Adaptive radiotherapy for nasopharyngeal carcinoma

Introduction

Intensity-modulated radiotherapy (IMRT) has become the standard treatment for nasopharyngeal carcinoma (NPC) in the precision radiotherapy era (1). It enables the delivery of highly conformal dose distribution to target volumes with superior normal tissue sparing (2). However, IMRT is often delivered on a single snapshot of the patient’s anatomy and position and does not take into account the potential changes occurring during a typical 7-week treatment course. With very steep dose fall-off at strategic locations around target volumes in NPC IMRT, significant target shrinkage and anatomical changes during the course of the treatment could increase the risk of geographical target miss and organs-at-risk (OARs) overdose (3,4). Data (5) have shown that systematic strategies addressing these patient-specific changes during a course of radiotherapy are particularly important for NPC due to its radio-/chemo-sensitive nature, proximity to multiple OARs, as well as limited salvage options in the event of subsequent local failure.

The concept of “adaptive radiotherapy” (ART) was introduced by Yan et al. in 1997 (6,7) as an imaging feedback control strategy with treatment plan modification in response to patient-specific treatment variation during the course of radiotherapy. It generally includes the following four steps: (I) treatment dose assessment, (II) treatment variation identification/evaluation, (III) treatment modification decisions, and (IV) adaptive treatment modification (8). Theoretically, ART can be employed on a daily basis to correct for any dose discrepancy from the original IMRT plan. However, in practice, ART strategies are often only implemented in selected cases at certain times. In part, this is due to the labor-intensive and time-consuming nature of the current ART processes. Identifying who may benefit from ART and when/how to implement ART remain active research areas. Significant progress has been made in recent years to improve and streamline the ART process for NPC in the clinical setting. This review summarizes recent advances in the implementation of ART for NPC relating to potential dosimetric and clinical benefits, how to trigger its use before or during the radiotherapy course, and efforts in streamlining ART such as improving deformable registration algorithms and refining computer-assisted auto-contouring tools.

Classification of ART

ART can be classified as reactive and proactive based on whether it is part of the initial treatment package. An example of reactive ART includes re-scan and re-plan to counter unstable treatment setup or significant observed anatomic changes caused by tumor shrinkage, weight loss, or internal motion. Proactive ART often incorporates re-planning as a part of the initial treatment package in anticipating significant tumor and normal tissue changes at certain time points. ART can be implemented for different purposes. To describe the ART intent, the following nomenclature has been proposed by Heukelom et al. (9): (I) ART_{ex_aequo}—serial plan verification to ensure initial plan parameters are maintained for tumor and OARs, (II) ART_OAR—reduced OAR dose with the same initial plan dosimetry to CTV, (III) ART_amplio—increased dose to tumor with isotoxic or lower OAR dose, (IV) ART_reduce—“shrinking CTV” for on-treatment responders, and (5) ART_totale—increase dose to sub-volume of initial CTV.

Necessity and benefits of ART

Many NPC patients can experience significant weight loss during the 6 to 7 weeks of radiotherapy. Patients having significant weight loss tend to be accompanied by reduced skin separation at various levels of the cervical spine and neck, causing significant inter-fractional setup instability. Excessive weight loss and tumor shrinkage may result in a significant deviation of accumulated delivered dose from the initially planned dose. Studies (10,11) have shown that these volumetric and geographic variations could compromise the conformality of IMRT plans and increase the dose to selected OARs. A prospective study of 19 NPC patients by Cheng et al. (11) evaluated volumetric and dosimetric changes during IMRT. Patients were rescanned at the 30 and 50 Gy time-points, and hybrid plans were generated by recontouring target volumes and OARs followed by applying the parameters of the original plan to the newly acquired CT at these two time-points. The authors reported a mean weight loss of 5.4% and 9.3%, a mean 9% and 16% reduction of gross tumor volume (GTV), and a mean volume reduction of the contralateral parotid gland by 0.7 and 3.4 cm³ and of the ipsilateral parotid by 5.3 and 8.4 cm³ at the 30 and 50 Gy dose points in the course, respectively. Compared to the original plan, the hybrid plan showed a significantly higher dose with greater dose inhomogeneity in most target volumes, and a higher maximum dose to the spinal cord and brainstem, as well as a higher mean dose to parotid glands.

