Introduction
Traditional radiotherapy for gynecologic malignancies relies on static "snapshot" CT scans for treatment planning. However, the pelvic anatomy is dynamic; organs at risk (OARs) such as the bladder, bowel, and rectum shift and deform daily. To account for this movement, clinicians typically use generous safety margins, which inadvertently increase the radiation dose to healthy tissues. The advent of magnetic resonance imaging-guided linear accelerator (MR-linac) systems offers a solution by enabling real-time visualization and daily plan adaptation. This study reports the first-in-human clinical implementation of a 1.5 Tesla MR-linac for gynecologic cancers, following the R-IDEAL framework for evaluating new medical technologies.
Methodology and Workflow
The study utilized the Elekta Unity system, which integrates a 7-MV flattening-filter-free (FFF) linear accelerator with a 1.5 Tesla MR scanner. Ten patients with various gynecologic malignancies—including cervical, endometrial, vaginal, ovarian, and peritoneal cancers—were treated over a total of 90 fractions.
Two primary daily adaptation strategies were evaluated:
- Adapt-to-Position (ATP): A rigid shift of the original plan to align with the tumor's daily coordinates.
- Adapt-to-Shape (ATS): A more intensive process involving manual re-contouring of the tumor and OARs on the daily MR image, followed by full plan re-optimization.
The researchers assessed feasibility through treatment duration, setup variability (isocenter shifts), and quality assurance (QA) pass rates. Additionally, deformable image registration was used to calculate the "accumulated dose"—the actual dose delivered across the treatment course compared to the initial reference plan.
Key Findings
Clinical Feasibility and Efficiency
The implementation proved highly reliable, with 90 of 92 planned fractions successfully delivered on the MR-linac. The median treatment duration was 32 minutes per fraction. While this is longer than conventional 15-minute IMRT slots, it remains within a clinically acceptable range for adaptive therapy. Notably, the ATP workflow was significantly faster than the ATS workflow (30 minutes vs. 42 minutes, p < 0.0001).
Precision and Quality Assurance
Despite the absence of traditional external laser alignment, the system demonstrated high geometric accuracy. The average isocenter shift was less than 0.5 cm in all axes, relying entirely on internal MR imaging for setup. Furthermore, 100% of the adapted plans passed IMRT QA testing, with a median gamma pass rate of 99.9% (using a 3%/3mm criterion).
Dosimetric Impact of Anatomical Variation
The study highlighted that the necessity of complex adaptation (ATS) depends largely on tumor location:
- Peripheral Lesions: Tumors located in the pelvic periphery or near fixed structures (e.g., the ischium) showed stable dosimetry, with accumulated doses within 1% of the reference plan.
- Central Lesions: Tumors such as vaginal cuff boosts were highly sensitive to daily fluctuations in bladder and rectal filling. For these cases, dose differences ranged from -5.8% to -9.6%, emphasizing the value of daily visualization and potential shape-based adaptation to maintain target coverage.
Discussion and Limitations
This R-IDEAL stage 1/2a study confirms that 1.5T MR-guided adaptive radiotherapy is feasible and provides high dosimetric stability for gynecologic patients. The ability to "see" the anatomy daily allows for confidence in target coverage even amidst significant organ motion.
However, the study's small sample size (n=10) and the heterogeneity of the patient cohort limit the ability to define specific protocols for margin reduction. Additionally, the 22-cm longitudinal field size limit of the Elekta Unity currently restricts its use for patients requiring extended-field radiation, such as those needing coverage of the para-aortic lymph nodes.
Conclusion
The MR-linac is a reliable tool for the treatment of gynecologic cancers, offering a path toward more personalized and precise radiotherapy. Future research will focus on "margin attenuation"—reducing safety margins to minimize toxicity—and exploring the use of MR-linacs as an alternative to brachytherapy for patients where the latter procedure is not an option.
Summary updated