Multiple regions-of-interest analysis of setup uncertainties for head-and-neck cancer radiotherapy

An Unsupervised Learning-Based Regional Deformable Model for Automated Multi-Organ Contour Propagation

The aim of this study is to evaluate a regional deformable model based on a deep unsupervised learning model for automatic contour propagation in breast cone-beam computed tomography-guided adaptive radiation therapy. A deep unsupervised learning model was introduced to map breast's tumor bed, clinical target volume, heart, left lung, right lung, and spinal cord from planning computed tomography to cone-beam CT. To improve the traditional image registration method's performance, we used a regional deformable framework based on the narrow-band mapping, which can mitigate the effect of the image artifacts on the cone-beam CT. We retrospectively selected 373 anonymized cone-beam CT volumes from 111 patients with breast cancer. The cone-beam CTs are divided into three sets. 311 / 20 / 42 cone-beam CT images were used for training, validating, and testing. The manual contour was used as reference for the testing set. We compared the results between the reference and the model prediction for evaluating the performance. The mean Dice between manual reference segmentations and the model predicted segmentations for breast tumor bed, clinical target volume, heart, left lung, right lung, and spinal cord were 0.78 ± 0.09, 0.90 ± 0.03, 0.88 ± 0.04, 0.94 ± 0.03, 0.95 ± 0.02, and 0.77 ± 0.07, respectively. The results demonstrated a good agreement between the reference and the proposed contours. The proposed deep learning-based regional deformable model technique can automatically propagate contours for breast cancer adaptive radiotherapy. Deep learning in contour propagation was promising, but further investigation was warranted.

The application of gradient dose segmented analysis of in-vivo EPID images for patients undergoing VMAT in a resource-constrained environment

The gamma analysis metric is a commonly used metric for VMAT plan evaluation. The major drawback of this is the lack of correlation between gamma passing rates and DVH values. The novel GDSA_mean metric was developed by Steers et al. to quantify changes in the PTV mean dose (D_mean ) for VMAT patients. The aim of this work is to apply the GDSA retrospectively on head-and-neck cancer patients treated on the newly acquired Varian Halcyon, to assess changes in GDSA_mean , and to evaluate the cause of day-to-day changes in the time-plot series. In-vivo EPID transmission images of head-and-neck cancer patients treated between August 2019 and July 2020 were analyzed retrospectively. The GDSA_mean was determined for all patients treated. The changes in patient anatomy and rotational errors were quantified using the daily CBCT images and added to a time-plot with the daily change in GDSA_mean . Over 97% of the delivered treatment fractions had a GDSA_mean  < 3%. Thirteen of the patients received at least one treatment fraction where the GDSA_mean  > 3%. Most of these deviations occurred for the later fractions of radiotherapy treatment. Additionally, 92% of these patients were treated for malignancies involving the larynx and oropharynx. Notable deviations in the effective separation diameters were observed for 62% of the patients where the change in GDSA_mean  > 3%. For the other five cases with GDSA_mean  < 3%, the mean pitch, roll, and yaw rotational errors were 0.90°, 0.45°, and 0.43°, respectively. A GDSA_mean  > 3% was more likely due to a change in separation, whereas a GDSA_mean  < 3% was likely caused by rotational errors. Pitch errors were shown to be the most dominant. The GDSA_mean is easily implementable and can aid in scheduling new CT scans for patients before significant deviations in dose delivery occur.

Introduce a rotational robust optimization framework for spot-scanning proton arc (SPArc) therapy

Objective. Proton dosimetric uncertainties resulting from the patient's daily setup errors in rotational directions exist even with advanced image-guided radiotherapy techniques. Thus, we developed a new rotational robust optimization SPArc algorithm (SPArc_rot) to mitigate the dosimetric impact of the rotational setup error in Raystation ver. 6.02 (RaySearch Laboratory AB, Stockholm, Sweden).Approach.The initial planning CT was rotated ±5° simulating the worst-case setup error in the roll direction. The SPArc_rotuses a multi-CT robust optimization framework by taking into account of such rotational setup errors. Five cases representing different disease sites were evaluated. Both SPArc_originaland SPArc_rotplans were generated using the same translational robust optimized parameters. To quantitatively investigate the mitigation effect from the rotational setup errors, all plans were recalculated using a series of pseudo-CT with rotational setup error (±1°/±2°/±3°/±5°). Dosimetric metrics such as D98% of CTV, and 3D gamma analysis were used to assess the dose distribution changes in the target and OARs.Main results.The magnitudes of dosimetric changes in the targets due to rotational setup error were significantly reduced by the SPArc_rotcompared to SPArc in all cases. The uncertainties of the max dose to the OARs, such as brainstem, spinal cord and esophagus were significantly reduced using SPArc_rot. The uncertainties of the mean dose to the OARs such as liver and oral cavity, parotid were comparable between the two planning techniques. The gamma passing rate (3%/3 mm) was significantly improved for CTV of all tumor sites through SPArc_rot.Significance.Rotational setup error is one of the major issues which could lead to significant dose perturbations. SPArc_rotplanning approach can consider such rotational error from patient setup or gantry rotation error by effectively mitigating the dose uncertainties to the target and in the adjunct series OARs.

QUANTIFICATION OF 6D INTER-FRACTION TUMOUR LOCALISATION ERRORS IN TONGUE AND PROSTATE CANCER USING DAILY KV-CBCT FOR 1000 IMRT AND VMAT TREATMENT FRACTIONS

This study aimed to evaluate the 6D inter-fraction tumour localisation errors in 20 tongue and 20 prostate cancer patients treated with intensity-modulated radiation therapy and volumetric-modulated arc therapy. The patient tumour localisation errors in lateral, longitudinal and vertical translation axes and pitch, roll and yaw rotational axes were analysed by automatic image registration of daily pretreatment kilovoltage cone-beam computed tomography (kV-CBCT) with planning CT in 1000 fractions. The overall mean error (M), systematic error (Σ), random error (σ) and planning target volume (PTV) margins were evaluated. The frequency distributions of setup errors were normally distributed about the mean except for pitch in the tongue and prostate. The overall 3D vector length ≥ 5 mm was 14.2 and 49.8% in the ca-tongue and ca-prostate, respectively. The frequency of rotational errors ≥1 degree was a maximum of 37 and 59.5%, respectively, in ca-tongue and ca-prostate. The M, Σ and σ for all translational and rotational axes decreased with increasing frequency of verification correction in ca-tongue and ca-prostate patients. Similarly, the PTV margin was reduced with no correction to alternate day correction from a maximum of 4.7 to 2.5 mm in ca-tongue and from a maximum of 8.6 to 4.7 mm in ca-prostate. The results emphasised the vital role of the higher frequency of kV-CBCT based setup correction in reducing M, Σ, σ and PTV margins in ca-tongue and ca-prostate patients.

Novel multi jet fusion 3D-printed patient immobilization for radiation therapy

PURPOSE

Thermoplastic immobilizers are used routinely in radiation therapy to achieve positioning accuracy. These devices are variable in quality as they are dependent on the skill of the human fabricator. We examine the potential multi jet fusion (MJF) 3D printing for the production immobilizers with a focus on the surface dosimetry of several MJF-printed PA12-based material candidates. Materials are compared with the goal of minimizing surface dose with comparison to standard thermoplastic. We introduce a novel metamaterial design for the shell of the immobilizer, with the aims of mechanical robustness and low-dose buildup. We demonstrate first examples of adult and pediatric cranial and head-and-neck immobilizers.

METHODS

Three different PA12 materials were examined and compared to fused deposition modeling-printed polylactic acid (PLA), PLA with density lowered by adding hollow glass microspheres, and to perforated or perforated/stretched and solid status quo thermoplastic samples. Build-up dose measurements were made using a parallel plate chamber. A metamaterial design was established based on a packed hexagonal geometry. Radiochromic film dosimetry was performed to determine the dependence of surface dose on the metamaterial design. Full cranial and head-and-neck prototype immobilizers were designed, printed, and assessed with regard to dimensional accuracy.

RESULTS

Build-up dose measurements demonstrated the superiority of the PA12 material with a light fusing agent, which yielded a ∼15% dose reduction compared to other MJF materials. Metamaterial samples provided dose reductions ranging from 11% to 40% compared to stretched thermoplastic. MJF-printed immobilizers were produced reliably, demonstrated the versatility of digital design, and showed dimensional accuracy with 97% of sampled points within ±2 mm.

CONCLUSIONS

MJF is a promising technology for an automated fabrication of patient immobilizers. Material selection and metamaterial design can be leveraged to yield surface dose reduction of up to 40%. Immobilizer design is highly customizable, and the first examples of MJF-printed immobilizers demonstrate excellent dimensional accuracy.

Development of a customisable 3D-printed intra-oral stent for head-and-neck radiotherapy

•

Advanced radiotherapy techniques have improved head-and-neck treatments.

•

More improvements are possible with intra-oral stents stabilising sensitive anatomy.

•

MRI imaging shows new modular 3D printed stents provide stable displacement.

•

Modular stents achieve positive outcomes within standard treatment workflow.

Intra-oral stents (including mouth-pieces and bite blocks) can be used to displace adjacent non-involved oral tissue and reduce radiation side effects from radiotherapy treatments for head-and-neck cancer. In this study, a modular and customisable 3D printed intra-oral stent was designed, fabricated and evaluated, to utilise the advantages of the 3D printing process without the interruption of clinical workflow associated with printing time. The stent design used a central mouth-opening and tongue-depressing main piece, with optional cheek displacement pieces in three different sizes, plus an anchor point for moulding silicone to fit individual patients’ teeth. A magnetic resonance imaging (MRI) study of one healthy participant demonstrated the tissue displacement effects of the stent, while providing a best-case indication of its comfort.

