agonists in persons without a diagnosis of diabetes. Methods Swedish descriptive register-based cohort study of persons who filled a prescription of a GLP-1 receptor agonist at least once during 2018-2022. Individuals were excluded from the study population if they had a diagnosis of diabetes or previous prescription fills of insulin/analogs at any time prior to the first filled prescription of a GLP-1 receptor agonist. Socioeconomic and demographic characteristics were described overall and stratified by sex and prior use of anti-obesity medications. Off-label use was defined by filled prescriptions of GLP-1 receptor agonists which are indicated for diabetes treatment. Results The study population included 16,436 individuals, of which 70.1% were women, 30.7% had previously filled a prescription of anti-obesity medications, 65.3% had Sweden as country of origin and 17.2% an Asian country. In the analyses stratified by sex women were more likely to have an education longer than 9 years (84.8% vs 78.3% in men). Nonetheless, women had lower annual individual (2891.3 vs 4004.9 in men) and family disposable income (5645.5 vs 6092.5 in men). Overall, on-label prescription was higher in women (49.2% vs 30.9% in men), while off-label was more common among men (69% vs 51% in women). Conclusion High family disposable income, and male sex are common among off-label glucagon-like peptide-1 receptor agonist users compared to users of the only on-label glucagon-like peptide-1 receptor agonist in Sweden. Regional differences indicate different clinical practices and guideline interpretations.

Polygenic risk scores in the clinic: a systematic review of stakeholders' perspectives, attitudes, and experiences.

Polygenic Risk Scores (PRS) are statistical methods estimating part of an individual's genetic susceptibility to various disease phenotypes. Their potential clinical applications to enhance the prediction, prevention, and risk management of complex conditions motivate current research efforts worldwide. While a growing body of literature has highlighted the scientific and ethical limitations of PRS, the technology's clinical translation will present both opportunities and challenges for the stakeholders involved. Here, a mixed-method systematic review of empirical studies was performed to gather evidence on the perspectives, attitudes, and experiences of healthcare providers, patients, and the public regarding the use of PRS in healthcare settings. The PRISMA reporting protocol was followed and 24 articles were included. Three major themes were identified. First, we reported on participants' familiarity with the test, including their knowledge, understanding, and education on PRS' clinical use. The second theme collects stakeholders' motivations for taking the test and their perspectives on sensitive issues related to the return of results. Participants' normative stances regarding the appropriate use of PRS, their benefits, and harms were presented in the third theme. The findings underscore significant knowledge gaps and challenges in the clinical interpretation of PRS among healthcare providers. On the other hand, the provision of genetic counseling benefitted patients' understanding of PRS results and in most cases, no psychosocial burden was reported. Finally, the review highlights that stakeholders' perspectives on the clinical use of PRS are highly context-dependent, shaped by population characteristics, disease type, and social factors, emphasizing the need for tailored approaches across diverse healthcare settings.

Health-related quality of life profiles of adults with arthritis and/or fibromyalgia: a cross-sectional study.

Adults with arthritis experience poor health-related quality of life (HRQOL), though research often focuses on single HRQOL outcomes or summary scores. We aimed to identify HRQOL profiles in adults with different arthritis types and determine risk and protective factors. Data including PROMIS-29 Profile v2.1 and PROMIS Short Form v2.0 - Emotional Support 4a were collected through a national foundation's online survey of adults with arthritis in the U.S. We used latent profile analysis (LPA) to characterize the heterogeneity in arthritis patients by clustering them into HRQOL profiles, based on statistical model fit and clinical interpretability. We fit a multinomial logistic regression model with HRQOL profile assignment as the outcome to determine associations with protective and risk factors. We included 25,305 adults with arthritis. The LPA results favored a five-HRQOL profile solution (entropy = 0.83). While some profiles displayed better HRQOL in some domains, 93% of the sample displayed impacted pain and physical functioning. One profile (20%) displayed mean T-scores nearly 2 standard deviations below the population mean. Despite poor physical HRQOL outcomes, one profile (10%) displayed average mental health. All demographic and clinical factors contributed significantly to the model, including risk factors (arthritis types, work status) and protective factors (more emotional support, starting exercise). We identified profiles with consistently impacted HRQOL in arthritis, though one displayed average mental health functioning despite poor physical functioning. These results highlight the value of considering the patient's HRQOL experience alongside treatment options, and the potentially positive impact of non-pharmacological interventions.

