Application of real-time MRI-guided linear accelerator in stereotactic ablative body radiotherapy for non-small cell lung cancer: one step forward to precise targeting

PURPOSE

Tumor motion is a major challenge in stereotactic ablative body radiotherapy (SABR) for non-small cell lung cancer (NSCLC), causing excessive irradiation to compensate for this motion. Real-time tumor tracking with a magnetic resonance imaging-guided linear accelerator (MR-Linac) could address this problem. This study aimed to assess the effects and advantages of MR-Linac in SABR for the treatment of lung tumors.

METHODS

Overall, 41 patients with NSCLC treated with SABR using MR-Linac between March 2019 and December 2021 were included. For comparison, 40 patients treated with SABR using computed tomography-based modalities were also enrolled. The SABR dose ranged from 48 to 60 Gy in 3-5 fractions. The primary endpoint was a lower radiation volume compared to CT-based SABR. The secondary endpoint was the local control rate of SABR using the MR-Linac.

RESULTS

The median follow-up time was 19 months (range: 3-105 months). There was no significant difference in the gross tumor volume between the MR and CT groups (7.1 ± 9.3 cm³ vs 8.0 ± 6.8 cm³, p = 0.643), but the planning target volume was significantly smaller in the MR group (20.8 ± 18.8 cm³ vs 34.1 ± 22.9 cm³, p = 0.005). The 1-year local control rates for the MR and CT groups were 92.1 and 75.4%, respectively (p = 0.07), and the 1-year overall survival rates were 87.4 and 87.0%, respectively (p = 0.30).

CONCLUSION

Lung SABR with MR-Linac can reduce the radiation field without compromising the local control rate. Further follow-up is needed to assess the long-term effects.

Development of AI-driven prediction models to realize real-time tumor tracking during radiotherapy

BACKGROUND

In infrared reflective (IR) marker-based hybrid real-time tumor tracking (RTTT), the internal target position is predicted with the positions of IR markers attached on the patient's body surface using a prediction model. In this work, we developed two artificial intelligence (AI)-driven prediction models to improve RTTT radiotherapy, namely, a convolutional neural network (CNN) and an adaptive neuro-fuzzy inference system (ANFIS) model. The models aim to improve the accuracy in predicting three-dimensional tumor motion.

METHODS

From patients whose respiration-induced motion of the tumor, indicated by the fiducial markers, exceeded 8 mm, 1079 logfiles of IR marker-based hybrid RTTT (IR Tracking) with the gimbal-head radiotherapy system were acquired and randomly divided into two datasets. All the included patients were breathing freely with more than four external IR markers. The historical dataset for the CNN model contained 1003 logfiles, while the remaining 76 logfiles complemented the evaluation dataset. The logfiles recorded the external IR marker positions at a frequency of 60 Hz and fiducial markers as surrogates for the detected target positions every 80-640 ms for 20-40 s. For each logfile in the evaluation dataset, the prediction models were trained based on the data in the first three quarters of the recording period. In the last quarter, the performance of the patient-specific prediction models was tested and evaluated. The overall performance of the AI-driven prediction models was ranked by the percentage of predicted target position within 2 mm of the detected target position. Moreover, the performance of the AI-driven models was compared to a regression prediction model currently implemented in gimbal-head radiotherapy systems.

RESULTS

The percentage of the predicted target position within 2 mm of the detected target position was 95.1%, 92.6% and 85.6% for the CNN, ANFIS, and regression model, respectively. In the evaluation dataset, the CNN, ANFIS, and regression model performed best in 43, 28 and 5 logfiles, respectively.

CONCLUSIONS

The proposed AI-driven prediction models outperformed the regression prediction model, and the overall performance of the CNN model was slightly better than that of the ANFIS model on the evaluation dataset.

Transarterial fiducial marker implantation for CyberKnife radiotherapy to treat pancreatic cancer: an experience with 14 cases

PURPOSE

The purpose of this study was to evaluate the safety and feasibility of transarterial fiducial marker implantation for CyberKnife radiotherapy to treat locally advanced pancreatic cancer.

MATERIALS AND METHODS

Fifteen pancreatic cancer patients were enrolled for transarterial marker implantation. Embolization platinum coils were implanted as a fiducial marker within 20 mm of the cancer edge, and preferably within 3 mm. The technical success of the implantation was defined as implantation of at least one fiducial marker within 20 mm of the target tumor. Irradiation was performed using the CyberKnife system.

RESULTS

For 14 of 15 patients, transarterial implantation was successfully performed, and for 13 of 14 patients, the tracking marker was implanted within 3 mm of the cancer. Tracking instability was observed in two patients, but irradiation was accomplished in all 14 patients. No major complications caused by the implantation procedure were observed. The median overall survival after irradiation was 13.8 months, and the 1- and 2-years survival rates were 62.9% and 32.3%, respectively.

CONCLUSION

Transarterial fiducial marker implantation for pancreatic cancer can be safely performed for tracking, and it will be a valuable alternative approach to percutaneous fiducial marker implantation.