Retrospective Clinical Evaluation of a Decision-Support Software for Adaptive Radiotherapy of Head and Neck Cancer Patients

Purpose

This study aimed to evaluate the clinical need for an automated decision-support software platform for adaptive radiation therapy (ART) of head and neck cancer (HNC) patients.

Methods

We tested RTapp (SegAna), a new ART software platform for deciding when a treatment replan is needed, to investigate a set of 27 HNC patients’ data retrospectively. For each fraction, the software estimated key components of ART such as daily dose distribution and cumulative doses received by targets and organs at risk (OARs) from daily 3D imaging in real-time. RTapp also included a prediction algorithm that analyzed dosimetric parameter (DP) trends against user-specified thresholds to proactively trigger adaptive re-planning up to four fractions ahead. The DPs evaluated for ART were based on treatment planning dose constraints. Warning (V₉₅<95%) and adaptation (V₉₅<93%) thresholds were set for PTVs, while OAR adaptation dosimetric endpoints of +10% (DE₁₀) were set for all D_max and D_mean DPs. Any threshold violation at end of treatment (EOT) triggered a review of the DP trends to determine the threshold-crossing fraction Fx when the violations occurred. The prediction model accuracy was determined as the difference between calculated and predicted DP values with 95% confidence intervals (CI₉₅).

Results

RTapp was able to address the needs of treatment adaptation. Specifically, we identified 18/27 studies (67%) for violating PTV coverage or parotid D_mean at EOT. Twelve PTVs had V₉₅<95% (mean coverage decrease of −6.8 ± 2.9%) including six flagged for adaptation at median Fx= 6 (range, 1–16). Seventeen parotids were flagged for exceeding D_mean dose constraints with a median increase of +2.60 Gy (range, 0.99–6.31 Gy) at EOT, including nine with DP>DE₁₀. The differences between predicted and calculated PTV V₉₅ and parotid D_mean was up to 7.6% (mean ± CI₉₅, −2.7 ± 4.1%) and 5 Gy (mean ± CI₉₅, 0.3 ± 1.6 Gy), respectively. The most accurate predictions were obtained closest to the threshold-crossing fraction. For parotids, the results showed that Fx ranged between fractions 1 and 23, with a lack of specific trend demonstrating that the need for treatment adaptation may be verified for every fraction.

Conclusion

Integrated in an ART clinical workflow, RTapp aids in predicting whether specific treatment would require adaptation up to four fractions ahead of time.

Human-level comparable control volume mapping with a deep unsupervised-learning model for image-guided radiation therapy

PURPOSE

To develop a deep unsupervised learning method with control volume (CV) mapping from patient positioning daily CT (dCT) to planning computed tomography (pCT) for precise patient positioning.

METHODS

We propose an unsupervised learning framework, which maps CVs from dCT to pCT to automatically generate the couch shifts, including translation and rotation dimensions. The network inputs are dCT, pCT and CV positions in the pCT. The output is the transformation parameter of the dCT used to setup the head and neck cancer (HNC) patients. The network is trained to maximize image similarity between the CV in the pCT and the CV in the dCT. A total of 554 CT scans from 158 HNC patients were used for the evaluation of the proposed model. At different points in time, each patient had many CT scans. Couch shifts are calculated for the testing by averaging the translation and rotation from the CVs. The ground-truth of the shifts come from bone landmarks determined by an experienced radiation oncologist.

RESULTS

The system positioning errors of translation and rotation are less than 0.47 mm and 0.17°, respectively. The random positioning errors of translation and rotation are less than 1.13 mm and 0.29°, respectively. The proposed method enhanced the proportion of cases registered within a preset tolerance (2.0 mm/1.0°) from 66.67% to 90.91% as compared to standard registrations.

CONCLUSIONS

We proposed a deep unsupervised learning architecture for patient positioning with inclusion of CVs mapping, which weights the CVs regions differently to mitigate any potential adverse influence of image artifacts on the registration. Our experimental results show that the proposed method achieved efficient and effective HNC patient positioning.

Management and work-up procedures of patients with head and neck malignancies treated by radiation

Radiotherapy alone or in association with systemic treatment plays a major role in the treatment of head and neck tumours, either as a primary treatment or as a postoperative modality. The management of these tumours is multidisciplinary, requiring particular care at every treatment step. We present the update of the recommendations of the French Society of Radiation Oncology on the radiotherapy of head and neck tumours from the imaging work-up needed for optimal selection of treatment volume, to optimization of the dose distribution and delivery.

Automated contour propagation of the prostate from pCT to CBCT images via deep unsupervised learning

PURPOSE

To develop and evaluate a deep unsupervised learning (DUL) framework based on a regional deformable model for automated prostate contour propagation from planning computed tomography (pCT) to cone-beam CT (CBCT).

METHODS

We introduce a DUL model to map the prostate contour from pCT to on-treatment CBCT. The DUL framework used a regional deformable model via narrow-band mapping to augment the conventional strategy. Two hundred and fifty-one anonymized CBCT images from prostate cancer patients were retrospectively selected and divided into three sets: 180 were used for training, 12 for validation, and 59 for testing. The testing dataset was divided into two groups. Group 1 contained 50 CBCT volumes, with one physician-generated prostate contour on CBCT image. Group 2 contained nine CBCT images, each including prostate contours delineated by four independent physicians and a consensus contour generated using the STAPLE method. Results were compared between the proposed DUL and physician-generated contours through the Dice similarity coefficients (DSCs), the Hausdorff distances, and the distances of the center-of-mass.

RESULTS

The average DSCs between DUL-based prostate contours and reference contours for test data in group 1 and group 2 consensus were 0.83 ± 0.04, and 0.85 ± 0.04, respectively. Correspondingly, the mean center-of-mass distances were 3.52 mm ± 1.15 mm, and 2.98 mm ± 1.42 mm, respectively.

CONCLUSIONS

This novel DUL technique can automatically propagate the contour of the prostate from pCT to CBCT. The proposed method shows that highly accurate contour propagation for CBCT-guided adaptive radiotherapy is achievable via the deep learning technique.

Development of a Tongue Immobilization Device Using a 3D Printer for Intensity Modulated Radiation Therapy of Nasopharyngeal Cancer Patients

Purpose

This study aimed to reduce radiation doses to the tongue, a patient-specific semi-customized tongue immobilization device (SCTID) was developed using a 3D printer for helical tomotherapy (HT) of nasopharyngeal cancer (NPCa). Dosimetric characteristics and setup stability of the SCTID were compared with those of a standard mouthpiece (SMP).

Materials and Methods

For displacement and robust immobilization of the tongue, the SCTID consists of four parts: upper and lower tooth stoppers, tongue guider, tongue-tip position guide bar, and connectors. With the SCTID and SMP, two sets of planning computed tomography and HT plans were obtained for 10 NPCa patients. Dosimetric and geometric characteristics were compared. Position reproducibility of the tongue with SCTID was evaluated by comparing with planned dose and adaptive accumulated dose of the tongue and base of the tongue based on daily setup mega-voltage computed tomography.

Results

Using the SCTID, the tongue was effectively displaced from the planning target volume compared to the SMP. The median mucosa of the tongue (M-tongue) dose was significantly reduced (20.7 Gy vs. 27.8 Gy). The volumes of the M-tongue receiving a dose of 15 Gy, 30 Gy, and 45 Gy and the volumes of the mucosa of oral cavity and oropharynx (M-OC/OP) receiving a dose of 45 Gy and 60 Gy were significantly lower than using the SMP. No significant differences was observed between the planned dose and the accumulated adaptive dose in any dosimetric characteristics of the tongue and base of tongue.

Conclusion

SCTID can not only reduce the dose to the M-tongue and M-OC/OP dramatically, when compared to SMP, but also provide excellent reproducibility and easy visual verification.

Evaluation of set-up errors and determination of set-up margin in pelvic radiotherapy by electronic portal imaging device (EPID)

Introduction and purpose:

The error in set-up of patients is an inherent part of treatment processes. The positioning errors can be used to determine the margins of the planning target volume (PTV) to cover the target volume, while minimising the radiation dose delivered to normal tissues. This study aimed to evaluate random and systematic errors occurring in inter-fraction set-ups of pelvic radiotherapy measured by electronic portal imaging device (EPID) and then to propose the optimum clinical target volume (CTV) to PTV margin in pelvic cancer patients.

Materials and methods:

This study examined 22 patients treated with pelvic radiotherapy. A total of 182 portal images were evaluated. Population random (σ) and systematic (Σ) errors were determined based on the portal images in three directions (X, Y and Z). The set-up margin for CTV to PTV was calculated by published margin formulae of International Commission on Radiation Units and measurements (ICRU) report No. 62 recommendation and formulas presented by Stroom and Heijmen and Van Herk et al.

Results:

Systematic set-up errors for radiotherapy to patients ranged between 2·36 and 4·99 mm, and random errors ranged between 1·51 and 2·74 mm. The margin required to cover the target volume retrospectively was calculated based on ICRU 62 and formulas presented by Stroom and Heijmen and Van Herk et al. were used to calculate the range 2·8–5·7 mm, 5·7–11·9 mm and 6·9–14·4 mm, respectively.

