A novel applicator design for intracavitary brachytherapy of the nasopharynx: Simulated reconstruction, image-guided adaptive brachytherapy planning, and dosimetry

Brachytherapy

-

(2018)

-

A novel applicator design for intracavitary brachytherapy of the nasopharynx: Simulated reconstruction, image-guided adaptive brachytherapy planning, and dosimetry Warren R. Bacorro1,2,*, Ryan Anthony F. Agas1, Stellar Marie R. Cabrera3, Maureen R. Bojador3, Paolo G. Sogono1, Michael Benedict A. Mejia1, Teresa T. Sy Ortin1,4 1

Head and Neck Unit, Department of Radiation Oncology, University of Santo Tomas Hospital, Benavides Cancer Institute, Manila, Philippines Brachytherapy Unit, Department of Radiation Oncology, University of Santo Tomas Hospital, Benavides Cancer Institute, Manila, Philippines 3 Medical Physics, Department of Radiation Oncology, University of Santo Tomas Hospital, Benavides Cancer Institute, Manila, Philippines 4 Department of Radiological Sciences, Faculty of Medicine and Surgery, University of Santo Tomas, Manila, Philippines

2

ABSTRACT

PURPOSE: In nasopharyngeal cancer, brachytherapy is given as boost in primary treatment or as salvage for recurrent or persistent disease. The Rotterdam nasopharyngeal applicator (RNA) allows for suboptimal reduction of soft palate radiation dose, based on image-guided brachytherapy plans. Building on the RNA, we propose a novel design, the Benavides nasopharyngeal applicator (BNA). METHODS AND MATERIALS: The virtual BNA was reconstructed on two cases (one T1, one T2) previously treated with intracavitary brachytherapy using the RNA. Dose was prescribed to the high-risk clinical target volumes (CTVs) and optimization was such that high-risk CTV D90 $ 100% of prescribed dose (PD), intermediate-risk-CTV D90 $ 75% PD, and soft palate D2cc # 120% PD. The optimized RNA and BNA image-guided brachytherapy plans were compared in terms of CTV coverage and organs-at-risk sparing. RESULTS: Optimization objectives were more easily met with the BNA. For the T1 case, all three planning objectives were easily achieved in both the RNA and BNA, but with 18e19% lower soft palate doses with the BNA. For the T2 case, the CTV planning objectives were achieved in both the RNA and BNA, but the soft palate constraint was only achieved with the BNA, with 38e41% lower soft palate doses. CONCLUSIONS: Compared to the RNA, the BNA permits easier optimization and improves therapeutic ratio by a significant reduction of soft palate doses, based on simulation using a proposed system for CTV/organs-at-risk delineation, prescription, and optimization for image-guided adaptive brachytherapy. Clinical piloting using a prototype is necessary to evaluate its feasibility and utility. Ó 2018 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.

Keywords:

Nasopharyngeal cancer; Intracavitary brachytherapy; Image-guided adaptive brachytherapy; Applicator design

Introduction Nasopharyngeal carcinoma (NPC) is rare globally but is notable for its ethnographic endemicity with the highest incidence rates reported in Southeast Asia, Southeastern China, and India (1). Standard treatment consists of Received 22 January 2018; received in revised form 20 March 2018; accepted 21 March 2018. Conflict of interest: None to declare. * Corresponding author. Department of Radiation Oncology, University of Santo Tomas Hospital, Benavides Cancer Institute, Espa~na Boulevard, 1008 Manila, Philippines. Tel./fax: +632-731-3001x2615. E-mail address: [email protected] (W.R. Bacorro).

radiotherapy (RT) alone for Stage I and chemoradiotherapy (CRT) for Stage II to IVB disease (2e4). While intensitymodulated radiotherapy (IMRT) is the current standard for curative RT, brachytherapy being reserved mostly for salvage treatment of recurrences or persistent disease (5,6); the latter is still used in several centers as a boost after RT to dose escalate (7e10). Multiple methods for personalized applicators (11,12) and applicator designs have been developed for intracavitary brachytherapy (ICBT). Personalized applicators require important resources and expertise. Among commercial applicators, the most commonly used are the nasopharyngeal balloon applicators (13) and the Rotterdam nasopharyngeal

