Offline adaptive radiation therapy in the treatment of prostate cancer: a case study

Adaptive assessment based on fractional CBCT images for cervical cancer

PURPOSE

Anatomical and other changes during radiotherapy will cause inaccuracy of dose distributions, therefore the expectation for online adaptive radiation therapy (ART) is high in effectively reducing uncertainties due to intra-variation. However, ART requires extensive time and effort. This study investigated an adaptive assessment workflow based on fractional cone-beam computed tomography (CBCT) images.

METHODS

Image registration, synthetic CT (sCT) generation, auto-segmentation, and dose calculation were implemented and integrated into ArcherQA Adaptive Check. The rigid registration was based on ITK open source. The deformable image registration (DIR) method was based on a 3D multistage registration network, and the sCT generation method was performed based on a 2D cycle-consistent adversarial network (CycleGAN). The auto-segmentation of organs at risk (OARs) on sCT images was finished by a deep learning-based auto-segmentation software, DeepViewer. The contours of targets were obtained by the structure-guided registration. Finally, the dose calculation was based on a GPU-based Monte Carlo (MC) dose code, ArcherQA.

RESULTS

The dice similarity coefficient (DSCs) were over 0.86 for target volumes and over 0.79 for OARs. The gamma pass rate of ArcherQA versus Eclipse treatment planning system was more than 99% at the 2%/2 mm criterion with a low-dose threshold of 10%. The time for the whole process was less than 3 min. The dosimetric results of ArcherQA Adaptive Check were consistent with the Ethos scheduled plan, which can effectively identify the fractions that need the implementation of the Ethos adaptive plan.

CONCLUSION

This study integrated AI-based technologies and GPU-based MC technology to evaluate the dose distributions using fractional CBCT images, demonstrating remarkably high efficiency and precision to support future ART processes.

Rectal deformation management with IGRT in prostate radiotherapy: Can it be managed with rigid alignment alone?

PURPOSE

It is challenging to achieve appropriate target coverage of the prostate with Image Guided Radiation Therapy (IGRT) while simultaneously constraining rectal doses within planned values when there is significant variability in rectal filling and shape. We investigated if rectum planning goals can be fulfilled using rigid CBCT-based on-board alignment to account for interfraction rectal deformations.

METHODS

Delivered rectal doses corresponding to prostate alignment ("PR") and anterior rectum alignment ("AR") for 239 daily treatments from 13 patients are reported. Rectal doses were estimated by rigidly mapping the planned dose on the daily CT derived from the daily CBCT according to respective alignment shifts. Rectum V95% (rV95%) was used for analyses.

RESULTS

Compared to "PR", "AR" alignment increased rV95% for an average of 34.4% across all patients. rV95% (cc) averaged over all fractions was significant from planning values for 10/13 patients for "PR" and for 9/13 for "AR". 3/13 patients had reproducible anatomy. Of patients with non-reproducible anatomy, three had dosimetrically more favorable, while seven had less favorable anatomies. Most shift differences (82.3%) between the "PR" and "AR" alignments larger than 2 mm resulted in rV95% changes larger than 2 cc. Most shift differences (82.2%) of 2 mm or less between the "PR" and "AR" alignments resulted in rV95% changes less than 2 cc. The average percentage of fractions among patients in which anterior or posterior shifts for "AR" and "PR" alignment was larger than the PTV margins was 9.1% (0.0%-37.5%) and 1.3% (0%-10%).

CONCLUSION

Rectal deformation and subsequent inconsistent interfraction separation between prostate and rectal wall translate into anatomical changes that cannot always be mitigated with rigid alignment. If systematic differences exist due to a non-reproducible planning anatomy, attempts to restore the planned rectal doses through anterior rectum alignment produce rather small improvements and may result in unacceptable target underdosage.

Transformer CycleGAN with uncertainty estimation for CBCT based synthetic CT in adaptive radiotherapy