Conclusion:

According to our findings, it can be concluded that by extending the CTV margin by 6·9–14·4 mm, we can ensure that 90% of the pelvic cancer patients will receive ≥ 95% of the prescribed dose in the CTV area.

Repositioning accuracy of a novel thermoplastic mask for head and neck cancer radiotherapy

PURPOSE

The aim of this study was to assess the reproducibility of patient shoulder position immobilized with a novel and innovative prototype mask (E-Frame, Engineering System).

METHODS

The E-frame mask fixes both shoulders and bisaxillary regions compared with that of a commercial mask (Type-S, CIVCO). Thirteen and twelve patients were immobilized with the Type-S and E-Frame mask systems, respectively. For each treatment fraction, cone-beam CT (CBCT) images of the patient were acquired and retrospectively analyzed. The CBCT images were registered to the planning CT based on the cervical spine, and then the displacements of the acromial extremity of the clavicle were measured.

RESULTS

The systematic and random errors between the two mask systems were evaluated. The differences of the systematic errors between the two mask systems were not statistically significant. The mean random errors in the three directions (AP, SI and LR) were 2.7 mm, 3.1 mm and 1.5 mm, respectively for the Type-S mask, and 2.8 mm 2.5 mm and 1.4 mm, respectively for the E-Frame mask. The random error of the E-Frame masks in the SI direction was significantly smaller than that of the Type-S. The number of cases showing displacements exceeding 10 mm in the SI direction for at least one fraction was eight (61% of 13 cases) and three (25% of 12 cases) for Type-S and E-Frame masks, respectively.

CONCLUSIONS

The E-Frame masks reduced the random displacements of patient's shoulders in the SI direction, effectively preventing large shoulder shifts that occurred frequently with Type-S masks.

[Adaptive radiotherapy for nasopharyngeal carcinomas: Where are we?]

Modern high-precision radiotherapy techniques have recently incorporated the notion of anatomical variations of the patient during treatment and have tried to adapt the treatment planning to them. Adaptive radiotherapy for nasopharyngeal tumors is starting to prove its benefit nowadays. His interest is constantly being evaluated. The variations encountered during the treatment are both geometric and dosimetric. They are represented by a reduction in the macroscopic tumors volume, a change in its position and a consequent dosimetric impact. The changes also concern organs at risk with a reduction of glandular structure volumes, and a different position which increases their doses. Delivered doses to noble structures (brainstem and spinal cord) may also increase. However, difficulties are encountered in its realization. There is a problem to perfectly reproduce the patient position during the second acquisition, which impacts the fusion quality between the two CT scans. This generates an imprecision in the definition of the same treatment isocentre on the second scanner. Also, there is a difficulty in accumulated doses calculation. The indication of adaptive radiotherapy remains a subject of controversy. It should be proposed for a subgroup of patients who could benefit from this new strategy. We present here an update on the state of the art of adaptive radiotherapy for nasopharyngeal cancer.

A Fast Online Replanning Algorithm Based on Intensity Field Projection for Adaptive Radiotherapy

Purpose: The purpose of this work was to propose an online replanning algorithm, named intensity field projection (IFP), that directly adjusts intensity distributions for each beam based on the deformation of structures. IFP can be implemented within a reasonably acceptable time frame. Methods and Materials: The online replanning method is based on the gradient-based free form deformation (GFFD) algorithm, which we have previously proposed. The method involves the following steps: The planning computed tomography (CT) and cone-beam CT image are registered to generate a three-dimensional (3-D) deformation field. According to the 3-D deformation field, the registered image and a new delineation are generated. The two-dimensional (2-D) deformation field of ray intensity in each beam direction is determined based on the 3-D deformation field in the region of interest. The 2-D ray intensity distribution in the corresponding beam direction is deformed to generate a new 2-D ray intensity distribution. According to the new 2-D ray intensity distribution, corresponding multi-leaf collimator (MLC), and jaw motion data are generated. The feasibility and advantages of our method have been demonstrated in 20 lung cancer intensity modulated radiation therapy (IMRT) cases. Results: Substantial underdosing in the CTV is seen in the original and the repositioning plans. The average prescription dose coverage (V100%) and D95 for CTV were 100% and 60.3 Gy for the IFP plans compared to 82.6% (P < 0.01) and 44.0 Gy (P < 0.01) for original plans, 86.7% (P < 0.01), and 58.5 Gy (P < 0.01) for repositioning plans. On average, the mean total lung doses were 12.2 Gy for the IFP plan compared to the 12.4 Gy (P < 0.01) and 12.6 Gy (P < 0.01) for the original and the repositioning plans. The entire process of IFP can be completed within 3 min. Conclusions: We proposed an online replanning strategy for automatically correcting interfractional anatomy variations. The preliminary results indicate that the IFP method substantially increased planning speed for online adaptive replanning.

Prospective observational study to estimate set-up errors and optimise PTV margins in patients undergoing IMRT for head and neck cancers from a Government cancer centre of Eastern India

Background:

The head and neck cancers as a whole are the most common cancers among males in India. Technological advancements have led to an improvement in radiation therapy (RT) techniques with subsequent reduction in normal tissue complications. To correct patient set-up errors, an off-line correction method like no action level (NAL) protocol may be used as a preferred protocol particularly for a busy department. The objectives of the study were to measure the translational set-up errors using kV cone-beam computed tomography (CBCT) in patients undergoing intensity modulated radiotherapy (IMRT) in head and neck cancers and also to optimise clinical target volume (CTV) to planning target volume (PTV) margin using NAL protocol.

Material and methods:

On the first 5 days of RT, patient’s position was verified by kV-CBCT and then weekly during the course of treatment. The comparison between the reference and kV-CBCT images was performed, and the shifts measured and recorded. The mean error from the initial five consecutive fractions was corrected on the sixth daily fraction. Displacements in all the directions were measured. The population systematic and random errors were determined and used to estimate PTV margins according to the van Herk formula.

Results:

A total of 322 images were analysed. Before correction, 15, 12 and 9% patients had systematic error ≥3 mm on X, Y and Z axes, but after correction this was reduced to 9, 0 and 0%. The total percentage of patients whose set-up margin was ≥5 mm before correction was 5, 6·25, 3·75%, but after correction it reduced to 1·88, 0, and 0·63%. The margins of total population were reduced to 63, 65 and 56% after correction on X, Y and Z axes, respectively.

Conclusion:

A simple off-line NAL protocol can correct the set-up errors without daily on-line imaging in patients undergoing IMRT and hence acting as a resource sparing alternative. Five millimetre margin to CTVs was adequate and safe to overcome the problem of set-up errors in head and neck IMRT.

A spatial neighborhood methodology for computing and analyzing lymph node carcinoma similarity in precision medicine

Precision medicine seeks to tailor therapy to the individual patient, based on statistical correlates from patients who are similar to the one under consideration. These correlates can and should go beyond genetics, and in general, beyond tabular or array data that can be easily represented computationally and compared. For example, in many types of cancer, cancer treatment and toxicity depend in large measure on the spatial disease spread-e.g., metastasizes to regional lymph nodes in head and neck cancer. However, there is currently a lack of methodology for integrating spatial information when considering patient similarity. We present a novel modeling methodology for the comparison of cancer patients within a cohort, based on the spatial spread of the lymph nodes affected in each patient. The method uses a topological map, bigrams, and hierarchical clustering to group patients based on their similarity. We compare this approach against a nonspatial (categorical) similarity approach where patients are binned solely by their affected nodes. We present similarity results on a 582 head and neck cancer patient cohort, along with two visual abstractions for analysis of the results, and we present clinician feedback. Our novel methodology partitions a patient cohort into clinically meaningful groups more susceptible to treatment side-effects. Such spatially-aware similarity approaches can help maximize the effectiveness of each patient's treatment.

Estimating PTV Margins in Head and Neck Stereotactic Ablative Radiation Therapy (SABR) Through Target Site Analysis of Positioning and Intrafractional Accuracy

PURPOSE

Recurrent or previously irradiated head and neck cancers (HNC) are therapeutically challenging and may benefit from high-dose, highly accurate radiation techniques, such as stereotactic ablative radiation therapy (SABR). Here, we compare set-up and positioning accuracy across HNC subsites to further optimize the treatment process and planning target volume (PTV) margin recommendations for head and neck SABR.

METHODS AND MATERIALS

We prospectively collected data on 405 treatment fractions across 79 patients treated with SABR for recurrent/previously irradiated HNC. First, interfractional error was determined by comparing ExacTrac x-ray to the treatment plan. Patients were then shifted and residual error was measured with repeat x-ray. Next, cone beam computed tomography (CBCT) was compared with ExacTrac for positioning agreement, and final shifts were applied. Lastly, intrafractional error was measured with x-ray before each arc. Results were stratified by treatment site into skull base, neck/parotid, and mucosal.

RESULTS

Most patients (66.7%) were treated to 45 Gy in 5 fractions (range, 21-47.5 Gy in 3-5 fractions). The initial mean ± standard deviation interfractional errors were -0.2 ± 1.4 mm (anteroposterior), 0.2 ± 1.8 mm (craniocaudal), and -0.1 ± 1.7 mm (left-right). Interfractional 3-dimensional vector error was 2.48 ± 1.44, with skull base significantly lower than other sites (2.22 vs 2.77; P = .0016). All interfractional errors were corrected to within 1.3 mm and 1.8°. CBCT agreed with ExacTrac to within 3.6 mm and 3.4°. CBCT disagreements and intrafractional errors of >1 mm or >1° occurred at significantly lower rates in skull base sites (CBCT: 16.4% vs 50.0% neck, 52.0% mucosal, P < .0001; intrafractional: 22.0% vs 48.7% all others, P < .0001). Final PTVs were 1.5 mm (skull base), 2.0 mm (neck/parotid), and 1.8 mm (mucosal).