The dosimetric and clinical benefits of ART in NPC have been demonstrated in several prospective and retrospective studies (4,10-19) (Table 1). Emerging data (12,13,15,17,19,20) have shown that adaptation of the treatment plan can result in improved target coverage and homogeneity, reduced dose maximum to critical structures like the spinal cord and brainstem, as well as volume reductions in target volumes and lower accumulated doses to parotid glands.

In addition to these dosimetric benefits, the clinical benefit of ART has been shown in several studies. Limited data suggest that ART has the potential to reduce normal-tissue toxicities and enhance locoregional control (LRC) although there is no significant difference in distant control and overall survival (4,16,18). A non-randomized prospective controlled cohort study by Yang et al. (18) showed that IMRT replanning improved quality of life and enhanced LRC in patients with NPC. However, the authors did not report the details regarding how the replanning was triggered. A propensity score-matched analysis conducted by Luo et al. (16) compared the outcome of T3-T4 NPC patients with (n=66) vs. without (n=66) replanning. The decision for replanning was made at the physician’s discretion and considered multiple factors such as proximity of GTV to critical OARs, significant weight loss, declining nutritional status, significant changes in tumor size and an ill-fitting mask, as well as severe acute toxicities. An average of two new ART plans (range, 1–3) was implemented. The time from re-simulation to implementation of the new plan was generally 1–3 days. The study showed that the replanning cohort had a higher LRC compared to the cohort without replanning, and the effect of replanning on LRC remained after adjusting for confounders. Distant metastasis rates were similar and remained the main pattern of treatment failure for both cohorts. No significant survival advantage was observed with ART.

Practical aspects of ART—triggers and timing

To incorporate ART into routine clinical practice, one needs to consider who would benefit from ART, and when to implement it. The latter often need to take into account any substantial volumetric changes that warrant ART and whether sufficient time remains in the treatment course to derive benefit from the adaptation.

Reasons or “triggers” for ART vary between studies (Table 2) (16,21-26). There is also no consensus regarding the optimal time to implement ART, and the “threshold” or “trigger” to mandate it. For reactive ART to account for time-dependent changes, triggered adaptations are frequently applied. Triggered adaptation refers to the process of adapting the treatment plan when exceeding a certain “threshold”, such as when a patient experiences considerable shrinkage of gross tumor or anatomical changes related to weight loss. Yao et al. (25) evaluated real-time volumetric and dosimetric changes in the parotid gland to determine the optimal replanning criteria (“trigger”) and timing for parotid protection-based adaptive IMRT in NPC. They suggested that when two out of the three following parameters reach their cut-off, an ART should be considered: (I) initial parotid volume >52.8 cm³, (II) initial parotid mean dose >32 Gy, and (III) weight loss rate >2.3% at the 11^th fraction or >3.6% at the 16^th fraction, or >4.4% at the 21^st fraction. In Huang et al.’s study (24), each patient had repeated CT scans after every five fractions and at treatment completion. They used auto-segmentation to re-contour the targets and OARs and performed deformable registration for CT-CT fusion. Two replans at the 5^th and 15^th fractions were suggested since significant volumetric changes occurred around these two time points.