1538-4721/$ - see front matter Ó 2018 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.brachy.2018.03.004

2

W.R. Bacorro et al. / Brachytherapy

applicator (RNA) (14). The balloon applicator allows easy transnasal insertion but does not allow preferential displacement of the catheter away from the soft palate and closer to the nasopharynx. Moreover, it is not compatible with multiple treatments over several days as the relation of the balloon catheters is not stable or fixed. The RNA requires retrograde insertion through the oropharynx, which can be more difficult, but allows for a more stable geometry and may remain in place for several days. It consists of two silicone catheters fixed on a silicone base, which displaces the catheters away from the soft palate (14). Before the advent of three-dimensional (3D) imageguided adaptive brachytherapy (IGABT) techniques, the Levendag points, marked on two standard orthogonal films, were used in two-dimensional (2D) treatment planning for ICBT using the RNA (14). The use of axial imaging now allows for better delineation of target volumes and organs at risk. Based on our initial experience, the design of the RNA allows for suboptimal reduction of radiation dose to the soft palate. Furthermore, while the RNA was designed to remain in place for several days, there is no mechanism for lavage or transnasal delivery of medications (i.e., local or topical anesthesia or analgesic), which could significantly improve patient hygiene and comfort. Building on the RNA, we propose a novel applicator design, hereafter referred to as Benavides nasopharyngeal applicator (BNA), and a system for delineation of clinical target volumes (CTVs) and organs-at-risk (OARs), dose prescription, optimization, and reporting. We then compare the dosimetric properties of the BNA with that of the RNA.

Methods Novel design The proposed design comprising a silicone base, two silicone tubes, four plastic catheters, and a silicone flange designed to facilitate insertion, improve radiation dosimetry, and improve patient hygiene and comfort during treatment. The silicone base, spatulate, flexible, and 8-mm thick, is oriented such that the narrower end (8 mm) faces the nasal cavities and the broader end (12 mm) faces the oropharynx. The central length is 30 mm. The thickness and shape allow easy retrograde maneuvering through the nasooropharyngeal passage. A pocket is incorporated into the base, which can house a 3-mm thick lead or tungsten alloy shield to provide additional soft palate protection, which is desirable in cases of reirradiation. Each silicon tube, with 8.8-mm outer diameter and 6-mm inner diameter, is joined to the base laterally, along each side, such that there is a 15-mm long free limb on the oropharyngeal end and a 180-mm long free limb on the nasal end. The curved junction is along the level of a septum that divides each tube into superior and inferior compartments. A pair (a medial and a lateral) of semirigid plastic catheters, with 2-mm outer diameter, are housed in

-

(2018)