CONCLUSIONS

Head and neck SABR PTV margins should be optimized by target site. PTV margins of 1.5 to 2 mm may be sufficient in the skull base, whereas 2 to 2.5 mm may be necessary for neck and mucosal targets. When using ExacTrac, skull base sites show significantly fewer uncertainties throughout the treatment process, but neck/mucosal targets may require the addition of CBCT to account for positioning errors and internal organ motion.

Radiotherapy-Compatible Robotic System for Multi-Landmark Positioning in Head and Neck Cancer Treatments

The spine flexibility creates one of the most significant challenges to proper positioning in radiation therapy of head and neck cancers. Even though existing immobilization techniques can reduce the positioning uncertainty, residual errors (2–3 mm along the cervical spine) cannot be mitigated by single translation-based approaches. Here, we introduce a fully radiotherapy-compatible electro-mechanical robotic system, capable of positioning a patient’s head with submillimeter accuracy in clinically acceptable spatial constraints. Key mechanical components, designed by finite element analysis, are fabricated with 3D printing and a cyclic loading test of the printed materials captures a great mechanical robustness. Measured attenuation of most printed components is lower than analytic estimations and radiographic imaging shows no visible artifacts, implying full radio-compatibility. The new system evaluates the positioning accuracy with an anthropomorphic skeletal phantom and optical tracking system, which shows a minimal residual error (0.7 ± 0.3 mm). This device also offers an accurate assessment of the post correction error of aligning individual regions when the head and body are individually positioned. Collectively, the radiotherapy-compatible robotic system enables multi-landmark setup to align the head and body independently and accurately for radiation treatment, which will significantly reduce the need for large margins in the lower neck.

Assessment of setup uncertainty in hypofractionated liver radiation therapy with a breath-hold technique using automatic image registration-based image guidance

Background

Target localization in radiation therapy is affected by numerous sources of uncertainty. Despite measures to minimize the breathing motion, the treatment of hypofractionated liver radiation therapy is further challenged by residual uncertainty coming from involuntary organ motion and daily changes in the shape and location of abdominal organs. To address the residual uncertainty, clinics implement image-guided radiation therapy at varying levels of soft-tissue contrast. This study utilized the treatment records from the patients that have received hypofractionated liver radiation therapy using in-room computed tomography (CT) imaging to assess the setup uncertainty and to estimate the appropriate planning treatment volume (PTV) margins in the absence of in-room CT imaging.

Methods

We collected 917 pre-treatment daily in-room CT images from 69 patients who received hypofractionated radiation therapy to the liver with the inspiration breath-hold technique. For each treatment, the daily CT was initially aligned to the planning CT based on the shape of the liver automatically using a CT-CT alignment software. After the initial alignment, manual shift corrections were determined by visual inspection of the two images, and the corrections were applied to shift the patient to the physician-approved treatment position. Considering the final alignment as the gold-standard setup, systematic and random uncertainties in the automatic alignment were quantified, and the uncertainties were used to calculate the PTV margins.

Results

The median discrepancy between the final and automatic alignment was 1.1 mm (0–24.3 mm), and 38% of treated fractions required manual corrections of ≥3 mm. The systematic uncertainty was 1.5 mm in the anterior-posterior (AP) direction, 1.1 mm in the left-right (LR) direction, and 2.4 mm in the superior-inferior (SI) direction. The random uncertainty was 2.2 mm in the AP, 1.9 mm in the LR, and 2.2 mm in the SI direction. The PTV margins recommended to be used in the absence of in-room CT imaging were 5.3 mm in the AP, 3.5 mm in the LR, and 5.1 mm in the SI direction.

Conclusions

Manual shift correction based on soft-tissue alignment is substantial in the treatment of the abdominal region. In-room CT can reduce PTV margin by up to 5 mm, which may be especially beneficial for dose escalation and normal tissue sparing in hypofractionated liver radiation therapy.

Assessment of the therapeutic accuracy of cone beam computed tomography-guided nasopharyngeal carcinoma radiotherapy

The aim of the present study was to determine the ability of cone beam computed tomography (CBCT) to improve the accuracy of nasopharyngeal carcinoma (NPC) radiotherapy by analyzing the setup and inter-fraction errors at different levels and directions of the target volumes. A total of 113 patients with NPC who were undergoing intensity-modulated radiotherapy were recruited for the present study. Each patient had at least three CBCT exams prior to the start of radiation therapy. Three anatomic bony landmarks, including the upper neck, lower neck and head, were used to represent the different levels of assessment. The positioning errors were registered in three planes throughout the course of radiotherapy: The right-left (RL), superior-inferior (SI) and anterior-posterior (AP) directions. The planning CT images were matched with the CBCT images to determine the naso-pharynx shifts. A receiver operating characteristic curve was plotted to establish the specificity and sensitivity of CBCT. The planning target volume margin (MPTV) for the head was 0.9 mm, 1.4 mm for the upper neck and 2.0 mm for the lower neck. MPTVs of 1.5, 0.6 and 2.2 mm in the RL, SI and AP directions, respectively, were detected. In addition, there was evidence of setup errors in the three planes (RL, SI and AP) with the greatest error observed in the AP direction. Furthermore, the setup uncertainties in the neck region were greater than those of the head. In conclusion, CBCT could greatly improve the accuracy of radiotherapy by minimizing the setup errors and MPTV.

Development of twist‐correction system for radiotherapy of head and neck cancer patients

To propose a concept for correcting the twist between the head and neck and the body frequently occurring in radiotherapy patients and to develop a prototype device for achieving this. Furthermore, the operational accuracy of this device under no load was evaluated. We devised a concept for correcting the twist of patients by adjustment of the three rotation (pitch, roll, and yaw) angles in two independent plates connected by a joint (fulcrum). The two plates (head and neck plate and body plate) rotate around the fulcrum by adjusting screws under each of them. A prototype device was created to materialize this concept. First, after all adjusting screws were set to the zero position, the rotation angle of each plate was measured by a digital goniometer. Repeatability was evaluated by performing 20 repeated measurements. Next, to confirm the rotational accuracy of each plate of the prototype device, the calculated rotation angles for 20 combinations of patterns of traveled distances of the adjusting screws were compared with those measured by the digital goniometer and cone‐beam computed tomography (CT). The repeatability (standard deviation: SD) of the pitch, roll, and yaw angles of the head and neck plate was 0.04°, 0.05°, and 0.03°, and the repeatability (SD) of the body plate was 0.05°, 0.04°, and 0.04°, respectively. The mean differences ± SD between the calculated and measured pitch, roll, and yaw angles for the head and neck plate with the digital goniometer were 0.00 ± 0.06°, −0.01 ± 0.06°, and −0.04 ± 0.04°, respectively. The differences for the body plate were −0.03 ± 0.04°, 0.03 ± 0.05°, and 0.02 ± 0.05°, respectively. Results of the cone‐beam CT were similar to those of the digital goniometer. The prototype device exhibited good performance regarding the rotational accuracy and repeatability under no load. The clinical implementation of this concept is expected to reduce the residual error of the patient position due to the twist.

Development of a semi-customized tongue displacement device using a 3D printer for head and neck IMRT

Purpose

To reduce radiation doses to the tongue, a patient-specific semi-customized tongue displacement device (SCTDD) was developed using a 3D printer (3DP) for head and neck (H&N) radiation therapy (RT). Dosimetric characteristics of the SCTDD were compared with those of a standard mouthpiece (SMP).

Materials and methods

The SCTDD consists of three parts: a mouthpiece, connector with an immobilization mask, and tongue displacer, which can displace the tongue to the contralateral side of the planning target volume. Semi-customization was enabled by changing the thickness and length of the SCTDD. The instrument was printed using a 3DP with a biocompatible material. With the SCTDD and SMP, two sets of planning computed tomography (CT) and tomotherapy plans were obtained for seven H&N cancer patients. Dosimetric and geometric characteristics were compared.

Results

Using the SCTDD, the tongue was effectively displaced from the planning target volume without significant tongue volume change compared to the SMP. The median tongue dose was significantly reduced (29.6 Gy vs. 34.3 Gy). The volumes of the tongue receiving a dose of 15 Gy, 30 Gy, 35 Gy, 45 Gy, and 60 Gy were significantly lower than using the SMP.

Conclusion

The SCTDD significantly decreased the radiation dose to the tongue compared to the SMP, which may potentially reduce RT-related tongue toxicity.

Precision Risk Analysis of Cancer Therapy with Interactive Nomograms and Survival Plots

We present the design and evaluation of an integrated problem solving environment for cancer therapy analysis. The environment intertwines a statistical martingale model and a K Nearest Neighbor approach with visual encodings, including novel interactive nomograms, in order to compute and explain a patient's probability of survival as a function of similar patient results. A coordinated views paradigm enables exploration of the multivariate, heterogeneous and few-valued data from a large head and neck cancer repository. A visual scaffolding approach further enables users to build from familiar representations to unfamiliar ones. Evaluation with domain experts show how this visualization approach and set of streamlined workflows enable the systematic and precise analysis of a patient prognosis in the context of cohorts of similar patients. We describe the design lessons learned from this successful, multi-site remote collaboration.