Factor Structure and Validity of Composite Scores Resulting From a Computerized Cognitive Test Battery in Healthy Adults and Patients With Primary Brain Tumors.

Computerized neuropsychological test batteries (CNTs), such as Central Nervous System Vital Signs (CNS VS), are increasingly used for measuring cognitive functioning, but empirical evidence of how they measure cognition is scarce. We investigated the factor structure of CNS VS using exploratory factor analyses four samples: healthy adults (n = 169), patients with meningioma (392), low-grade glioma (99), and high-grade glioma (247). We tested model fit and investigated measurement invariance. Differences in factor interpretation existed between healthy participants and patients. Factor structures among patient groups were approximately the same but differed in non-zero loadings. Overall, factor structures largely did not support the "clinical domains" provided by CNS VS for clinical interpretation. Confirmatory models did not have a good fit, and measurement invariance could not be established. Our results indicate that (weighted) sum scores of CNS VS results may lack validity. We recommend researchers and clinicians to use scores on individual test measures.

Cardiopulmonary exercise testing in pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension.

Cardiopulmonary exercise testing allows for a comprehensive assessment of the mechanism of exercise limitation in pulmonary arterial hypertension and chronic thromboembolic pulmonary hypertension. Competitive pathophysiologic mechanisms may affect the clinical interpretation of cardiopulmonary disease as they relate to dyspnea, leg fatigue, and exercise intolerance.

Artificial Intelligence Diagnosing of Oral Lichen Planus: A Comparative Study.

Early diagnosis of oral lichen planus (OLP) is challenging, which traditionally is dependent on clinical experience and subjective interpretation. Artificial intelligence (AI) technology has been widely applied in objective and rapid diagnoses. In this study, we aim to investigate the potential of AI diagnosis in OLP and evaluate its effectiveness in improving diagnostic accuracy and accelerating clinical decision making. A total of 128 confirmed OLP patients were included, and lesion images from various anatomical sites were collected. The diagnosis was performed using AI platforms, including ChatGPT-4O, ChatGPT (Diagram-Date extension), and Claude Opus, for AI directly identification and AI pre-training identification. After OLP feature training, the diagnostic accuracy of the AI platforms significantly improved, with the overall recognition rates of ChatGPT-4O, ChatGPT (Diagram-Date extension), and Claude Opus increasing from 59%, 68%, and 15% to 77%, 80%, and 50%, respectively. Additionally, the pre-training recognition rates for buccal mucosa reached 94%, 93%, and 56%, respectively. However, the AI platforms performed less effectively when recognizing lesions in less common sites and complex cases; for instance, the pre-training recognition rates for the gums were only 60%, 60%, and 20%, demonstrating significant limitations. The study highlights the strengths and limitations of different AI technologies and provides a reference for future AI applications in oral medicine.

BRCAIndica: a resource for ACMG/AMP classified BRCA1 and BRCA2 variants.

As genetic testing becomes increasingly accessible and affordable, the uniform and accurate interpretation of genetic variants becomes essential. The ACMG/AMP joint guidelines provide the basis for systematic and uniform interpretation of pathogenicity of genetic variants. However, the application of these in routine clinical interpretation at-scale has largely been limited by the lack of resources providing harmonized data especially at a population-scale. Here we describe BRCAIndica, a resource for BRCA1 and BRCA2 variants conforming to the ACMG & AMP joint guidelines to aid uniform clinical interpretation of genetic tests with a specific focus on variants reported in the Indian population. We collected and harmonized variants from across several resources including population-scale datasets, literature survey and other variant datasets. We then classified them according to the ACMG/AMP guidelines.We have collected a total of 10,490 unique variants, of which 2261 Pathogenic and 43 Likely Pathogenic variants belong to BRCA1 and 2694 Pathogenic and 20 Likely Pathogenic variants to BRCA2 respectively. BRCAIndica can be accessed at:https://clingen.igib.res.in/brcaindica/ . In conclusion, BRCAIndica is a powerful resource that offers researchers and clinicians with ACMG/AMP annotated BRCA variants.

Trajectories of antidepressant use after tamoxifen initiation among young and middle-aged women with breast cancer.