-

the superior compartment, running along the length of the tube, and fixed at the oropharyngeal end. The superior compartment, designed to have a catheter deploy-and-lock mechanism, is open at two segments, exposing the catheters. The anteriormost segment, roofed and 30-mm long, serves to secure the free ends of the catheters when in storage or not in use. The second segment, roofless and 50-mm long, provides for a mechanism for advancing each catheter forward by up to 30 mm and locking them in place. Along this length, in storage, the catheters have five locking beads 10-mm apart, the first bead situated at 30 mm from the free tip of the catheter. In storage, the length bounded by the fourth and fifth beads is secured by a locking mechanism. In use, each catheter may be advanced forward in 10-mm increments and locking it in place using the beads. The third segment, roofed, serves to keep the catheters within the tunnel such that advancing them results in a protrusion outside the tube only at the level of the fourth, roofless segment. The latter, 35 mm in length, runs along the curved junction between the tube and the base. Along this segment, the catheters are oriented such that when advanced, the medial catheter protrudes superiorly and the lateral catheter protrudes superolaterally, into the nasopharyngeal recess. The superior compartment terminates in a 15-mm roofed segment that secures the fixed ends of the catheters. The catheters are detachable and replaceable. The inferior compartment, designed to have a lavage system, is open on both ends and has five apertures on its inferior aspect along the junction with the base. The anterior end serves to receive a plastic tube, which is inserted through the nostril, into the oropharynx and inserted and fixed into the silicone tube by tying a knot using a silk suture. The nasal catheter is then retracted at the same time maneuvering the applicator through the oropharynx and into the nasopharynx. The apertures allow for lavage or delivery of medications (such as a local anesthetic). Once the applicator is positioned in the nasopharynx, the nasal limbs of the silicon tubes are secured together using a silicon flange at approximately midway the length of the limbs, such that the flange abuts the nasal columella. The catheters are then deployed and locked to desired lengths. Simulated reconstruction Two NPC patients, one T1 and one T2 tumor (American Joint Committee on Cancer seven classification), treated with ICBT boost using the RNA, were retrieved. For both cases, noncontrast 2-mm thick axial CT scans were coregistered with preexternal beam radiotherapy (EBRT) contrastenhanced T1-weighted MRI axial images. Two 3D plans were created for each patientdone with the reconstructed RNA and the other with the virtual BNA simulated, using the RNA base and tubes as references and reconstructing the medial and lateral catheters according to the above-detailed dimensions and relations, and patient anatomy.

W.R. Bacorro et al. / Brachytherapy

-

(2018)

-

3

Table 1 Comparative dosimetric profiles Rotterdam applicator Levendag 2D pointsa

Physical dose (Gy)

Case 1: T1 disease Prescription dose: 350 cGy Target Na(R)/Na(L) BOS(R)/BOS(L) R

3.4/2.6 1.0/1.1 6.0

Benavides applicator EQD2b (Gy)

IGBT volumes

Physical dose (Gy)

EQD2b (Gy)

3.9 4.3 4.6 2.2 2.7 3.0

4.5 5.1 5.6 2.2 2.9 3.3

HR-CTV D95 HR-CTV D90 HR-CTV D85 IR-CTV D95 IR-CTV D90 IR-CTV D85

3.9 4.4 4.7 2.2 2.6 3.0

4.5 5.3 5.8 2.2 2.7 3.3

0.9 0.6 0 0 0 0.1 2.0 1.6 1.3 1.5 3.9 3.1

0.7 0.4 0 0 0 0.1 2.0 1.5 1.1 1.4 5.4 3.8

Brainstem D2cc Spinal cord D2cc Pituitary D2cc Optic chiasm D2cc Retina right D2cc Retina left D2cc Clivus D2cc Clivus DMean Atlanto-axial joint D2cc Atlanto-axial joint DMean Soft palate D2cc Soft palate DMean

0.9 0.6 0 0 0 0.1 2.1 1.6 1.3 1.5 3.2 2.5

0.7 0.4 0 0 0 0.1 2.1 1.5 1.1 1.4 4.0 2.8

3.2 3.5 3.8 2.4 2.7 3.0

3.5 3.9 4.4 2.5 2.9 3.3

HR-CTV D95 HR-CTV D90 HR-CTV D85 IR-CTV D95 IR-CTV D90 IR-CTV D85

3.4 3.8 4.1 2.3 2.7 3.0

3.8 4.4 4.8 2.4 2.9 3.3

1.1 0 0 0 1.7 1.1 2.3 1.9 2.0 2.1 6.4 5.2

0.9 0 0 0 1.6 0.9 2.4 1.9 2.0 2.1 12.0 8.5

Brainstem D2cc Spinal cord D2cc Pituitary D2cc Optic chiasm D2cc Retina right D2cc Retina left D2cc Clivus D2cc Clivus DMean Atlanto-axial joint D2cc Atlanto-axial joint DMean Soft palate D2cc Soft palate DMean

1.0 0 0 0 1.7 1.1 2.4 1.9 1.7 1.8 3.8 3.2

0.8 0 0 0 1.6 0.9 2.6 1.9 1.6 1.7 5.2 4.0

Organs-at-risk C P OC Re(R) Re(L)

2.7 0.6 0.4 0.3 0.3

Pa(R)/Pa(L)