Generating amorphous target margins in radiation therapy to promote maximal target coverage with minimal target size

BACKGROUND AND SIGNIFICANCE

This work provides proof-of-principle for two versions of a heuristic approach that automatically creates amorphous radiation therapy planning target volume (PTV) margins considering local effects of tumor shape and motion to ensure adequate voxel coverage with while striving to minimize PTV size. The resulting target thereby promotes disease control while minimizing the risk of normal tissue toxicity.

METHODS

This work describes the mixed-PDF algorithm and the independent-PDF algorithm which generate amorphous margins around a radiation therapy target by incorporating user-defined models of target motion. Both algorithms were applied to example targets - one circular and one "cashew-shaped." Target motion was modeled by four probability density functions applied to the target quadrants. The spatially variant motion model illustrates the application of the algorithms even with tissue deformation. Performance of the margins was evaluated in silico with respect to voxelized target coverage and PTV size, and was compared to conventional techniques: a threshold-based probabilistic technique and an (an)isotropic expansion technique. To demonstrate the algorithm's clinical utility, a lung cancer patient was analyzed retrospectively. For this case, 4D CT measurements were combined with setup uncertainty to compare the PTV from the mixed-PDF algorithm with a PTV equivalent to the one used clinically.

RESULTS

For both targets, the mixed-PDF algorithm performed best, followed by the independent-PDF algorithm, the threshold algorithm, and lastly, the (an)isotropic algorithm. Superior coverage was always achieved by the amorphous margin algorithms for a given PTV size. Alternatively, the margin required for a particular level of coverage was always smaller (8-15%) when created with the amorphous algorithms. For the lung cancer patient, the mixed-PDF algorithm resulted in a PTV that was 13% smaller than the clinical PTV while still achieving ≥99.9% coverage.

CONCLUSIONS

The amorphous margin algorithms are better suited for the local effects of target shape and positional uncertainties than conventional margins. As a result, they provide superior target coverage with smaller PTVs, ensuring dose delivered to the target while decreasing the risk of normal tissue toxicity.

Dosimetric influence of pitch in patient positioning for radiotherapy of long treatment volumes; the usefulness of six degree of freedom couch

OBJECTIVE:

Pitch, the rotation around the transversal axis of the patient during radiotherapy has little impact on the dose distribution of small spherical treatment volumes; however it might affect treatment of long volumes requiring a correction with a six degree of freedom couch.

METHODS:

We included 10 patients each with nasopharyngeal carcinoma (NPC) and esophageal cancer, treated with volumetric modulated arc therapy. Pitch was simulated by tilting the planning CT in ventral and dorsal direction by ± 1.5° and ± 3°. Verification plans were calculated on the tilted datasets and were compared to the original plan and the dose constraints of the organs at risk (OAR).

RESULTS:

The deviation in dose to the planning target volume is increasing with the degree of pitch with mean changes of up to 2% for NPC and 1% for esophageal cancer. The most affected OAR in NPC patients are brainstem (max. dose +6.0%) and spinal cord (max. dose +10.0%) when tilted by 3° dorsally and lenses (max. dose +3.3%), oral mucosa (mean dose +2.6%) and parotid glands (mean dose +4.3%) when tilted by 3° ventrally. For esophageal cancer patients, there was no significant change in dose to any OAR. Whereas for esophageal cancer, all tilted treatment plans were still clinically acceptable regarding OAR, 5 NPC plans would no longer be acceptable with a pitch of 1.5° ventral (N = 1), 3° ventral (N = 2) and 3° dorsal (N = 2).

CONCLUSION:

Planning target volume coverage in both tumor entities was only slightly affected, but pitch errors could be relevant for OAR in NPC patients.

ADVANCES IN KNOWLEDGE:

A correction with a six degree of freedom couch is recommended for NPC patients with a pitch mismatch of more than 1.5° to avoid exceeded doses to the OAR.

Establishing action threshold for change in patient anatomy using EPID gamma analysis and PTV coverage for head and neck radiotherapy treatment

PURPOSE

To present a new adaptive radiotherapy (ART) method based on relative gamma analysis and patient classification for the identification of anatomical changes that induce a sufficient dosimetric impact to affect the treatment delivery and require complete replanning.

METHODS

This retrospective study includes 55 patients treated for a head and neck cancer with IMRT, VMAT, or 3D conformal RT. Electronic Portal Imaging Device images for all treatment fields were acquired daily at every fraction. CBCTs were collected at least once a week. Gamma analysis was performed using the first fraction of the treatment as a reference once validated that it was delivered without error. Gamma analysis parameters (<γ>, standard deviation and the Top 1% γ) were used to define categories using statistic from a k-means clustering analysis. From these categories an action threshold was defined and correlated with dosimetric changes. For 23 of 55 patients, the V100% for PTV was computed for both, the planning CT and original contours deformed onto CBCT acquired at the last fraction. These values were then compared with 2D image relative γ-analysis of EPID images. Sensitivity and specificity of the method for the detection of dosimetric changes were computed.

RESULTS

Three categories indicating an increasing level of change with the planned treatment were identified. A threshold was established for which patients were at risk of deviation at <γ> = 0.42. From 23 recomputing plans, it has been confirmed that patients with a strong dosimetric impact were above this threshold, with a specificity of 0.80 and a sensitivity of 0.84.

CONCLUSIONS

The specificity and the sensitivity value confirmed the performance of the method to detect anatomical changes. The γ-analysis threshold correlated well with morphological changes that have a relevant dosimetric impact. Analysis of daily EPID images provides a method to identify patients at risk of deviation from their planned treatment and can support an early replanning decision.

Dosimetric Evaluation of Incorporating Patient Geometric Variations Into Adaptive Plan Optimization Through Probabilistic Treatment Planning in Head and Neck Cancers

PURPOSE

Four-dimensional (4D) adaptive radiation therapy (ART) treatment planning is an alternative to the conventional margin-based treatment planning approach. In 4D ART, interfraction patient geometric variations, gathered from computed tomography (CT) or cone beam CT (CBCT) images acquired during the patient treatment course, are directly incorporated into the adaptive plan optimization using a probabilistic treatment planning method. The goal of the present planning study was to evaluate the dosimetric differences between 4D ART and conventional margin-based adaptive planning strategies for head and neck cancers. In addition, we examined whether the dose differences achieved with 4D ART would translate into clinically relevant toxicity reductions using the existing normal tissue complication probability (NTCP) models.

METHODS AND MATERIALS

For 18 head and neck cancer patients, the treatment plans were retrospectively generated for 4 different treatment strategies, including a solely image guided radiation therapy (IGRT) strategy (IGRT-only), 2 conventional adaptive treatment planning strategies using 3- and 0-mm planning target volume (PTV) margins, and the 4D ART strategy. In the IGRT-only strategy, a conventional 3-mm PTV margin treatment plan was applied for the entire treatment course. In the 2 conventional adaptive strategies, 2 new treatment plans were generated during the treatment course using diagnostic planning CT scans acquired after the 10th and 22nd fractions. The 4D ART followed the same adaptive schedule, except that the 4D adaptive plan was generated using 5 CBCT images acquired during the 5 most recent treatment fractions. For each strategy, the actual delivered dose for the entire treatment course was constructed by calculating the daily doses on 35 CBCT scans, deforming back to the pretreatment planning CT scan, and accumulating over all 35 fractions. The target coverage was evaluated using the percentage of target volume receiving ≥100% of the prescription dose (V_100%) and the minimum dose to 99% of the target volume (D₉₉). It was considered adequate if the V_100% was ≥95% and the dose deficit in D₉₉ was ≤2 Gy (with respect to the prescription dose). For each strategy, the dose received by the organs at risk (OARs) was also evaluated, and the corresponding NTCP values were subsequently calculated using 3 NTCP models.

RESULTS

Adequate target coverage was achieved for the primary clinical target volume (CTV1) and elective nodal CTV (CTV2) with a 3-mm PTV margin, regardless of adaptation. The 3-mm ART plan reduced the OAR mean dose by 1 to 2 Gy compared with the IGRT-only plan. The 0-mm ART plan further reduced the OAR dose by another 2 to 3 Gy at the expense of target coverage: 3 and 1 patient had V_100% <95%, and 6 and 5 patients had a >2 Gy dose deficit in D₉₉ for the CTV1 and CTV2, respectively. Use of 4D ART improved target coverage and attained OAR sparing similar to that with 0-mm ART. The number of patients with V_100% <95% and >2 Gy D₉₉ deficit decreased to 0 and 0 for CTV1 and 0 and 2 for CTV2, respectively. The NTCP calculations suggested that 4D ART could benefit a substantial portion of patients compared with IGRT-only because 17 and 12 patients had ≥5% and ≥10% NTCP reductions for parotid toxicity and 18 and 3 patients had ≥5% and ≥10% NTCP reductions for swallowing toxicity, respectively.

CONCLUSIONS

Compared with margin-based adaptive planning strategies, 4D ART provides a better balance between target coverage and OAR sparing. NTCP estimation predicted for theoretical clinical benefits that warrant further clinical validation.

Set-up errors in head and neck cancer treated with IMRT technique assessed by cone-beam computed tomography: a feasible protocol

PURPOSE

To investigate set-up errors, suggest the adequate planning target volume (PTV) margin and image-guided radiotherapy frequency in head and neck (H&N) cancer treated with intensity-modulated radiotherapy (IMRT) assessed by kV cone-beam computed tomography (CBCT).

