Digital twins in hypertrophic cardiomyopathy from sarcomeric mutation to predictive personalized care

To the Editor,

Hypertrophic cardiomyopathy (HCM) is a highly prevalent monogenic heritable myocardial disorder driven by sarcomeric-gene mutations, predominantly in MYH7 and MYBPC3. These autosomal-dominant pathogenic variants cause impairment of sarcomeric function and modification of myofilament dynamics by enhancing calcium sensitivity, impairing cross-bridge cycling, reducing energetic efficiency, and inducing the expansion of extracellular matrix. This collectively results in myocyte hypertrophy, disorganization, fibrosis, myocardial stiffness, and progressive diastolic dysfunction^[1,2]. Clinically, affected individuals may develop heart failure, arrhythmias, and sudden cardiac death (SCD). Cardiac magnetic resonance (CMR) with late-gadolinium enhancement (LGE) enables sensitive detection of myocardial fibrosis, and mutation-positive patients (particularly those with MYH7 or MYBPC3 variants) always have a higher LGE burden and more extensive fibrosis than mutation-negative patients with HCM^[1,3].

High-fidelity digital twin (DT) models that incorporate patient-specific sarcomeric mutations (MYH7, MYBPC3) provide a more accurate framework for HCM management. DTs allow multiscale modeling of myocardial mechanics under various physiological and pathological conditions by integrating continuous physiological measurements of wearable sensors, quantitative CMR strain, and LGE parameters. These models, which are developed through machine learning algorithms, predict fiber strain, wall stress, diastolic deterioration, and fibrotic progression to make personalized predictions of optimal time and modality to implement interventions such as pharmacotherapy, septal reduction, or device therapy. This changing paradigm also has applicability to other myofilament-mediated cardiomyopathies, such as dilated cardiomyopathy^[4].

Increasing clinical experience and novel computational studies clearly enhance the predictive utility of a digital-twin (DT) framework for hypertrophic cardiomyopathy. Machine-learning integration of physiologic data, CMR-based strain metrics, and tissue-level signatures allows these models to detect early mechanical deterioration with greater temporal resolution than traditional risk scores. In silico experimental studies with in silico cardiomyocytes with MYH7 mutation are highly reminiscent of the contractile defects observed in patient-derived tissue and recapitulate the mutation-specific susceptibility of myofilaments to stress^[5]. Similarly, genotype-oriented imaging observations reveal that MYH7 and MYBPC3 variants are invariably associated with concentric cellular hypertrophy and expansion of extracellular matrix phenotypes that can be monitored over time in disease progression through the use of DT platforms^[6]. The growing literature on radiomics also supports this modality: in large cohorts of HCM, ML-enhanced cine-CMR methods can differentiate LGE-positive from LGE-negative myocardium with an AUC around 0.83 and sensitivities exceeding 90%^[6,7]. When these imaging-derived markers are incorporated into a DT workflow, they can meaningfully guide the timing of therapeutic decisions. Observational outcome data demonstrate that appropriately selected patients undergoing septal-reduction procedures or early medical management achieve sustained symptomatic improvement with low complication rates, supporting the safety and clinical utility of DT-based stratification^[7].

Despite the major advances in the sphere of digital twin (DT) modeling, a range of practical and scientific issues has a constrained effect on its effectiveness to personalize the process of managing hypertrophic cardiomyopathy (HCM). Although it is a promising step that includes patient-specific genotypes, wearable hemodynamic data, and regular CMR strain analyses, enabling the prediction of disease progression with precision, there is a lack of standardized longitudinal datasets. Myofilament-level, high-resolution simulations have high computational demands and so could not be used extensively. In addition, there is poor familiarity among clinicians, and the costs involved are high; strong prospective validation between digital-twin predictions and other alternatives like HCM Risk-SCD is still unavailable. The related challenges should be addressed to transform the computational knowledge into safer and more effective patient care^[4].

The introduction of a genotypically anchored digital twin model, integrating cardiac magnetic resonance (CMR) strain, late gadolinium enhancement (LGE), and physiological stress data, transforms the management of hypertrophic cardiomyopathy (HCM) from a reactive to a proactive and individualized care paradigm. This sophisticated modelling approach enables highly personalized risk stratification and treatment planning. To confirm the predictive accuracy and clinical efficacy of this framework, large-scale multinational collaborative clinical trials are required. Furthermore, collaborative governmental and industrial efforts must focus on reducing the significant computational costs involved, thereby ensuring this maximum degree of personalized accuracy is equitably accessible to all high-risk patients.

This letter to the editor adheres to the Transparency in the Reporting of Artificial Intelligence in Research (TITAN) guidelines^[8].

Ethical approval

Ethical approval was not required for this article as it is a letter to the editor and did not involve human participants, animals, or identifiable patient data.

Consent

Not applicable.

Sources of funding

No external funding was received for the preparation of this manuscript.

Conflicts of interest disclosure

The authors declare no conflicts of interest.

Acknowledgements

Not applicable.