Objective. Clinical implementation of synthetic CT (sCT) from cone-beam CT (CBCT) for adaptive radiotherapy necessitates a high degree of anatomical integrity, Hounsfield unit (HU) accuracy, and image quality. To achieve these goals, a vision-transformer and anatomically sensitive loss functions are described. Better quantification of image quality is achieved using the alignment-invariant Fréchet inception distance (FID), and uncertainty estimation for sCT risk prediction is implemented in a scalable plug-and-play manner.Approach. Baseline U-Net, generative adversarial network (GAN), and CycleGAN models were trained to identify shortcomings in each approach. The proposed CycleGAN-Best model was empirically optimized based on a large ablation study and evaluated using classical image quality metrics, FID, gamma index, and a segmentation analysis. Two uncertainty estimation methods, Monte-Carlo Dropout (MCD) and test-time augmentation (TTA), were introduced to model epistemic and aleatoric uncertainty.Main results. FID was correlated to blind observer image quality scores with a Correlation Coefficient of -0.83, validating the metric as an accurate quantifier of perceived image quality. The FID and mean absolute error (MAE) of CycleGAN-Best was 42.11 ± 5.99 and 25.00 ± 1.97 HU, compared to 63.42 ± 15.45 and 31.80 HU for CycleGAN-Baseline, and 144.32 ± 20.91 and 68.00 ± 5.06 HU for the CBCT, respectively. Gamma 1%/1 mm pass rates were 98.66 ± 0.54% for CycleGAN-Best, compared to 86.72 ± 2.55% for the CBCT. TTA and MCD-based uncertainty maps were well spatially correlated with poor synthesis outputs.Significance. Anatomical accuracy was achieved by suppressing CycleGAN-related artefacts. FID better discriminated image quality, where alignment-based metrics such as MAE erroneously suggest poorer outputs perform better. Uncertainty estimation for sCT was shown to correlate with poor outputs and has clinical relevancy toward model risk assessment and quality assurance. The proposed model and accompanying evaluation and risk assessment tools are necessary additions to achieve clinically robust sCT generation models.

Adaptive radiation therapy strategies in the treatment of prostate cancer patients using hypofractionated VMAT

PURPOSE

To perform a comprehensive evaluation of eight adaptive radiation therapy strategies in the treatment of prostate cancer patients who underwent hypofractionated volumetric modulated arc therapy (VMAT) treatment.

MATERIAL AND METHODS

The retrospective study included 20 prostate cancer patients treated with 40 Gy total dose over five fractions (8 Gy/fraction) using VMAT. Daily cone beam computed tomography images were acquired before the delivery of every fraction and then, with the application of deformable image registration used for the estimation of daily dose, contouring and plan re-optimization. Dosimetric benefits of the various ART strategies were quantified by the comparison of dose and dose-volume metrics derived from treatment planning objectives for original treatment plan and adapted plans with the consideration of target volumes (PTV and CTV) as well as critical structures (bladder, rectum, left, and right femoral heads).

RESULTS

Percentage difference (ΔD) between planning objectives and delivered dose in the D99%  > 4000cGy (CTV) metric was -3.9% for the non-ART plan and 2.1% to 4.1% for ART plans. For D99%  > 3800cGy and Dmax  < 4280cGy (PTV), ΔD was -11.2% and -6.5% for the non-ART plan as well as -3.9% to -1.6% and -0.2% to 1.8% for ART plans, respectively. For D15%  < 3200 cGy and D20%  < 2800 cGy (bladder), ΔD was -62.4% and -68.8% for the non-ART plan as well as -60.0% to -57.4% and -67.0% to -64.0% for ART plans. For D15%  < 3200 cGy and D20%  < 2800 cGy (rectum), ΔD was -11.4% and -8.15% for non-ART plan as well as -14.9% to -9.0% and -11.8% to -5.1% for ART plans.

CONCLUSIONS

Daily on-line adaptation approaches were the most advantageous, although strategies adapting every other fraction were also impactful while reducing relative workload as well. Offline treatment adaptations were shown to be less beneficial due to increased dose delivered to bladder and rectum compared toother ART strategies.

Application of an automated dose accumulation workflow in high-risk prostate cancer - validation and dose-volume analysis between planned and delivered dose