METHODS

We analyzed 360 CBCTs in 60 patients with H&N cancer treated with IMRT. The target delineation was contoured according to ICRU62. PTVs were generated by adding a 3-5 mm margin in all directions to the respective clinical target volumes. The kV CBCT images were obtained at first three days of irradiation and weekly thereafter. The overall mean displacement, range, systematic (∑) and random (σ) errors were calculated. Adequate PTV margins were calculated according to the van Herk formula (2.5∑ + 0.7r).

RESULTS

The mean of set-up errors was less than 2 mm in any direction. The overall frequency of set-up displacements greater than 3 mm was 3.9% in medial-lateral (ML) direction, 8% in superior-inferior (SI) direction, and 15.5% in anterior-posterior (AP) direction. The range of translations shifts was 0-9 mm in ML direction, 0-5 mm in SI direction and 0-10 mm in AP direction, respectively. After systematic set-up errors correction, the adequate margin to overcome the problem of set-up errors was found to be less than 3 mm.

CONCLUSION

Image-guided kV CBCT was effective for the evaluation of set-up accuracy in H&N cancer. The kV CBCT at first three fractions and followed-by weekly appears adequate for reducing significantly set-up errors in H&N cancer treated with IMRT technique. Finally, 3-5 mm PTV margins appear adequate and safe to overcome the problem of set-up errors.

Patient positioning in head and neck cancer : Setup variations and safety margins in helical tomotherapy

OBJECTIVE

To evaluate the interfractional variations of patient positioning during intensity-modulated radiotherapy (IMRT) with helical tomotherapy in head and neck cancer and to calculate the required safety margins (sm) for bony landmarks resulting from the necessary table adjustments.

MATERIALS AND METHODS

In all, 15 patients with head and neck cancer were irradiated using the Hi-Art II tomotherapy system between April and September 2016. Before therapy sessions, patient position was frequently checked by megavolt computed tomography (MV-CT). Necessary table adjustments (ta) in the right-left (rl), superior-inferior (si) and anterior-posterior (ap) directions were recorded for four anatomical points: second, fourth and sixth cervical vertebral body (CVB), anterior nasal spine (ANS). Based upon these data sm were calculated for non-image-guided radiotherapy, image-guided radiotherapy (IGRT) and image guidance limited to a shortened area (CVB 2).

RESULTS

Based upon planning CT the actual treatment required ta from -0.05 ± 1.31 mm for CVB 2 (ap) up to 2.63 ± 2.39 mm for ANS (rl). Considering the performed ta resulting from image control (MV-CT) we detected remaining ta from -0.10 ± 1.09 mm for CVB 4 (rl) up to 1.97 ± 1.64 mm for ANS (si). After theoretical adjustment of patients position to CVB 2 the resulting ta ranged from -0.11 ± 2.44 mm for CVB6 (ap) to 2.37 ± 2.17 mm for ANS (si). These data imply safety margins: uncorrected patient position: 3.63-9.95 mm, corrected positioning based upon the whole target volume (IGRT): 1.85-6.63 mm, corrected positioning based upon CVB 2 (IGRT): 3.13-6.66 mm.

CONCLUSIONS

The calculated safety margins differ between anatomic regions. Repetitive and frequent image control of patient positioning is necessary that, however, possibly may be focussed on a limited region.

Evaluation of the Efficacy of Rotational Corrections for Standard-Fractionation Head and Neck Image-Guided Radiotherapy

PURPOSE

Modern linear accelerators are equipped with cone beam computed tomography and robotic couches that can correct for errors in the translational (X, Y, Z) and rotational (α, β, γ) axes prior to treatment delivery. Here, we compared the positional accuracy of 2 cone beam registration approaches: (1) employing translational shifts only in 3 degrees of freedom (X, Y, Z), versus; (2) using translational-rotational shifts in 6 degrees of freedom (X, Y, Z, α, β, γ).

METHODS

This retrospective study examined 140 interfraction cone beam images from 20 patients with head and neck cancer treated with standard intensity-modulated radiation therapy. The cone beam images were matched to planning simulation scans in 3, then in 6 degrees of freedom, using the mandible, clivus, and C2 and C7 vertebrae as surrogate volumes. Statistical analyses included a generalized mixed model and was used to assess whether there were significant differences in acceptable registrations between the 2 correction methods.

RESULTS

The rates of improvement with corrections in 6 degrees of freedom for the mandible with a 5-mm expansion margin were 54.55% ( P = .793), for the clivus 85.71% ( P = .222), and for C7 87.50% ( P = .015). There was a 100% increase in acceptability for the C2 vertebra within the 5-mm margin ( P < .001). For the 3-mm expansion margin, the rates of improvement for the mandible, clivus, C2, and C7 were 63.16% ( P = .070), 91.30% ( P = .011), 84.21% ( P = .027), and 76.92% ( P < .001), respectively.

CONCLUSIONS

Significant registration improvements with the use of rotational corrections with a 5-mm expansion margin are only seen in the C7 vertebra. At the 3-mm margin, significant improvements are found for the C2, C7, and clivus registrations, suggesting that intensity-modulated radiotherapy treatments for head and neck cancers with 3-mm planning target volume margins may benefit from corrections in 6 degrees of freedom.

Minimal mask immobilization with optical surface guidance for head and neck radiotherapy

PURPOSE

Full face and neck thermoplastic masks provide standard-of-care immobilization for patients receiving H&N IMRT. However, these masks are uncomfortable and increase skin dose. The purpose of this pilot trial was to investigate the feasibility and setup accuracy of minimal face and neck mask immobilization with optical surface guidance.

METHODS

Twenty patients enrolled onto this IRB-approved protocol. Patients were immobilized with masks securing only forehead and chin. Shoulder movement was restricted by either moldable cushion or hand held strap retractors. Positional information, including isocenter location and CT skin contours, were imported to a commercial surface image guidance system. Patients typically received standard-of-care IMRT to 60-70 Gy in 30-33 fractions. Patients were first set up to surface markings with optical image guidance referenced to regions of interest (ROIs) on simulation CT images. Positioning was confirmed by in-room CBCT. Following six-dimensional robotic couch correction, a new optical real-time surface image was acquired to track intrafraction motion and to serve as a reference surface for setup at the next treatment fraction. Therapists manually recorded total treatment time as well as couch shifts based on kV imaging. Intrafractional ROI motion tracking was automatically recorded by the optical image guidance system. Patient comfort was assessed by self-administered surveys.

RESULTS

Setup error was measured as six-dimensional shifts (vertical/longitudinal/lateral/rotation/pitch/roll). Mean error values were -0.51 ± 2.42 mm, -0.49 ± 3.30 mm, 0.23 ± 2.58 mm, -0.15 ± 1.01o , -0.02 ± 1.19o , and 0.06 ± 1.08o , respectively. Average treatment time was 21.6 ± 8.4 mins). Subjective comfort during surface-guided treatment was confirmed on patient surveys.

CONCLUSION

These pilot results confirm feasibility of minimal mask immobilization combined with commercially available optical image guidance. Patient acceptance of minimal mask immobilization has been encouraging. Follow-up validation, with direct comparison to standard mask immobilization, appears warranted.

A Pilot Study Evaluating the Effectiveness of Dual-Registration Image-Guided Radiotherapy in Patients with Oropharyngeal Cancer

PURPOSE

The purpose of the article was to determine the impact of Dual Registration (DR) image-guided radiotherapy (IGRT) on clinical judgement and treatment delivery for patients with oropharyngeal cancer before implementation.

METHODS

Ninety cone beam computed tomography images from 10 retrospective patients were matched using standard clipbox registration (SCR) and DR. Three IGRT specialist radiographers performed all registrations and evaluated by intraclass correlation to determine inter-rater agreement, Bland-Altman with 95% limits of agreement to determine differences between SCR and DR procedures, changes in clinical judgment, time taken to perform registrations, and radiographer satisfaction.

RESULTS

Inter-rater agreement between radiographers using both SCR and DR was high (0.867 and 0.917, P ≤ .0001). The 95% limits of agreement between SCR and DR procedures in the mediolateral, cranial-caudal, and ventrodorsal translational directions were -6.40 to +4.91, -7.49 to +6.05, and -7.00 to +5.44 mm, respectively. The mediolateral direction demonstrated significant proportional bias (P ≤ .001) suggesting non-agreement between SCR and DR. Eighty percent of DR matches resulted in a change in clinical judgement to ensure maximum target coverage. Mean registration times for SCR and DR were 94 and 115 seconds, respectively, and radiographers found DR feasible and satisfactory.

CONCLUSION

The standard method using SCR in patients with oropharyngeal cancer underestimates the deviation in the lower neck. In these patients, DR is an effective IGRT tool to ensure target coverage of the inferior neck nodes and has demonstrated acceptability to radiotherapy clinical practice.

Setup uncertainties and PTV margins at different anatomical levels in intensity modulated radiotherapy for nasopharyngeal cancer

AIM

To determine the systematic error (∑), random error (σ) and derive PTV margin at different levels of the target volumes in Nasopharyngeal Cancer (NPC).