4D in vivo dose verification for real-time tumor tracking treatments using EPID dosimetry

The use of sophisticated techniques such as gating and tracking treatments requires additional quality assurance to mitigate increased patient risks. To address this need, we have developed and validated an in vivo method of dose delivery verification for real-time aperture tracking techniques, using an electronic portal imaging device (EPID)-based, on-treatment patient dose reconstruction and a dynamic anthropomorphic phantom. Using 4DCT scan of the phantom, ten individual treatment plans were created, 1 for each of the 10 separate phases of the respiratory cycle. The 10 MLC apertures were combined into a single dynamic intensity-modulated radiation therapy (IMRT) plan that tracked the tumor motion. The tumor motion and linac delivery were synchronized using an RPM system (Varian Medical Systems) in gating mode with a custom breathing trace. On-treatment EPID frames were captured using a data-acquisition computer with a dedicated frame-grabber. Our in-house EPID-based in vivo dose reconstruction model was modified to reconstruct the 4D accumulated dose distribution for a dynamic MLC (DMLC) tracking plan using the 10-phase 4DCT dataset. Dose estimation accuracy was assessed for the DMLC tracking plan and a single-phase (50% phase) static tumor plan, represented a static field test to verify baseline accuracy. The 3%/3 mm chi-comparison between the EPID-based dose reconstruction for the static tumor delivery and the TPS dose calculation for the static plan resulted in 100% pass rate for planning target volume (PTV) voxels while the mean percentage dose difference was 0.6%. Comparing the EPID-based dose reconstruction for the DMLC tracking to the TPS calculation for the static plan gave a 3%/3 mm chi pass rate of 99.3% for PTV voxels and a mean percentage dose difference of 1.1%. While further work is required to assess the accuracy of this approach in more clinically relevant situations, we have established clinical feasibility and baseline accuracy of using the transmission EPID-based, in vivo patient dose verification for MLC-tracking treatments.

respiTrack: Patient-specific real-time respiratory tumor motion prediction using magnetic tracking

Purpose

An intraoperative real-time respiratory tumor motion prediction system with magnetic tracking technology is presented. Based on respiratory movements in different body regions, it provides patient and single/multiple tumor-specific prediction that facilitates the guiding of treatments.

Methods

A custom-built phantom patient model replicates the respiratory cycles similar to a human body, while the custom-built sensor holder concept is applied on the patient’s surface to find optimum sensor number and their best possible placement locations to use in real-time surgical navigation and motion prediction of internal tumors. Automatic marker localization applied to patient’s 4D-CT data, feature selection and Gaussian process regression algorithms enable off-line prediction in the preoperative phase to increase the accuracy of real-time prediction.

Results

Two evaluation methods with three different registration patterns (at fully/half inhaled and fully exhaled positions) were used quantitatively at all internal target positions in phantom: The statical method evaluates the accuracy by stopping simulated breathing and dynamic with continued breathing patterns. The overall root mean square error (RMS) for both methods was between \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0.32\pm 0.06~\hbox {mm}$$\end{document}0.32±0.06mm and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$3.71\pm 0.79~\hbox {mm}$$\end{document}3.71±0.79mm. The overall registration RMS error was \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0.6\pm 0.4~\hbox {mm}$$\end{document}0.6±0.4mm. The best prediction errors were observed by registrations at half inhaled positions with minimum \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0.27\pm 0.02~\hbox {mm}$$\end{document}0.27±0.02mm, maximum \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2.90\pm 0.72~\hbox {mm}$$\end{document}2.90±0.72mm. The resulting accuracy satisfies most radiotherapy treatments or surgeries, e.g., for lung, liver, prostate and spine.

Conclusion

The built system is proposed to predict respiratory motions of internal structures in the body while the patient is breathing freely during treatment. The custom-built sensor holders are compatible with magnetic tracking. Our presented approach reduces known technological and human limitations of commonly used methods for physicians and patients.

Improving the intra-fraction update efficiency of a correlation model used for internal motion estimation during real-time tumor tracking for SBRT patients: fast update or no update?

BACKGROUND AND PURPOSE

For tumor tracking, a correlation model is used to estimate internal tumor position based on external surrogate motion. When patients experience an internal/external surrogate drift, an update of the correlation model is required to continue tumor tracking. In this study, the accuracy of the internal tumor position estimation for both the clinical available update at discrete points in time (rebuild) and an in-house developed non-clinical online update approach was investigated.

METHODS

A dynamic phantom with superimposed baseline drifts and 14 SBRT patients, treated with real-time tumor tracking (RTTT) on the Vero system, were retrospectively simulated for three update scenarios, respectively no update, clinical rebuild and 0.5 Hz automated online update of the correlation model. By comparing the target positions based on 0.5 Hz verification X-ray images with the estimated internal tumor positions regarding all three update scenarios, 95th percentile modeling errors (ME95), incidences of full geometrical coverage of the CTV by a 5 mm extended PTV (P₅mm) and population-based PTV margins were calculated. Further, the treatment time reduction was estimated when switching from the clinical rebuild approach to the online correlation model update.

RESULTS

For dynamic phantom motion with baseline drifts up to 0.4 mm/min, a 0.5 Hz intra-fraction update showed a similar accuracy in terms of ME95 and P5 mm compared to clinical rebuild. For SBRT patients treated on Vero with RTTT, accuracy was improved by 0.5 Hz online update compared to the clinical rebuild protocol, yielding smaller PTV margins (from 3.2 mm to 2.7 mm), reduced ME95,3D (from 4.1 mm to 3.4 mm) and an increased 5th percentile P5 mm (from 90.7% to 96.1%) for the entire patient group. Further, 80% of treatment sessions were reduced in time with on average 5.5 ± 4.1(1 SD)min.

CONCLUSION

With a fast (0.5 Hz) automated online update of the correlation model, an efficient RTTT workflow with improved geometrical accuracy was obtained.