The MIRAGE trial showed that reduced PTV margins (when using real-time MRI gating) for prostate SBRT significantly reduced GU toxicity compared to standard CT-guided SBRT. An interesting question is why? Before you say, “Well duh, the margins were smaller,” understand that the authors found fairly modest differences in dosimetry between the MRI and CT arms. For example, there was no difference in bladder Dmax nor V20, the V40 was lower in the MRI arm at 0.3cc v 0.7cc, and the V39 was also lower (1.9cc v 3.7cc). Did these differences drive the nearly 50% relative reduction in grade 2 GU toxicity? Here’s a look at prostate motion in the MIRAGE trial. MRI cine images were acquired 4 times per second during treatment, and the beam was automatically held if >10% of the prostate area moved outside a 3mm window. They used a measure called the gating duty cycle (GDC) to quantify the fraction of time that the prostate was within the gating window. Across 391 fractions, the median GDC was 0.974, which means the prostate was within the gating window for 97.4% of all cine images acquired during the fraction. However, in just under a quarter of fractions (23%), the GDC was <0.9, and in 9% it was <0.8. On a per-patient basis, 62.3% had at least one fraction with a GDC <0.9, and 30.4% had a fraction with a GDC <0.8. What does this mean? They found that a lower minimum GDC (more motion outside the gating window) was associated with worse GU toxicity. During low GDC fractions, OARs like the bladder neck often moved inferior and posterior further into the high dose volume. This brings up an interesting idea about how gating reduces toxicity. Yes, gating allows for smaller margins, but gating also reduces effective prostate motion during beam-on. Without gating, larger margins are needed to adequately account for random motion, but also the beam isn’t held when that motion occurs. | Neylon, Int J Radiat Oncol Biol Phys 2023