MATERIALS AND METHODS

A retrospective offline review was done for patients who underwent IMRT for NPC from June 2015 to May 2016 at our institution. Alternate day kV images were matched with digitally reconstructed radiographs to know the setup errors. All radiographs were matched at three levels - the clivus, third cervical (C3) and sixth cervical (C6) vertebra. The shifts in positions along the vertical, longitudinal and lateral axes were noted and the ∑ and σ at three levels were calculated. PTV margins were derived using van Herk's formula.

RESULTS

Twenty patients and 300 pairs of orthogonal portal films were reviewed. The ∑ for the clivus, C3 and C6 along vertical, longitudinal and lateral directions were 1.6 vs. 1.8 vs. 2 mm; 1.2 vs. 1.4 vs. 1.4 mm and 0.9 vs. 1.6 and 2.3 mm, respectively. Similarly, the random errors were 1.1 vs. 1.4 vs. 1.8 mm; 1.1 vs. 1.2 vs. 1.2 mm and 1.2 vs. 1.3 vs. 1.6 mm. The PTV margin at the clivus was 4.4 mm along the vertical, 4 mm along the longitudinal direction and 3.2 m in the lateral direction. At the C3 level, it was 5.5 mm in the vertical, 5 mm in the lateral direction and 4.4 mm in the longitudinal direction. At the C6 level, it was 6.4 mm in the vertical, 6.9 mm in the lateral direction and 4.4 mm in the longitudinal direction.

CONCLUSION

A differential margin along different levels of target may be necessary to adequately cover the target.

Evaluation of Imaging Dose From Different Image Guided Systems During Head and Neck Radiotherapy: A Phantom Study

This work evaluated and compared the absorbed doses to selected organs in the head and neck region from the three image guided radiotherapy systems: cone-beam computed tomography (CBCT) and kilovoltage (kV) planar imaging using the On-board Imager® (OBI) as well as the ExacTrac® X-ray system, all available on the Varian Novalis TX linear accelerator. The head and neck region of an anthropomorphic phantom was used to simulate patients' head within the imaging field. Nanodots optically stimulated luminescent dosemeters were positioned at selected sites to measure the absorbed doses. CBCT was found to be delivering the highest dose to internal organs while OBI-2D gave the highest doses to the eye lenses. The setting of half-rotation in CBCT effectively reduces the dose to the eye lenses. Daily high-quality CBCT verification was found to increase the secondary cancer risk by 0.79%.

Cost-effective immobilization for whole brain radiation therapy

To investigate the inter‐ and intra‐fraction motion associated with the use of a low‐cost tape immobilization technique as an alternative to thermoplastic immobilization masks for whole‐brain treatments. The results of this study may be of interest to clinical staff with severely limited resources (e.g., in low‐income countries) and also when treating patients who cannot tolerate standard immobilization masks. Setup reproducibility of eight healthy volunteers was assessed for two different immobilization techniques. (a) One strip of tape was placed across the volunteer's forehead and attached to the sides of the treatment table. (b) A second strip was added to the first, under the chin, and secured to the table above the volunteer's head. After initial positioning, anterior and lateral photographs were acquired. Volunteers were positioned five times with each technique to allow calculation of inter‐fraction reproducibility measurements. To estimate intra‐fraction reproducibility, 5‐minute anterior and lateral videos were taken for each technique per volunteer. An in‐house software was used to analyze the photos and videos to assess setup reproducibility. The maximum intra‐fraction displacement for all volunteers was 2.8 mm. Intra‐fraction motion increased with time on table. The maximum inter‐fraction range of positions for all volunteers was 5.4 mm. The magnitude of inter‐fraction and intra‐fraction motion found using the “1‐strip” and “2‐strip” tape immobilization techniques was comparable to motion restrictions provided by a thermoplastic mask for whole‐brain radiotherapy. The results suggest that tape‐based immobilization techniques represent an economical and useful alternative to the thermoplastic mask.

The influence of non-rigid anatomy and patient positioning on endoscopy-CT image registration in the head and neck

PURPOSE

To assess the influence of non-rigid anatomy and differences in patient positioning between CT acquisition and endoscopic examination on endoscopy-CT image registration in the head and neck.

METHODS

Radiotherapy planning CTs and 31-35 daily treatment-room CTs were acquired for nineteen patients. Diagnostic CTs were acquired for thirteen of the patients. The surfaces of the airways were segmented on all scans and triangular meshes were created to render virtual endoscopic images with a calibrated pinhole model of an endoscope. The virtual images were used to take projective measurements throughout the meshes, with reference measurements defined as those taken on the planning CTs and test measurements defined as those taken on the daily or diagnostic CTs. The influence of non-rigid anatomy was quantified by 3D distance errors between reference and test measurements on the daily CTs, and the influence of patient positioning was quantified by 3D distance errors between reference and test measurements on the diagnostic CTs. The daily CT measurements were also used to investigate the influences of camera-to-surface distance, surface angle, and the interval of time between scans.

RESULTS

Average errors in the daily CTs were 0.36 ± 0.61 cm in the nasal cavity, 0.58 ± 0.83 cm in the naso- and oropharynx, and 0.47 ± 0.73 cm in the hypopharynx and larynx. Average errors in the diagnostic CTs in those regions were 0.52 ± 0.69 cm, 0.65 ± 0.84 cm, and 0.69 ± 0.90 cm, respectively. All CTs had errors heavily skewed towards 0, albeit with large outliers. Large camera-to-surface distances were found to increase the errors, but the angle at which the camera viewed the surface had no effect. The errors in the Day 1 and Day 15 CTs were found to be significantly smaller than those in the Day 30 CTs (P < 0.05).

CONCLUSIONS

Inconsistencies of patient positioning have a larger influence than non-rigid anatomy on projective measurement errors. In general, these errors are largest when the camera is in the superior pharynx, where it sees large distances and a lot of muscle motion. The errors are larger when the interval of time between CT acquisitions is longer, which suggests that the interval of time between the CT acquisition and the endoscopic examination should be kept short. The median errors found in this study are comparable to acceptable levels of uncertainty in deformable CT registration. Large errors are possible even when image alignment is very good, indicating that projective measurements must be made carefully to avoid these outliers.

Exploratory study of the association of volumetric modulated arc therapy (VMAT) plan robustness with local failure in head and neck cancer

This work is to show which is more relevant to cause local failures (LFs) due to patient setup uncertainty between the planning target volume (PTV) underdosage and the potential target underdosage subject to patient setup uncertainties in head and neck (H&N) cancer treated with volumetric‐modulated arc therapy (VMAT). Thirteen LFs in 10 H&N patients treated by VMAT were analyzed. Measures have been taken to minimize the chances of insufficient target delineation for these patients and the patients were clinically determined to have LF based on the PET/CT scan results by an experienced radiologist and then reviewed by a second experienced radiation oncologist. Two methods were used to identify the possible locations of LF due to underdosage: (a) examining the standard VMAT plan, in which the underdosed volume in the nominal dose distribution (UVN) was generated by subtracting the volumes receiving the prescription doses from PTVs, and (b) plan robustness analysis, in which in addition to the nominal dose distribution, six perturbed dose distributions were created by translating the CT iso‐center in three cardinal directions by the PTV margin. The coldest dose distribution was represented by the minimum of the seven doses in each voxel. The underdosed volume in the coldest dose distribution (UVC) was generated by subtracting the volumes receiving the prescription doses in the coldest dose distribution from the volumes receiving the prescription doses in the nominal dose distribution. UVN and UVC were subsequently examined for spatial association with the locations of LF. The association was tested using the binominal distribution and the Fisher's exact test of independence. We found that of 13 LFs, 11 were associated with UVCs (P = 0.011), while three were associated with UVNs (P = 0.99). We concluded that the possible target underdosage due to patient setup uncertainties appeared to be a more relevant factor associated with LF in VMAT for H&N cancer than the compromised PTV coverage at least for the patients included in this study.

Quantification of the accuracy limits of image registration using peak signal-to-noise ratio

A new method was developed for quantifying the accuracy limits of image registration devices and the distortion of anatomical structures in verification images without image registration. A correlation was found between peak signal-to-noise ratio (PSNR) and the amount of parallel movement (1-10 mm at 1-mm intervals) of a rectangular parallelepiped phantom [correlation coefficient (CC) -0.91, contribution ratio (CR) 0.83]. Rotating the phantom from 1° to 10° at 1° intervals produced a similar correlation with PSNR (CC -0.91, CR 0.83). To allow for manual registration, the grid pattern of the Mylar top plate was extracted from 455 pelvic portal images of 21 patients using a band-pass filtering technique. This revealed a different correlation between the original data (CC -0.62, CR 0.38) and averaged data (CC -0.96, CR 0.92), but this is considered to have been caused by structural distortion and manual matching errors. Thus, PSNR can be used to evaluate the accuracy limits of image registration and provide a judgment index that can be used in re-planning or re-setup in adaptive radiotherapy.

Comparison of Safety Margin Generation Concepts in Image Guided Radiotherapy to Account for Daily Head and Neck Pose Variations

Purpose

Intensity modulated radiation therapy (IMRT) of head and neck tumors allows a precise conformation of the high-dose region to clinical target volumes (CTVs) while respecting dose limits to organs a risk (OARs). Accurate patient setup reduces translational and rotational deviations between therapy planning and therapy delivery days. However, uncertainties in the shape of the CTV and OARs due to e.g. small pose variations in the highly deformable anatomy of the head and neck region can still compromise the dose conformation. Routinely applied safety margins around the CTV cause higher dose deposition in adjacent healthy tissue and should be kept as small as possible.