The impact of anatomic change on actual delivered dose is highly patient-dependent and appears to affect OAR sparing (e.g., parotid) to a relatively greater extent compared to the impact on GTV coverage (27). A prospective study of weekly volumetric changes during chemoradiation on 20 head and neck cancer patients (4 were NPC) from Bhide et al (21) showed that the most significant volumetric and dosimetric alteration occurred at week 2 of IMRT. There was a significant parotid volume reduction by week 2 (15%, P<0.001) and week 4 (31%) (both P<0.001), and an increment of the mean dose to the ipsilateral parotid gland at week 4 of IMRT (2.7 Gy, P=0.006). For NPC patients with large nodes receiving definitive chemoradiotherapy, Brown et al. (22) recommended introducing ART at week 3 for NPC. From a parotid protection point of view, replanning in the fourth week seems appropriate since parotid shrinkage occurs in a linear pattern initially and reaches its peak at the 16^th fractions as shown by Ren et al. (28). A study by Gai et al. (23) showed that 85% NPC patients had ≥50% of GTV shrinkage before the 21^st fractions and parotid volume decreased significantly in the first 4 weeks, thereby suggesting replanning between the 21^st to 25^th fractions. It appears that the most common time-frame for ART is between 30–50 Gy (i.e., the 15^th–25^th fractions) during a course of 33–35 fractions (17,29).

Since ART currently remains a labor-intensive effort and not all NPC patients would significantly benefit, proactive identification of patients who might benefit using pretreatment clinical characteristics remains a research focus. Advanced NPC with bulky primaries or nodal disease seem to be a candidate subset for proactive ART. Brown et al. (30) found that higher N-category, larger pre-treatment largest involved lymph node (LN) size, and greater initial body weight (BW) were predictors for ART. They classified NPC patients into 3 risk groups for ART: low-risk: LN <6 cm with BW <100 kg or LN <1.5 cm with BW >100 kg; intermediate risk: N2-N3 disease or LN >6.0 cm with BW >100 kg; high-risk: N2-N3 disease, or BW >100 kg or LN >1.5 cm with BW >100 kg. The study by Zhao et al. (4) showed that patients with a T3-T4 primary or N2-N3 neck disease had an improved 3-year LRC with ART compared to case-matched control patients. A single-arm phase II study (JCOG1015, UMIN000005448) of two-step IMRT (ART at 46 Gy) for 75 stage II-IVB NPC patients showed excellent overall survival (3-year: 88%) with an acceptable toxicity profile. However, 13 patients (17%) experienced locoregional failure, which seems unexpectedly high compared to other contemporary series; this raises the question whether volume-based adaption based on the second CT scan is safe. Yu et al. (26) studied pre-treatment MRI of 70 NPC patients and identified several pre-treatment MRI-based radiomic features (2 shape, 3 texture and 1 first-order features) from the GTV that suggested promising capability of identifying a subset of NPC patients who may benefit from ART. However, whether these features are a surrogate for GTV or truly independent additional features remains to be validated.

Challenges and opportunities for ART

Several challenges exist in implementing ART in routine clinical practice, including accuracy in image registration and dose accumulation, resource-demanding image acquisition, labor-intensive and time-consuming recontouring and replanning, and streamlining optimal ART workflow.

Accuracy in image registration is pivotal for assessing dose accumulation. Since any subsequent CT scan would have a different clinical target volume and normal tissue volume shapes, deformable image registration (DIR) is often preferred over rigid registration to obtain a better estimation of accumulated dose (31). However, registration errors could still exist in DIR, especially for structures that are small with lack of contrast with the background (e.g., air spaces, such as nasal cavity and paranasal sinuses), which could result in significant dosimetric deviation relating to target volumes and OARs, especially in spinal cord and optic apparatus in some NPC patients (32). The accuracy of DIR also depends on the DIR methods and interface area (33). Currently, several DIR algorithms are under investigation, which use different transformation frameworks, DIR registration algorithms, and mapping direction (34).

Currently, ART requires reimaging, recontouring and replanning using a diagnostic quality scan (e.g., planning CT). Since NPC patients often require daily volumetric imaging for setup verification, which could provide another potential source for dose calculation. However, the quality of verification volumetric images is still suboptimal and subject to noise and artifacts which could results in errors and uncertainties for deformable registration (35,36). In addition, the field of view of verification imaging is often narrow and unable to capture the anatomical information of all LNs in NPC patients, which also a limitation of using them for ART (37).