2.2/2.8

Case 2: T2 disease Prescription dose: 350 cGy Target Na(R)/Na(L) BOS(R)/BOS(L) R

3.8/4.0 2.7/2.6 6.3

Organs-at-risk C P OC Re(R) Re(L)

2.5 0.84 0.59 0.54 0.52

Pa(R)/PA(L)

4.5/4.9

2D 5 two-dimensional; EQD2 = equivalent dose in 2 Gy; ICBT 5 intracavitary brachytherapy; (L) 5 left; (R) 5 right; HR-CTV 5 high-risk CTV; IRCTV 5 intermediate-risk CTV; CTV 5 clinical target volume. a Nadnasopharynx (intersection of the Pa-BOS line with the bony outline of the base of skull), Pdpituitary (0.5 cm from center of sella), OCdoptic chiasm (1.5 cm ventrally from P), Redretina (1 cm posterior to the line drawn from the contralateral outer canthus and tragus), Cdcord (posterior to R at the posterior border of corpus C1), Rdnode of Rouviere (ventral part of corpus C1), Padpalate (junction of soft and hard palate), BOSdbase of skull (intersection of the line drawn between the anterior clinoid process and point R and the line drawn from the contralateral outer canthus and tragus), (Levendag, 1997(14)). b Equivalent dose in 2 Gy using ab ratio of 10 for the tumor, and ab ratio of three for organs-at-risk.

Delineation of CTV and OAR There is not yet a consensus on target volume delineation for IGABT for NPC. For our purposes, the gross tumor volume was delineated as the residual tumor on the CT simulation scan, the high-risk CTV (HR-CTV) as the gross tumor volume plus the whole nasopharynx (15); and the intermediate-risk

CTV (IR-CTV) as 5 mm around the HR-CTV plus the initial extent of disease, carving out bone and air. The brainstem, spinal cord, pituitary, optic chiasm, retina, clivus, atlanto-axial joint, and soft palate were contoured as OARs using the MRI as aid for delineation. The atlanto-axial joint was drawn using the following borders:

4

W.R. Bacorro et al. / Brachytherapy

-

(2018)

-

Fig. 1. Novel applicator design. For the labeled parts, refer to section Illustration of the mechanisms and clinical application of the novel applicator. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

upper limit of the dens (superior), 2 mm below the lower limit of the atlas (inferior), ventral limit of the dens and including the ligaments visualized on the MRI or corresponding area on CT (anterior), dorsal limit of the dens (posterior), and inner limits of the axis including the ligaments visualized on the MRI or corresponding area on CT (lateral). The soft palate was drawn from its junction with the hard palate down to its free border and along its junction with the tonsillar pillars, at least 6 mm below the inferior limit of the constructed catheters. Prescription, optimization, and dosimetry Dose prescription was to the HR-CTV. Graphical optimization was performed with the following planning objectives: HR-CTV D90% $ 100% of prescribed dose (PD) (high priority); IR-CTV D90% $ 75% PD (intermediate priority); and soft palate D2cc !120% PD or as low as possible (intermediate priority). Using the optimized RNA plans, the Levendag OAR point doses (legend in Table 1) (14) were plotted and compared with 3D OAR D2cc values. The dosimetric profiles of the RNA and BNA plans were then compared using the following parameters: HR-CTV D95%, D90%, and D85%; IR-CTV D95%, D90%, and D85%; D2cc of the brainstem, spinal cord, pituitary, optic chiasm, and retina; and D2cc and mean dose of the clivus, atlanto-axial joint, and soft palate. Treatment planning was conducted using Oncentra Brachy, Version 4.5.3, 2018, Elekta AB, Stockholm, Sweden.