Antidepressant treatment patterns may change after women with breast cancer (BC) initiate tamoxifen, potentially impacting health outcomes. We characterized trajectories of antidepressant use after initiating tamoxifen among young and middle-aged women with BC, identifying risk factors for trajectory group membership. A retrospective cohort included women 18-64years-old with BC and antidepressant treatment history who received a new tamoxifen dispensing (index date). We measured longitudinal antidepressant use post-index date as 12, monthly, proportion of days covered (PDC) measurements in a 25% random sample of IQVIA PharMetrics® Plus for Academics US claims, 2006-2022. Group-based trajectory models identified latent subgroups of antidepressant use by testing 2-6-group representations; the best model fit determined by the lowest Bayesian Information Criterion, clinical interpretability, and each subgroup comprising ≥ 5% of the cohort. Using multinomial logistic regression, baseline covariates including demographics, depression status and the CYP2D6-inhibitory strength of antidepressants were evaluated as risk factors for the trajectory of antidepressant use after tamoxifen initiation. Our sample of 851 women followed four distinct antidepressant adherence trajectories after tamoxifen initiation: 12% exhibited immediately decreasing use [mean PDC (sd) 8% (± 7)]; 7% exhibited delayed decreasing use [41% (± 14)]; 20% exhibited dynamic-moderate use [54% (± 15)]; and 60% exhibited consistently high use [91% (+ 7)]. Age, depression, and treatment with non CYP2D6-inhibiting antidepressants were associated with women's trajectory of antidepressant use after initiating tamoxifen. Nearly 40% of women were nonadherent to antidepressants after tamoxifen initiation. Future research should explore cancer-related and mental health implications of this nonadherence.

Interpreting Population Mean Treatment Effects in the Kansas City Cardiomyopathy Questionnaire

The Kansas City Cardiomyopathy Questionnaire (KCCQ) is a commonly used outcome in heart failure trials. While comparing means between treatment groups improves statistical power, mean treatment effects do not necessarily reflect the clinical benefit experienced by individual patients. To evaluate the association between mean KCCQ treatment effects and the proportions of patients experiencing clinically important improvements across a range of clinical trials and heart failure etiologies. A patient-level analysis of 11 randomized clinical trials, including 9977 patients, was performed to examine the association between mean treatment effects and the KCCQ Overall Summary Score (OSS) and the absolute differences in the proportions of patients experiencing clinically important (≥5 points) and moderate to large (≥10 points) improvements. There was no target date range, and included studies were those for which patient-level data were available. Validation was performed in 7 additional trials. The data were analyzed between July 1 and September 15, 2023. Proportion of patients experiencing an improvement of 5 or more and 10 or more points in their KCCQ score (with each domain transformed to a range of 0 to 100 points, where higher scores represent better health status). Group mean KCCQ-OSS differences were strongly correlated with absolute differences in clinically important changes (Spearman correlations 0.76-0.92). For example, a mean KCCQ-OSS treatment effect of 2.5 points (half of a minimally important difference for an individual patient) was associated with an absolute difference of 6.0% (95% prediction interval [PI], 4.0%-8.1%) in the proportion of patients improving 5 or more points and 5.0% (95% PI, 3.1%-7.0%) in the proportion improving 10 or more points, corresponding to a number needed to treat of 17 (95% PI, 12-25) and 20 (95% PI, 14-33), respectively. Inferences about clinical impacts based on population-level mean treatment effects may be misleading, since even small between-group differences may reflect clinically important treatment benefits for individual patients. Results of this study suggest that clinical trials should explicitly describe the distributions of KCCQ change at the patient level within treatment groups to support the clinical interpretation of their results.

Phenotypic Resistance to Rifampicin of Mycobacterium tuberculosis with the rpoB Leu430Pro Mutation

The objective: to determine the role of the rpoB Leu430Pro mutation in the degree of phenotypic resistance of Mycobacterium tuberculosis to rifampicin.Subjects and Methods. The WHO classifies the rpoB Leu430Pro mutation of Mycobacterium tuberculosis as a borderline resistance mutation but of clinical significance. From an array of cultures, we selected and analyzed 14 isolates of M. tuberculosis with discrepancies in the results of molecular genetic and phenotypic testing, carrying only this mutation in rpoB. For all samples, the phenotypic resistance level (MIC) of these isolates to rifampicin was determined using BACTEC MGIT 960 and Middlebrook 7H11 medium.Results. It was found out that 12 of 14 isolates had the rifampicin MIC below the current critical concentration when tested by both methods, thus they were phenotypically sensitive. One isolate was resistant when tested with Middlebrook 7H11 but susceptible when tested with BACTEC MGIT 960. Only one isolate which had the additional rpoB F425L mutation demonstrated high-level phenotypic resistance when tested by the same tests. Our data indicate that the clinical significance of this mutation requires clarification since even a decrease in the critical concentration of rifampicin does not lead to unambiguous results of molecular genetic and phenotypic tests. It is necessary to standardize the use of various molecular genetic tests and principles of their clinical interpretation to optimize strategies for managing patients with tuberculosis caused by M. tuberculosis with the rpoB Leu430Pro mutation.