One of the most labor-intensive and time-consuming steps in ART is manually contouring the target volumes and OARs (38). Auto-contouring software has the potential to enhance the efficiency of ART and reduce the variation among radiation oncologists (39,40). Several vendors are developing auto-segmentation software for clinical use of ART; however, they are not available yet for routine clinical use in NPC due to the complexity of the anatomy of this location of the head and neck region and minimal tissue density difference for satisfactory auto-segmentation. Studies by Fung et al. (38) showed that auto contouring OARs could reduce the total replanning time by more than 30%, and the geometrical discrepancies between the auto- and manual contours were insignificant when compared to inter-observer variations. However, the dosimetric impacts of such contour differences could still be substantial in some NPC patients. This suggests the need for manual review and edit of auto-contours in a real clinical setting, which may not always be a time-saving measure compared to traditional approaches. Studies have shown that atlas-based auto-segmentation for OARs and neck volumes are feasible, but human intervention and quality assurance is also required (41-44).

Conclusions

NPC patients remain a vulnerable group from the standpoint of anatomical changes during a 6–7-week course of IMRT. ART shows promising potential to reduce toxicities while enhancing LRC. However, ART is currently still at an early stage of development in terms of precise method, workflow, and clinical implementation. ART is yet to be implemented routinely in clinical practice for all NPC patients since it is a time consuming and labor-intensive process. Timing and thresholds to trigger reactive ART remain active research areas. ART is most frequently implemented between 15–25 fractions over a course of 6–7 weeks radiotherapy to take into account potential significant anatomical changes and also allow sufficient time to adapt. Computer-assisted auto-contouring seems promising to address the labor-intensive aspect of ART; however, it is still at its nascent stage of development and further refinement is warranted. Caution must be taken when performing DIR and dose accumulation for cases with significant volume changes (45-47). Rigorous quality assurance should be implemented to assess the accuracy of auto-contouring for OARs and target volumes. Technical advances, such as machine learning and artificial intelligence to refine deformable registration, dose accumulation, and auto-contouring algorithms, may pave the way for adopting pragmatic approaches in implementing ART routinely for NPC patients; this could further improve their oncologic and functional outcomes.