Results Illustration of the mechanisms and clinical application of the novel applicator Figure 1A and 1B show the nasopharyngeal portion of the applicator viewed from above, in the storage (1A) and deployed (1B) states. It consists of the silicone base (101), and silicone tubes (102), and the medial (103) and lateral (104) plastic catheters. In storage, the catheters are housed inside the superior compartment (105) of the silicone tubes. Deployed, the medial and lateral catheters protrude out superiorly and superolaterally, respectively. The catheters are fixed within the oropharyngeal free limb (106). Figure 1C and 1D show the nasopharyngeal portion of the applicator viewed from below, with the soft palate shield (201) outside (1C) and inside (1D) the shield pocket (202) of the spatulate silicon base (203). The apertures (204) of the inferior compartment (205) of the silicone tube (206) are also seen along its junction with the base. Figure 1E and 1F show the nasopharyngeal (301), nasal (truncated) (302), and external (303) portions of the applicator viewed from the side, in storage (1E) and deployed (1F) states. The roofless fourth segment (304) serves as the window through which the catheters protrude. The roofless second segment (305) houses the lock (306) and the beaded (307) portions of the catheters. The beads are 10mm apart and thus allow advancement of the catheter in 10-mm increments. In Fig. 1F, the catheters are shown

W.R. Bacorro et al. / Brachytherapy

-

(2018)

-

5

Fig. 2. Reconstruction of the applicators, relation to the target volumes and organs-at-risk, and comparison of target volume coverage and soft palate sparing in a T1 tumor. The source dwell points are represented by red circles inside the reconstructed applicators in light blue. The HR-CTV is in dark pink, IR-CTV in dark blue mesh, brainstem in solid orange, clivus in solid light blue, atlanto-axial joint in solid yellow, and soft palate in solid beige. The 350 cGy isodose volume is in solid green. HR-CTV 5 high-risk CTV; IR-CTV 5 intermediate-risk CTV; CTV 5 clinical target volume. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

deployed to 30 mm. The free ends (308) of the catheters are secured in the roofed first segment (309) in storage and are exposed on deployment. The opening of the inferior compartment on the external end (310) serves to receive a plastic nasal catheter during insertion. Figure 1G and 1H show the applicator being inserted (1G) and in place (1H) in sagittal representations of the patient. A plastic nasal catheter (such as a feeding tube) (406) is inserted through the nostril (401) and into the oropharynx (405) and is then connected with the applicator (407) and secured with a silk suture (408). The nasal catheter is retracted from the nasal cavity at the same time maneuvering the applicator through the oropharynx and into the nasopharyngeal recess (410). The applicator is secured in place with the silicon flange (409), and the plastic catheters are deployed such that they abut the nasopharynx (403). The catheters are then connected to the brachytherapy projector machine (411) through a source tube (412). A syringe (413) may be used to inject into the inferior compartment saline solution for lavage (414) or medication. Simulated reconstruction and CTV/OAR delineation, prescription, and optimization Figures 2 and 3 show the reconstructed RNA and virtual BNA with the catheters deployed, within the nasopharyngeal

cavity. The catheters are deployed further away from the soft palate and closer to the target volume with the BNA. For the T1 case, all planning objectives were easily achieved in both the RNA and BNA plans. For the T2 case, the CTV planning objectives were achieved in both the RNA and BNA plans, but the soft palate constraint was only achieved with the BNA plan. Table 1 details the prescription and dosimetric profiles for the optimized plans using each applicator for each of the cases. Comparison of 2D points and 3D volume doses In the RNA plans for the T1 and T2 cases, the nasopharynx (Na) point doses underestimate (by 21e40%) and overestimate (by 9e14%) the HR-CTV D90, respectively. The base of skull (BOS) point doses underestimate the HR-CTV D90 by at least 74% and 23% in the T1 and T2 cases, respectively. In both cases, the spinal cord, pituitary, and optic chiasm receive negligible doses, but the C point doses overestimate the cord D2cc by at least 350%. In T2 disease, the retinae receive significant doses (D2cc 1.1e1.7 Gy), which are underestimated by the retina (Re) point doses by up to 68%. More importantly, in both cases, the soft palate receives very important doses (D2cc O 3 Gy), which are underestimated by the palate (Pa) point doses by 23e44%.

