TL;DR
This study investigates the dosimetric impact of patient motion during Deep Inspiration Breath-Hold (DIBH) radiotherapy for left-sided breast cancer. By modifying their treatment planning system to analyze real-time motion data, the researchers found that while heart and lung doses were largely unaffected, internal mammary node (IMN) coverage could be modestly reduced due to these subtle movements.
What Problem Does This Paper Address?
Radiotherapy for left-sided breast cancer poses a risk of cardiac toxicity due to the heart's proximity to the treatment area. Deep Inspiration Breath-Hold (DIBH) is a technique used to mitigate this by having patients hold their breath at maximal inspiration, which moves the heart away from the radiation field. However, even during DIBH, patients can experience subtle movements (translational and rotational) of their chest surface. The problem this paper addresses is quantifying the dosimetric consequences of this residual motion on critical organs like the heart and lungs, as well as target coverage, specifically for internal mammary nodes (IMNs) which are often included in the treatment field. Previous studies had focused on motion itself but not its direct dosimetric impact in a clinical DIBH setting.
Methods
The researchers utilized data from 30 left-sided DIBH breast cancer patients treated using the AlignRT surface imaging system. AlignRT captures real-time translational and rotational displacements of the patient's chest surface, referred to as Real Time Deltas (RTDs). The core of their methodology involved modifying their in-house treatment planning system (TPS) to incorporate these RTD measurements.
Instead of planning based on a static CT scan, they retrospectively recalculated the delivered dose for each patient by considering the actual beam-on portions of the RTDs. For each recorded motion instance, the dose was recomputed, and the final delivered dose was an equally weighted average of these multiple resultant doses. This allowed them to compare the planned dose (assuming zero motion) with a more accurate estimate of the delivered dose, accounting for patient movement during treatment. They specifically analyzed the mean heart dose, lung V20 (volume of lung receiving > 20 Gy), and for patients with IMNs, the percentage of IMN covered by 40 Gy and the mean IMN dose.
Datasets
The study utilized retrospective data from 30 left-sided supine DIBH breast cancer patients treated between 2012 and 2013. Motion log files containing translational and rotational displacement data (RTDs) were collected from the AlignRT system for these patients. No new benchmarks were introduced; the study focused on analyzing existing clinical data.
Key Insights and Ideas
- DIBH is effective, but not perfect: While DIBH significantly moves the heart away from the radiation field, the residual motion, even if small (<1 mm and <1°), can still have a measurable impact.
- Heart and lungs are relatively robust to DIBH motion: The dosimetric impact on the heart and lungs was found to be minimal, with delivered doses closely matching planned doses. This suggests that the primary benefit of DIBH in sparing these organs is largely maintained despite patient movement.
- IMNs are more sensitive to motion: Internal mammary nodes, being closer to the edge of the tangential beams, are more susceptible to underdosage when patient motion occurs. This highlights the importance of considering motion when targeting these nodes.
- Motion analysis is crucial for accurate dose estimation: Modifying the TPS to incorporate real-time motion data provides a more realistic assessment of the delivered dose compared to static plan evaluation.
Key Results
- Motion levels: The averaged mean motion during DIBH was small, with translational movements averaging 0.81 ± 1.29 mm (vertical), 0.68 ± 0.85 mm (longitudinal), and 0.76 ± 0.85 mm (lateral), and rotational movements averaging 0.96° ± 0.49° (yaw), 0.93° ± 0.43° (roll), and 1.03° ± 0.50° (pitch).
- Heart and Lung Doses: The average planned mean heart dose was 99 cGy, and the delivered was 101 cGy. Lung V20 was 6.59% planned and 6.74% delivered. These differences were considered reasonably close to the planned values.
- IMN Coverage: For patients with IMNs included, the average planned 40 Gy coverage was 83%, while the delivered coverage was 77%. The average IMN mean dose was 4642 cGy planned and 4518 cGy delivered. This indicates a modest reduction in IMN coverage due to motion.
Limitations and Future Work
- Rigid Body Assumption: The study assumed a rigid body model for the chest/breast and heart, which is a simplification. The actual relative positions can change.
- Deformable Registration: The authors suggest that a more sophisticated model, such as deformable registration, could potentially improve the accuracy of the analysis.
- Visual Feedback: Audio coaching was used for all patients. The authors note that visual coaching has been shown to improve reproducibility and suggest considering a visual feedback system in their clinic.
- IMN Coverage Variability: While the average IMN coverage was reduced, the authors acknowledge that this reduction might be more significant in certain cases, potentially necessitating electron boosts.
- Comparison with Prone Position: The study focused on supine DIBH. Future work could involve comparing these findings with prone position treatments, which have also been shown to reduce cardiac dose.
- Interfraction vs. Intrafraction Motion: The study primarily focused on intrafraction motion (within a single breath-hold). The authors note that interfraction motion (patient setup variations between fractions) can also be significant and warrants further investigation.
Summary updated