Acknowledgments

Funding: We acknowledge the O. Harold Warwick Prize of the Canadian Cancer Society for supporting the author’s (B O’Sullivan) academic activities. We also acknowledge the Sanming Project of Medicine in Shenzhen Fund (SZSM201612024) for supporting J Li, Z Xu, B O’Sullivan, SH Huang academic activities. Finally, we acknowledge the Bartley-Smith/Wharton, the Gordon Tozer, the Wharton Head and Neck Translational, the Dr. Mariano Elia, the Petersen-Turofsky Funds, and the “Joe & Cara Finley Center for Head & Neck Cancer Research”, the “Discovery Fund” at the Princess Margaret Cancer Foundation for supporting research by B O’Sullivan, SH Huang, and A Pilar.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Annals of Nasopharynx Cancer for the series “Precision Radiotherapy in Nasopharyngeal Carcinoma”. The article has undergone external peer review.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at http://dx.doi.org/10.21037/anpc.2020.03.01). The series “Precision Radiotherapy in Nasopharyngeal Carcinoma” was commissioned by the editorial office without any funding or sponsorship. SHH served as the unpaid Guest Editor of the series and serves as an unpaid editorial board member of Annals of Nasopharynx Cancer. BOS serves as an unpaid editorial board member of Annals of Nasopharynx Cancer. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Ng WT, Lee MC, Hung WM, et al. Clinical outcomes and patterns of failure after intensity-modulated radiotherapy for nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2011;79:420-8. [Crossref] [PubMed]
2. Grégoire V, De Neve W, Eisbruch A, et al. Intensity-modulated radiation therapy for head and neck carcinoma. Oncologist 2007;12:555-64. [Crossref] [PubMed]
3. Brouwer CL, Steenbakkers RJ, Langendijk JA, et al. Identifying patients who may benefit from adaptive radiotherapy: Does the literature on anatomic and dosimetric changes in head and neck organs at risk during radiotherapy provide information to help? Radiother Oncol 2015;115:285-94. [Crossref] [PubMed]
4. Zhao L, Wan Q, Zhou Y, et al. The role of replanning in fractionated intensity modulated radiotherapy for nasopharyngeal carcinoma. Radiother Oncol 2011;98:23-7. [Crossref] [PubMed]
5. Capelle L, Mackenzie M, Field C, et al. Adaptive radiotherapy using helical tomotherapy for head and neck cancer in definitive and postoperative settings: initial results. Clin Oncol (R Coll Radiol) 2012;24:208-15. [Crossref] [PubMed]
6. Yan D, Vicini F, Wong J, et al. Adaptive radiation therapy. Phys Med Biol 1997;42:123-32. [Crossref] [PubMed]
7. Yan D, Wong J, Vicini F, et al. Adaptive modification of treatment planning to minimize the deleterious effects of treatment setup errors. Int J Radiat Oncol Biol Phys 1997;38:197-206. [Crossref] [PubMed]
8. Yan D. Adaptive radiotherapy: merging principle into clinical practice. Semin Radiat Oncol 2010;20:79-83. [Crossref] [PubMed]
9. Heukelom J, Fuller CD. Head and Neck Cancer Adaptive Radiation Therapy (ART): Conceptual Considerations for the Informed Clinician. Semin Radiat Oncol 2019;29:258-73. [Crossref] [PubMed]
10. Mnejja W, Daoud H, Fourati N, et al. Dosimetric impact on changes in target volumes during intensity-modulated radiotherapy for nasopharyngeal carcinoma. Rep Pract Oncol Radiother 2020;25:41-5. [Crossref] [PubMed]
11. Cheng HC, Wu VW, Ngan RK, et al. A prospective study on volumetric and dosimetric changes during intensity-modulated radiotherapy for nasopharyngeal carcinoma patients. Radiother Oncol 2012;104:317-23. [Crossref] [PubMed]
12. Chitapanarux I, Chomprasert K, Nobnaop W, et al. A dosimetric comparison of two-phase adaptive intensity-modulated radiotherapy for locally advanced nasopharyngeal cancer. J Radiat Res 2015;56:529-38. [Crossref] [PubMed]
13. Wang RH, Zhang SX, Zhou LH, et al. Volume and dosimetric variations during two-phase adaptive intensity-modulated radiotherapy for locally advanced nasopharyngeal carcinoma. Biomed Mater Eng 2014;24:1217-25. [Crossref] [PubMed]