6

W.R. Bacorro et al. / Brachytherapy

-

(2018)

-

Fig. 3. Reconstruction of the applicators, relation to the target volumes and organs-at-risk, and comparison of target volume coverage and soft palate sparing in a T2 tumor. The source dwell points are represented by red circles inside the reconstructed applicators in light blue. The HR-CTV is in dark pink, IR-CTV in dark blue mesh, brainstem in solid orange, clivus in solid light blue, atlanto-axial joint in solid yellow, and soft palate in solid beige. The 350 cGy isodose volume is in solid green. HR-CTV 5 high-risk CTV; IR-CTV 5 intermediate-risk CTV; CTV 5 clinical target volume. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Comparison of optimized RNA and BNA IGABT plans For the T1 case, the OARs receive similar doses, but with lower soft palate D2cc (by 18%) and Dmean (by 19%) values with the BNA when compared to the RNA plan. For the T2 case, the BNA plan, compared to the RNA plan, has lower soft palate D2cc (by 41%) and Dmean (by 38%) values. Figure 3 shows the soft palate volumes (beige) encompassed within the prescription isodose volumes (red) in the optimized RNA and BNA plans for the T2 case. In both cases and both plans, the clivus and the atlantoaxial joint receive significant doses in the 1.6e2.4 Gy and 1.3e2.1 Gy ranges, respectively. Figure 4 shows the relative sparing of the soft palate, clivus, and atlanto-axial joint using the RNA and BNA for the T2 case.

Discussion Standard treatment of NPC consists of RT alone for Stage I and CRT for Stage II to IVB disease (2e4). While IMRT is the current standard for curative RT, brachytherapy may be used to dose escalate (7e10) and to salvage recurrences or persistent disease (5,6). A significant RT doseeresponse relationship was observed in retrospective studies among patients treated with 2D EBRT techniques, with better local control (LC) among

patients treated to $70Gy (16,17). A total dose of 77e81 Gy have been recommended if treating with RT alone (18). For T1eT2 disease, dose escalation by ICBT after EBRT has led not only to improved overall survival (OS) and LC but also decreased toxicity (7e9). For Stage III and IV disease, a multicenter trial examining dose escalation by ICBT after neoadjuvant chemotherapy and concurrent CRT did not demonstrate significantly improved outcomes, including LC for T1eT2 tumors, compared to the latter regimen alone (19). However, the poor outcomes in both arms, which are inferior in comparison to those of other published series, may be due to suboptimal 2D EBRT techniques. In the era of IMRT and CRT, excellent LC rates (up to 85.8% overall 8-year local failure free survival; T1: 91.7%, T2: 88.2%, T3: 87.2%; T4, 71.6%) (20) have been achieved leading to a decline in ICBT use (4). However, a recent study examining the benefit of dose escalation with ICBT after concurrent CRT using IMRT techniques, with or without adjuvant chemotherapy, among patients with T1eT3 tumors that have completely responded to IMRT, showed improved LC in T1 tumors with no apparent increase in local toxicity (10), supporting the benefit of dose escalation even in the era of IMRT and CRT. On the other hand, dose escalation using IMRT techniques exclusively in the setting of CRT is associated with increased toxicity (21e24), underscoring the fact that the highly localized

W.R. Bacorro et al. / Brachytherapy

Fig. 4. Comparison of soft palate sparing and isodose profiles of the Rotterdam and proposed Benavides applicator in a T2 Tumor in representative axial, sagittal, and coronal planes. Volumes: The HR-CTV is in red; IRCTV, dark blue; brainstem, orange; clivus, light blue; atlanto-axial joint, yellow; and soft palate, beige. Isodose lines: light pink, 700 cGy; dark pink, 525 cGy; red, 350 cGy; yellow, 315 cGy; green, 263 cGy; blue, 140 cGy. The red line represents the prescription dose, 350 cGy. HRCTV 5 high-risk CTV; IR-CTV 5 intermediate-risk CTV; CTV 5 clinical target volume. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

and conformal doses attainable with ICBT are difficult to achieve with IMRT techniques. Brachytherapy, alone or in combination with EBRT, is particularly useful in the reirradiation of recurrences due to stringent OAR dose limitations. Two retrospective studies reported 5-year LC and OS rates of 61e63% and 40e50%, respectively, with brachytherapy alone (25,26). Inferior outcomes (5-year locoregional failure and OS rates of 45e88% and 28e40%, respectively) with either interstitial (27) or intracavitary (28) brachytherapy boost have been reported, possibly relating to more advanced local recurrences. Major