Radiological assessment of Sella Turcica morphology correlates with skeletal classes in an Austrian population: an observational study.

This study aimed to analyze variations in the sella turcica (ST) concerning its size, shape, and bridging, providing first reference values in Austrian individuals. Additionally, it assessed associations between these morphological and demographic parameters and their correlation with patients' skeletal class. 208 lateral cephalometric radiographs (154 female, 54 male; age 8-58 years) from DPU Dental Clinic (Austria) were included. Size, skeletal class, shape, age, and gender of ST were tested for significance in correlation using, (M)ANOVA, and chi-square. Linear dimensions of ST ranged from 11.1 to 12.9mm across readers, with a standard deviation of 2.0-2.2mm. Normal ST (49.76%) and round ST (58.77%) were the most frequent. ST bridging was detected in 6.97%. Skeletal class I appeared most frequently (54.8%). Statistical significance was observed between age, gender, and ST length, with further significant age effects on ST shape. Moreover, age showed significant modification of ST shape, while skeletal parameters appeared unaffected by other ST parameters. These preliminary findings define normal ST dimensions in an Austrian population, offering reference values for clinical interpretation and broadening the available European data. Clear associations between morphological and demographic parameters were detected. Additionally, these findings may contribute to diagnostic and therapeutic strategies in orthodontics and craniofacial pathology. Future studies employing cone beam computed tomography (CBCT) along a larger sample size could enhance the generalizability of these findings.

Abstract 4138690: A Case of Hypertrophic Cardimyopathy: Digenic Variants of Uncertain Significance Mutations in MHY7 and RYR2 Genes

Introduction: Hypertrophic Cardiomyopathy (HCM) is primarily a disease caused by mutations in single genes, but around 5% of HCM patients, exhibit 2 (digenic) or more (oligogenic) causal mutations in the same gene or in different genes. We present a case of familial HCM with variants of uncertain significance (VUS) mutations in the MYH7 and RYR2 genes, highlighting the complexity of genetic inheritance and its implications on disease severity. Case presentation: A 40-year-old male presented with a history of hypertension, HCM, and two prior myocardial infarctions. He had syncope, non-sustained ventricular tachycardia, and a family history of sudden cardiac death (SCD) in his maternal grandfather. A dual-chamber ICD was placed for primary prevention of SCD because of his high-risk profile. Echocardiography demonstrated normal LV size and function with moderate asymmetric septal hypertrophy. Cardiac MRI indicated diffuse asymmetric hypertrophy and a 10.9% LV myocardial scar burden. Left heart catheterization revealed normal coronary arteries with LVOT gradient up to 60 mmHg, consistent with HCM. Genetic testing revealed VUS in the MYH7 (c.1305G&gt;A, p.Met435Ile) and RYR2 (c.10528C&gt;A, p.Arg3510Ser) genes (Figure 1). Further family genetic testing confirmed these VUS mutations in multiple relatives (Figure 2). Discussion: The identification of VUS mutations in MYH7 and RYR2 genes in multiple family members highlights the complexity of genetic inheritance in HCM. These findings underscore the challenges in clinical interpretation and risk stratification, as the co-occurrence of mutations in different genes may have synergistic effects on disease severity and phenotype expression. Comprehensive genetic evaluation, including family studies, is crucial for understanding the clinical significance of these mutations and guiding personalized management strategies. Further research is needed to elucidate the impact of VUS mutations on the clinical outcomes of HCM patients.

Mortality Prediction Modeling for Patients with Breast Cancer Based on Explainable Machine Learning.

Background/Objectives: Breast cancer is the most common cancer in women worldwide, requiring strategic efforts to reduce its mortality. This study aimed to develop a predictive classification model for breast cancer mortality using real-world data, including various clinical features. Methods: A total of 11,286 patients with breast cancer from the National Cancer Center were included in this study. The mortality rate of the total sample was approximately 6.2%. Propensity score matching was used to reduce bias. Several machine learning models, including extreme gradient boosting, were applied to 31 clinical features. To enhance model interpretability, we used the SHapley Additive exPlanations method. ML analyses were also performed on the samples, excluding patients who developed other cancers after breast cancer. Results: Among the ML models, the XGB model exhibited the highest discriminatory power, with an area under the curve of 0.8722 and a specificity of 0.9472. Key predictors of the mortality classification model included occurrence in other organs, age at diagnosis, N stage, T stage, curative radiation treatment, and Ki-67(%). Even after excluding patients who developed other cancers after breast cancer, the XGB model remained the best-performing, with an AUC of 0.8518 and a specificity of 0.9766. Additionally, the top predictors from SHAP were similar to the results for the overall sample. Conclusions: Our models provided excellent predictions of breast cancer mortality using real-world data from South Korea. Explainable artificial intelligence, such as SHAP, validated the clinical applicability and interpretability of these models.