14. Bahl A, Elangovan A, Dracham CB, et al. Analysis of volumetric and dosimetric changes in mid treatment CT scan in carcinoma nasopharynx: implications for adaptive radiotherapy. J Exp Ther Oncol 2019;13:33-9. [PubMed]
15. Deng S, Liu X, Lu H, et al. Three-Phase Adaptive Radiation Therapy for Patients With Nasopharyngeal Carcinoma Undergoing Intensity-Modulated Radiation Therapy: Dosimetric Analysis. Technol Cancer Res Treat 2017; [Crossref] [PubMed]
16. Luo Y, Qin Y, Lang J. Effect of adaptive replanning in patients with locally advanced nasopharyngeal carcinoma treated by intensity-modulated radiotherapy: a propensity score matched analysis. Clin Transl Oncol 2017;19:470-6. [Crossref] [PubMed]
17. Wang W, Yang H, Hu W, et al. Clinical study of the necessity of replanning before the 25th fraction during the course of intensity-modulated radiotherapy for patients with nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2010;77:617-21. [Crossref] [PubMed]
18. Yang H, Hu W, Wang W, et al. Replanning during intensity modulated radiation therapy improved quality of life in patients with nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2013;85:e47-54. [Crossref] [PubMed]
19. Lu J, Ma Y, Chen J, et al. Assessment of anatomical and dosimetric changes by a deformable registration method during the course of intensity-modulated radiotherapy for nasopharyngeal carcinoma. Journal of radiation research 2014;55:97-104. [Crossref] [PubMed]
20. Hu YC, Tsai KW, Lee CC, et al. Which nasopharyngeal cancer patients need adaptive radiotherapy? BMC Cancer 2018;18:1234. [Crossref] [PubMed]
21. Bhide SA, Davies M, Burke K, et al. Weekly volume and dosimetric changes during chemoradiotherapy with intensity-modulated radiation therapy for head and neck cancer: a prospective observational study. Int J Radiat Oncol Biol Phys 2010;76:1360-8. [Crossref] [PubMed]
22. Brown E, Owen R, Harden F, et al. Head and neck adaptive radiotherapy: Predicting the time to replan. Asia Pac J Clin Oncol 2016;12:460-7. [Crossref] [PubMed]
23. Gai X, Wei Y, Tao H, et al. Clinical study of the time of repeated computed tomography and replanning for patients with nasopharyngeal carcinoma. Oncotarget 2017;8:27529-40. [Crossref] [PubMed]
24. Huang H, Lu H, Feng G, et al. Determining appropriate timing of adaptive radiation therapy for nasopharyngeal carcinoma during intensity-modulated radiation therapy. Radiat Oncol 2015;10:192. [Crossref] [PubMed]
25. Yao WR, Xu SP, Liu B, et al. Replanning Criteria and Timing Definition for Parotid Protection-Based Adaptive Radiation Therapy in Nasopharyngeal Carcinoma. Biomed Res Int 2015;2015:476383 [Crossref] [PubMed]
26. Yu TT, Lam SK, To LH, et al. Pretreatment Prediction of Adaptive Radiation Therapy Eligibility Using MRI-Based Radiomics for Advanced Nasopharyngeal Carcinoma Patients. Front Oncol 2019;9:1050. [Crossref] [PubMed]
27. Marzi S, Pinnaro P, D'Alessio D, et al. Anatomical and dose changes of gross tumour volume and parotid glands for head and neck cancer patients during intensity-modulated radiotherapy: effect on the probability of xerostomia incidence. Clin Oncol (R Coll Radiol) 2012;24:e54-62. [Crossref] [PubMed]
28. Ren G, Xu SP, Du L, et al. Actual anatomical and dosimetric changes of parotid glands in nasopharyngeal carcinoma patients during intensity modulated radiation therapy. Biomed Res Int 2015;2015:670327 [Crossref] [PubMed]
29. Yan D, Yan S, Wang Q, et al. Predictors for replanning in loco-regionally advanced nasopharyngeal carcinoma patients undergoing intensity-modulated radiation therapy: a prospective observational study. BMC Cancer 2013;13:548. [Crossref] [PubMed]
30. Brown E, Owen R, Harden F, et al. Predicting the need for adaptive radiotherapy in head and neck cancer. Radiother Oncol 2015;116:57-63. [Crossref] [PubMed]