Materials and Methods

In this work we evaluate and compare three approaches for margin generation 1) a clinically used approach with a constant isotropic 3 mm margin, 2) a previously proposed approach adopting a spatial model of the patient and 3) a newly developed approach adopting a biomechanical model of the patient. All approaches are retrospectively evaluated using a large patient cohort of over 500 fraction control CT images with heterogeneous pose changes. Automatic methods for finding landmark positions in the control CT images are combined with a patient specific biomechanical finite element model to evaluate the CTV deformation.

Results

The applied methods for deformation modeling show that the pose changes cause deformations in the target region with a mean motion magnitude of 1.80 mm. We found that the CTV size can be reduced by both variable margin approaches by 15.6% and 13.3% respectively, while maintaining the CTV coverage. With approach 3 an increase of target coverage was obtained.

Conclusion

Variable margins increase target coverage, reduce risk to OARs and improve healthy tissue sparing at the same time.

Water equivalent path length calculations using scatter-corrected head and neck CBCT images to evaluate patients for adaptive proton therapy

Proton therapy has dosimetric advantages due to the well-defined range of the proton beam over photon radiotherapy. When the proton beams, however, are delivered to the patient in fractionated radiation treatment, the treatment outcome is affected by delivery uncertainties such as anatomic change in the patient and daily patient setup error. This study aims at establishing a method to evaluate the dosimetric impact of the anatomic change and patient setup error during head and neck proton therapy. Range variations due to the delivery uncertainties were assessed by calculating water equivalent path length (WEPL) to the distal edge of tumor volume using planning CT and weekly treatment cone-beam CT (CBCT) images. Specifically, mean difference and root mean squared deviation (RMSD) of the distal WEPLs were calculated as the weekly range variations. To accurately calculate the distal WEPLs, an existing CBCT scatter correction algorithm was used. An automatic rigid registration was used to align the planning CT and treatment CBCT images, simulating a six degree-of-freedom couch correction at treatments. The authors conclude that the dosimetric impact of the anatomic change and patient setup error was reasonably captured in the differences of the distal WEPL variation with a range calculation uncertainty of 2%. The proposed method to calculate the distal WEPL using the scatter-corrected CBCT images can be an essential tool to decide the necessity of re-planning in adaptive proton therapy.

Robustness quantification methods comparison in volumetric modulated arc therapy to treat head and neck cancer

BACKGROUND

To compare plan robustness of volumetric modulated arc therapy (VMAT) with intensity modulated radiation therapy (IMRT) and to compare the effectiveness of 3 plan robustness quantification methods.

METHODS AND MATERIALS

The VMAT and IMRT plans were created for 9 head and neck cancer patients. For each plan, 6 new perturbed dose distributions were computed using ±3 mm setup deviations along each of the 3 orientations. Worst-case analysis (WCA), dose-volume histogram (DVH) band (DVHB), and root-mean-square dose-volume histogram (RVH) were used to quantify plan robustness. In WCA, a shaded area in the DVH plot bounded by the DVHs from the lowest and highest dose per voxel was displayed. In DVHB, we displayed the envelope of all DVHs in band graphs of all the 7 dose distributions. The RVH represents the relative volume on the vertical axis and the root-mean-square-dose on the horizontal axis. The width from the first 2 methods at different target DVH indices (such as D95% and D5%) and the area under the RVH curve for the target were used to indicate plan robustness. Results were compared using Wilcoxon signed-rank test.

RESULTS

The DVHB showed that the width at D95% of IMRT was larger than that of VMAT (unit Gy) (1.59 vs 1.18) and the width at D5% of IMRT was comparable to that of VMAT (0.59 vs 0.54). The WCA showed similar results between IMRT and VMAT plans (D95%: 3.28 vs 3.00; D5%: 1.68 vs 1.95). The RVH showed the area under the RVH curve of IMRT was comparable to that of VMAT (1.13 vs 1.15). No statistical significance was found in plan robustness between IMRT and VMAT.

CONCLUSIONS

The VMAT is comparable to IMRT in terms of plan robustness. For the 3 quantification methods, WCA and DVHB are DVH parameter-dependent, whereas RVH captures the overall effect of uncertainties.

Towards 3D printed multifunctional immobilization for proton therapy: Initial materials characterization

PURPOSE

3D printing technology is investigated for the purpose of patient immobilization during proton therapy. It potentially enables a merge of patient immobilization, bolus range shifting, and other functions into one single patient-specific structure. In this first step, a set of 3D printed materials is characterized in detail, in terms of structural and radiological properties, elemental composition, directional dependence, and structural changes induced by radiation damage. These data will serve as inputs for the design of 3D printed immobilization structure prototypes.

METHODS

Using four different 3D printing techniques, in total eight materials were subjected to testing. Samples with a nominal dimension of 20 × 20 × 80 mm³ were 3D printed. The geometrical printing accuracy of each test sample was measured with a dial gage. To assess the mechanical response of the samples, standardized compression tests were performed to determine the Young's modulus. To investigate the effect of radiation on the mechanical response, the mechanical tests were performed both prior and after the administration of clinically relevant dose levels (70 Gy), multiplied with a safety factor of 1.4. Dual energy computed tomography (DECT) methods were used to calculate the relative electron density to water ρ_e, the effective atomic number Z_eff, and the proton stopping power ratio (SPR) to water SPR. In order to validate the DECT based calculation of radiological properties, beam measurements were performed on the 3D printed samples as well. Photon irradiations were performed to measure the photon linear attenuation coefficients, while proton irradiations were performed to measure the proton range shift of the samples. The directional dependence of these properties was investigated by performing the irradiations for different orientations of the samples.

RESULTS

The printed test objects showed reduced geometric printing accuracy for 2 materials (deviation > 0.25 mm). Compression tests yielded Young's moduli ranging from 0.6 to 2940 MPa. No deterioration in the mechanical response was observed after exposure of the samples to 100 Gy in a therapeutic MV photon beam. The DECT-based characterization yielded Z_eff ranging from 5.91 to 10.43. The SPR and ρ_e both ranged from 0.6 to 1.22. The measured photon attenuation coefficients at clinical energies scaled linearly with ρ_e. Good agreement was seen between the DECT estimated SPR and the measured range shift, except for the higher Z_eff. As opposed to the photon attenuation, the proton range shifting appeared to be printing orientation dependent for certain materials.

CONCLUSIONS

In this study, the first step toward 3D printed, multifunctional immobilization was performed, by going through a candidate clinical workflow for the first time: from the material printing to DECT characterization with a verification through beam measurements. Besides a proof of concept for beam modification, the mechanical response of printed materials was also investigated to assess their capabilities for positioning functionality. For the studied set of printing techniques and materials, a wide variety of mechanical and radiological properties can be selected from for the intended purpose. Moreover the elaborated hybrid DECT methods aid in performing in-house quality assurance of 3D printed components, as these methods enable the estimation of the radiological properties relevant for use in radiation therapy.

Radiation Oncology--New Approaches in Squamous Cell Cancer of the Head and Neck

The many advances in radiotherapy for squamous cell cancer of the head and neck described in this article will have significant effects on the ultimate outcomes of patients who receive this treatment. The technological and clinical advances should allow one to maintain or improve disease control, while moderating the toxicity associated with head and neck radiation therapy.

A Systematic Analysis of Errors in Target Localization and Treatment Delivery for Stereotactic Radiosurgery Using 2D/3D Image Registration

PURPOSE

To determine the localization uncertainties associated with 2-dimensional/3-dimensional image registration in comparison to 3-dimensional/3-dimensional image registration in 6 dimensions on a Varian Edge Linac under various imaging conditions.

METHODS

The systematic errors in 6 dimensions were assessed by comparing automatic 2-dimensional/3-dimensional (kV/MV vs computed tomography) with 3-dimensional/3-dimensional (cone beam computed tomography vs computed tomography) image registrations under various conditions encountered in clinical applications. The 2-dimensional/3-dimensional image registration uncertainties for 88 patients with different treatment sites including intracranial and extracranial were evaluated by statistically analyzing 2-dimensional/3-dimensional pretreatment verification shifts of 192 fractions in stereotactic radiosurgery and stereotactic body radiotherapy.

RESULTS

The systematic errors of 2-dimensional/3-dimensional image registration using kV-kV, MV-kV, and MV-MV image pairs were within 0.3 mm and 0.3° for the translational and rotational directions within a 95% confidence interval. No significant difference ( P > .05) in target localization was observed with various computed tomography slice thicknesses (0.8, 1, 2, and 3 mm). Two-dimensional/3-dimensional registration had the best accuracy when pattern intensity and content filter were used. For intracranial sites, means ± standard deviations of translational errors were -0.20 ± 0.70 mm, 0.04 ± 0.50 mm, and 0.10 ± 0.40 mm for the longitudinal, lateral, and vertical directions, respectively. For extracranial sites, means ± standard deviations of translational errors were -0.04 ± 1.00 mm, 0.2 ± 1.0 mm, and 0.1 ± 1.0 mm for the longitudinal, lateral, and vertical directions, respectively. Two-dimensional/3-dimensional image registration for intracranial and extracranial sites had comparable systematic errors that were approximately 0.2 mm in the translational direction and 0.08° in the rotational direction.

CONCLUSION

The standard 2-dimensional/3-dimensional image registration tool available on the Varian Edge radiosurgery device, a state-of-the-art system, is helpful for robust and accurate target positioning for image-guided stereotactic radiosurgery.