Abstract 4136785: Interpreting Population Mean Treatment Effects in the Kansas City Cardiomyopathy Questionnaire

Introduction: The Kansas City Cardiomyopathy Questionnaire (KCCQ) is a common outcome in heart failure trials. While comparing means between treatment groups improves statistical power, mean effects do not necessarily reflect the clinical benefit experienced by patients. Objectives: To evaluate the relationship between mean KCCQ treatment effects in clinical trials and the proportions of patients experiencing clinically important improvements across a range of heart failure etiologies and interventions. Methods: We performed a patient-level meta-analysis of 11 randomized trials (9,977 patients) to examine the relationship between mean KCCQ Overall Summary Score (OSS) treatment effects and the absolute differences in the proportions of patients experiencing clinically important (≥5 points) and moderate or larger (≥10 points) improvements. Validation was performed in 7 additional trials. Results: Group mean KCCQ-OSS differences were strongly correlated with absolute differences in clinically important changes (Spearman correlations 0.76-0.92). For example, a mean KCCQ-OSS treatment effect of 2.5 points (half of a minimally important difference for an individual patient) was associated with an absolute difference of 6.0% (95% prediction interval [PI] 4.0-8.1) in the proportion of patients improving ≥5 points and 5.0% (95% PI 3.1-7.0) in the proportion improving ≥10 points, corresponding to NNTs of 17 (95% PI 12-25) and 20 (95% PI 14-33), respectively. Similar relationships were observed in the validation studies. Conclusions: Inferences on clinical impact based on population-level mean treatment effects are misleading since even small between group differences may reflect clinically important treatment benefits at the patient level. Clinical trials should explicitly describe the distributions of patient-level KCCQ changes to support clinical interpretation of their results.

Abstract 4127664: Influences of Age on Improvement of Health-Related Quality of Life in Patients with Atrial Fibrillation following Initial Treatment

Background: The prevalence of Atrial Fibrillation (AF), the most cardiac common arrhythmia, increases with age. Various treatment strategies are available to improve health-related quality of life (HRQoL) in patients with AF; however, the impact of age on HRQoL treatment benefits is unknown. Aims: To investigate the association of age with HRQoL benefit across different treatment strategies. Methods: Using data from a multicenter registry for outpatients with early recognized AF, HRQoL at enrollment and 1 year later was measured with the Atrial Fibrillation Effect on Quality-of-life quesTionnaire (AFEQT). The association between age and the 1-year changes in AFEQT-overall summary (OS) and subdomain scores with either rate or rhythm control strategies were assessed using generalized additive models (GAM) to address non-linear relationships. To support clinical interpretation, multivariable linear regression models were performed with categories of age ranges. Results: Among 2786 patients (31.7% women, 54.7% rhythm control); 946 were &lt;65 of age, 1019 were ≥65–75, and 821 were ≥75 years old. Rhythm (vs. rate) control was pursued in 76.6%, 58.0%, and 33.6% and mean AFEQT-OS changes were 11.1 (SD 16.6), 8.0 (17.0), and 4.7 (18.2) in those aged &lt;65, 65–74, and ≥75, respectively. The GAM showed improvements in AFEQT-OS attenuated with advancing age with both treatment strategies ( Figure ). Linear regression analyses showed lesser improvements in AFEQT-OS as higher age categories (≥65–75 vs. &lt;65, β -1.12 [95% CI, -2.33 – 0.09]; ≥75 vs. &lt;65, β -3.86 [-5.29 – -2.42]), which were mainly driven by lesser improvement in AFEQT-daily activities scores, without differences in AFEQT scores for symptom or treatment concerns ( Table ). Conclusions: In AF patients, HRQoL benefits after treatment were attenuated by advancing age, primarily driven by less improvement in functional limitations, underscoring the need for further research to identify optimal treatment strategies for older adults.