31. Kai Y, Arimura H, Toya R, et al. Comparison of rigid and deformable image registration for nasopharyngeal carcinoma radiotherapy planning with diagnostic position PET/CT. Jpn J Radiol 2020;38:256-64. [PubMed]
32. Poon M, Holborn C, Cheng KF, et al. Evaluation of deformed image-based dose calculations for adaptive radiotherapy of nasopharyngeal carcinoma. Med Dosim 2017;42:273-81. [Crossref] [PubMed]
33. Nobnop W, Chitapanarux I, Neamin H, et al. Evaluation of Deformable Image Registration (DIR) Methods for Dose Accumulation in Nasopharyngeal Cancer Patients during Radiotherapy. Radiol Oncol 2017;51:438-46. [Crossref] [PubMed]
34. Nobnop W, Neamin H, Chitapanarux I, et al. Accuracy of eight deformable image registration (DIR) methods for tomotherapy megavoltage computed tomography (MVCT) images. J Med Radiat Sci 2017;64:290-8. [Crossref] [PubMed]
35. Lawson JD, Schreibmann E, Jani AB, et al. Quantitative evaluation of a cone-beam computed tomography-planning computed tomography deformable image registration method for adaptive radiation therapy. J Appl Clin Med Phys 2007;8:2432. [Crossref] [PubMed]
36. Li Y, Zhu J, Liu Z, et al. A preliminary study of using a deep convolution neural network to generate synthesized CT images based on CBCT for adaptive radiotherapy of nasopharyngeal carcinoma. Phys Med Biol 2019;64:145010 [Crossref] [PubMed]
37. MacFarlane M, Wong D, Hoover DA, et al. Patient-specific calibration of cone-beam computed tomography data sets for radiotherapy dose calculations and treatment plan assessment. J Appl Clin Med Phys 2018;19:249-57. [Crossref] [PubMed]
38. Fung NTC, Hung WM, Sze CK, et al. Automatic segmentation for adaptive planning in nasopharyngeal carcinoma IMRT: Time, geometrical, and dosimetric analysis. Med Dosim 2020;45:60-5. [Crossref] [PubMed]
39. Chao KS, Bhide S, Chen H, et al. Reduce in variation and improve efficiency of target volume delineation by a computer-assisted system using a deformable image registration approach. Int J Radiat Oncol Biol Phys 2007;68:1512-21. [Crossref] [PubMed]
40. Daoud B, Morooka K, Kurazume R, et al. 3D segmentation of nasopharyngeal carcinoma from CT images using cascade deep learning. Comput Med Imaging Graph 2019;77:101644 [Crossref] [PubMed]
41. Voet PW, Dirkx ML, Teguh DN, et al. Does atlas-based autosegmentation of neck levels require subsequent manual contour editing to avoid risk of severe target underdosage? A dosimetric analysis. Radiother Oncol 2011;98:373-7. [Crossref] [PubMed]
42. Walker GV, Awan M, Tao R, et al. Prospective randomized double-blind study of atlas-based organ-at-risk autosegmentation-assisted radiation planning in head and neck cancer. Radiother Oncol 2014;112:321-5. [Crossref] [PubMed]
43. Tong N, Gou S, Yang S, et al. Fully automatic multi-organ segmentation for head and neck cancer radiotherapy using shape representation model constrained fully convolutional neural networks. Med Phys 2018;45:4558-67. [Crossref] [PubMed]
44. Lim JY, Leech M. Use of auto-segmentation in the delineation of target volumes and organs at risk in head and neck. Acta Oncol 2016;55:799-806. [Crossref] [PubMed]
45. Zhong H, Chetty IJ. Caution Must Be Exercised When Performing Deformable Dose Accumulation for Tumors Undergoing Mass Changes During Fractionated Radiation Therapy. Int J Radiat Oncol Biol Phys 2017;97:182-3. [Crossref] [PubMed]
46. Joint Head and Neck MRI-Radiotherapy Development Cooperative. Prospective quantitative quality assurance and deformation estimation of MRI-CT image registration in simulation of head and neck radiotherapy patients. Clin Transl Radiat Oncol 2019;18:120-7. [Crossref] [PubMed]
47. Mohamed AS, Ruangskul MN, Awan MJ, et al. Quality assurance assessment of diagnostic and radiation therapy-simulation CT image registration for head and neck radiation therapy: anatomic region of interest-based comparison of rigid and deformable algorithms. Radiology 2015;274:752-63. [Crossref] [PubMed]