Cardiovascular magnetic resonance characterisation of early myocardial changes following allogeneic hematopoietic stem cell transplantation and their clinical predictors
K Chan, B T Costello, H T Dillon, N J Saner, A La Gerche, E J HowdenAbstract
Background
Allogeneic haematopoietic stem cell transplantation (alloHSCT) represents a curative yet high-risk therapeutic intervention for advanced haematological malignancies. Although treatment-related cardiotoxicity is traditionally monitored via echocardiography, cardiovascular magnetic resonance (CMR) serves as the gold standard for non-invasive tissue characterisation. Longitudinal CMR data evaluating early changes in myocardial structure and tissue characteristics following alloHSCT remain limited, with clinical and treatment-related predictors not well established.
Aims
This study aimed to characterise early post-alloHSCT alterations in cardiac structure and function via CMR and to identify potential clinical predictors associated with these changes.
Methods
In this post hoc analysis of the ALLO-Active trial, 44 patients (mean age 52.2 ± 15.3 years) underwent CMR pre- and early post-alloHSCT (mean 11.7 ± 18.8 and 146.5 ± 29.0 days pre- and post-alloHSCT, respectively). CMR parameters included ventricular volumes, ejection fraction (LVEF), global longitudinal strain (GLS), myocardial mass, wall thickness, and noncontrast T1 parametric mapping via ShMOLLI and SASHA sequences. Baseline clinical data, including demographics, medical history, cardiorespiratory fitness (VO2peak) and treatment exposures were recorded for all participants
Results
Following alloHSCT, significant increases were observed in indexed myocardial mass (53.2 ± 9.7 to 59.7 ± 11.1 g/m²; p < 0.001) and septal wall thickness (9.0 ± 2.3 to 10.8 ± 2.3 mm; p < 0.001) on CMR. Findings are detailed in Table 1. Native T1 mapping times increased significantly using both ShMOLLI (septal ∆ 49.2 ± 70.8 ms; p < 0.001) and SASHA (septal ∆ 24.6 ± 49.3 ms; p = 0.003) sequences. These structural and tissue changes occurred without significant reductions in LVEF or GLS. Univariate analyses identified no significant clinical predictors for the change seen in the majority of CMR parameters (all p >0.05). A history of hypertension was associated with a significantly greater increase in septal wall thickness compared to normotensive patients (median 3. mm [IQR 1.85–3.82] vs. 1.20 mm [IQR 0.40–2.35]; p = 0.011). This effect was localised, with no significant associations found for changes in total myocardial mass or contralateral wall thickness.
Conclusion
AlloHSCT is associated with early concentric cardiac remodelling and T1 prolongation, likely reflecting subclinical myocardial oedema. These alterations appear to be a generalised response to treatment, largely independent of traditional cardiac risk factors or specific comorbidities, apart from a modest signal of more pronounced septal thickening with a history of hypertension. While current treatment protocols remain appropriate, these findings support careful cardiac monitoring peri-alloHSCT.