Preliminary LGE- CMR defined scar consolidation after stereotactic cardiac radiotherapy for refractory ventricular tachycardia: a retrospective pilot analysis

Background

Ventricular tachycardia (VT) in structural heart disease remains a major therapeutic challenge. Stereotactic arrhythmia radiotherapy (STAR) has emerged as a rescue treatment in highly selected patients after failure of antiarrhythmic drugs and radiofrequency catheter ablation. While reduction in VT burden has been reported, mid-term substrate modification after STAR remains insufficiently characterized.

Purpose

To quantify pre- to post-STAR changes in late gadolinium enhancement CMR–defined substrate within the clinical target volume (CTV) compared with remote myocardium, using a standardized analysis workflow.

Methods

Consecutive patients undergoing STAR for refractory VT were screened for availability of both pre- and post-treatment LGE-CMR. STAR was delivered as a single fraction (20–25 Gy) to a VT substrate–guided clinical target volume (CTV).

LGE-CMR (PSIR sequence, 1.5T, wideband protocol for device carriers) was analyzed using a standardized workflow (ADAS-VT, 10 transmural shells). Pixel signal intensity maps defined dense scar (core) and border zone (BZ), expressed as percentage of regional myocardial volume.

Primary endpoints were percentage-point changes (Δpp) in core% and BZ% within the CTV; secondary endpoints were corresponding changes outside the CTV.

Results

Pre- and post-treatment LGE-CMR were available in 13 patients; 5 datasets had completed standardized processing at the time of submission and were included in the present analysis.

Within the CTV, 3 of 5 patients demonstrated a consistent remodeling pattern characterized by reduction in BZ% (−10 to −24 percentage points) accompanied by increased core% (+18 to +60 percentage points), compatible with scar densification and reduced tissue heterogeneity (Figure 1, Table 1). Two patients exhibited divergent remodeling patterns. Changes outside the CTV were variable and did not show a consistent trend.

Conclusion

A standardized, quality-controlled LGE-CMR workflow is feasible to assess myocardial remodeling after STAR. In this preliminary analysis, most evaluable cases demonstrated reduction in border zone with concomitant increase in dense scar within the irradiated CTV, suggesting scar densification and reduced tissue heterogeneity.

However, this interpretation likely oversimplifies the biological effects of STAR, which may include acute electrophysiological reprogramming, diffuse interstitial remodeling, and anti- and pro-inflammatory changes not fully captured by LGE alone. Integration of additional CMR sequences (e.g., T1/T2 mapping, strain analysis) may provide a more comprehensive characterization of post-STAR remodeling.

Underlying disease progression and intervening RF ablation procedures must also be considered when interpreting longitudinal substrate changes. Larger prospective analyses are required to clarify the relationship between imaging remodeling and clinical VT outcomes.