Inter-fraction organ variations cause deviations between planned and delivered doses in patients receiving radiotherapy for prostate cancer. This study compared planned (D_P) vs accumulated doses (D_A) obtained from daily cone-beam computed tomography (CBCT) scans in high-risk- prostate cancer with pelvic lymph nodes irradiation. An intensity-based deformable image registration algorithm used to estimate contours for D_A was validated using geometrical agreement between radiation oncologist's and deformable image registration algorithm propagated contours. Spearman rank correlations (r_s) between geometric measures and changes in organ volumes were evaluated for 20 cases. Dose-volume (DV) differences between D_A and D_P were compared (Wilcoxon rank test, p < 0.05). A novel region-of-interest (ROI) method was developed and mean doses were analyzed. Geometrical measures for the prostate and organ-at-risk contours were within clinically acceptable criteria. Inter-group mean (± SD) CBCT volumes for the rectum were larger compared to planning CT (pCT) (51.1 ± 11.3 cm³vs 46.6 ± 16.1 cm³), and were moderately correlated with variations in pCT volumes, rs = 0.663, p < 0.01. Mean rectum DV for D_A was higher at V30-40 Gy and lower at V70-75 Gy, p < 0.05. Mean bladder CBCT volumes were smaller compared to pCT (198.8 ± 55 cm³vs 211.5 ± 89.1 cm³), and was moderately correlated with pCT volumes, rs = 0.789, p < 0.01. Bladder D_A was higher at V30-65 Gy and lower at V70-75 Gy (p < 0.05). For the ROI method, rectum and bladder D_A were lower at 5 to 10 mm (p < 0.01) as compared to D_P, whilst bladder D_A was higher than D_P at 20 to 50 mm (p < 0.01). Generated D_A demonstrated significant differences in organ-at-risk doses as compared to D_P. A well-constructed workflow incorporating a ROI DV-extraction method has been validated in terms of efficiency and accuracy designed for seamless integration in the clinic to guide future plan adaptation.

Patient-specific radiotherapy quality assurance for estimating actual treatment dose

This study aimed to evaluate the optimal method for planning computed tomography (CT) for prostate cancer radiotherapy to avoid a dose difference of ≥3% between the actual and planned treatments using multiple acquisition planning CT (MPCT). We calculated the 3-dimensional (3D) displacement error between the pelvic bone and matching fiducial marker on MPCT and cone-beam CT scans of 25 patients who underwent prostate volumetric-modulated arc therapy for prostate cancer. The correlation of the 3D displacement error and the dose difference between planned and actual treatments was calculated using least squares second-order polynomial model. The 3D displacement error showed a moderate correlation with differences between planned and accumulated treatment doses (r = 0.587, p < 0.0001). Moreover, the improvement rate of the minimum 3D displacement error showed a strong correlation with the relative error between each MPCT image (r = 0.793, p < 0.0001). Significant differences were observed between planned and actual treatment doses (p < 0.0001) in the relative 3D displacement errors of <1 mm, 1 to 3 mm, and >3 mm. The 3D displacement error on MPCT (as the selection estimation index for optimal planning CT) is useful for monitoring patient-specific intensity-modulated radiation therapy quality assurance. This new method allows to estimate dose differences from the planned dose before commencing treatment, thereby ensuring high-quality therapy. As radiotherapy quality is critical for patient outcome, these findings may contribute to better management of prostate cancer.

Optimizing planning CT using past CT images for prostate cancer volumetric modulated arc therapy

This study aimed to evaluate a new method to optimize planning computed tomography (CT) using three-dimensional (3D) displacement error between the planning and diagnosed past CT scans. Thirty-two patients undergoing volumetric modulated arc therapy for prostate cancer were evaluated for a 3D displacement error between bone- and prostate-matching spatial coordinates using multiple acquisition planning CT (MPCT) scans. Each MPCT image and a past CT image were used to perform rigid image registration (RIR) and deformable image registration (DIR), and the 3D displacement error was calculated. Correlations of the 3D displacement error in each MPCT scan and between the MPCT and past CT were evaluated based on RIR and DIR, respectively. The 3D displacement error in the MPCT images exhibited moderate correlation with the 3D displacement error between MPCT and past CT for both RIR (adjusted r² = 0.495) and DIR (adjusted r² = 0.398). In the correlation analysis between MPCT and past CT, image pairs with 3D displacement errors ≥ 6 mm were significantly different from those with errors < 6 mm (p < 0.0001). Past CT images were different from the planning CT images, which can be attributed to setup tools, flat-top plates, and physical differences due to the presence or absence of urine as well as prescription effects. The relationship between bone and prostate exhibited small deviations between the planning and past CT regardless of pretreatment. The prostate, which only has a slight effect on the displacement between it and bladder volume, was covered with a stiff pelvic bone. As a result, MPCT images exhibited correlations with past CT images of various difference states such as body positions. Finally, large 3D displacement errors in prostate position were caused by pelvic tension and stress, which can be detected using diagnosed past CT images instead of requiring MPCT scans. By comparing past and planning CT images, the random displacement error in the planning CT scan can be avoided by evaluating 3D displacement errors. The new method using the past CT images can estimate the displacement error of the prostate during the treatment period with 1 plan CT scan only, and it helps improve the treatment accuracy.