-

(2018)

-

7

complication rates were still significant, reported in 35e45% of patients. On the other hand, a retrospective comparison of EBRT alone (median dose, 59 Gy) versus EBRT (median dose, 45 Gy) with ICBT boost (20 Gy low dose-rate) showed similar LC, event-free survival, and OS rates but lower Grade 3 toxicities (73% vs. 8%) with ICBT boost (6). Personalized applicators for ICBT of the nasopharynx require important resources and expertise (11,12), and multiple commercial applicators have been developed. The balloon applicator, inserted transnasally (13), is fixed in place by inflating a balloon concentrically around the catheter. This does not allow preferential displacement of the catheter away from the soft palate and toward the nasopharynx. Eccentrically placed balloons have been proposed to allow better apposition of the catheter against the nasopharynx (29). These applicators are not compatible with multiple treatments over several days as the relation of the balloon catheters is not stable or fixed. The RNA, which is inserted in a retrograde fashion through the oropharynx, may remain in place for up to 6 days of treatment. It consists of two silicone catheters fixed on a silicone base that maintains a curved path following the curvature of the nasopharyngeal recess and displaces the catheters away from the soft palate. The applicator placement is secured externally using a silicone flange. A newer design features displacement of the two catheters laterally and away from each other allowing a better coverage of the nasopharyngeal recess (14). In both designs, however, the catheters are not optimally displaced toward the recess as they are fixed on the base, the maximum height of which is limited by the diameter of the nasooropharyngeal passage, that is, the passage bounded by the posterior and lateral walls of the pharynx and the soft palate. As there is not yet a consensus on CTV and OAR delineation for IGABT for the nasopharynx, we have devised a system incorporating that previously employed by Ren et al (15). and the principles underlying the Groupe Europeen de Curietherapie-European Society for Radiotherapy and Oncology recommendations for IGABT for the cervix (30). Instead of using the Na, BOS, and R target points defined in the Levendag system, we prescribe to the HR-CTV and IR-CTV and optimize such that the 90% volumes receive $100% and 75% of the PD, respectively, and that the soft palate (due to its close proximity to the applicator) D2cc is kept #120% of the PD or as low as possible. In place of the C, P, optic chiasm, Re, and Pa OAR points, we monitor the D2cc for the delineated cord, pituitary, optic chiasm, retina, and soft palate volumes. To compare the two systems, we plotted the Levendag target and OAR points in the optimized RNA plans. The Na and BOS point doses correlate poorly with the HRCTV D90. More importantly, the former overestimates the HR-CTV D90 by 9e14% in the T2 case, which could lead to clinically significant HR-CTV underdosage. The C points overestimate the D2cc for the spinal cord, which in

8

W.R. Bacorro et al. / Brachytherapy

both cases receive negligible doses. Similarly, the pituitary and optic chiasm receive negligible doses in both cases; we believe that monitoring for these OARs are optional and that the D2cc would suffice. In the T2 case, the retina D2cc is significant but is underestimated by the Re point dose; we recommend routine monitoring of the retina D2cc. Significant volumes of the soft palate receive important doses (both D2cc and Dmean O3 Gy), which are underestimated by the Pa point. Similarly, significant volumes of the clivus and the atlanto-axial joint receive significant doses (D2cc and Dmean in the 1.3e2.3 Gy range) but are not monitored in OARs in the Levendag system. For these OARs that are in close proximity to the applicator, we recommend routine D2cc and Dmean monitoring and optimizing to keep the doses to minimum. In cases of reirradiation, we recommend routine monitoring and minimization of doses to all of the above OARs. The design of the RNA allows for suboptimal reduction of soft palate dose and low therapeutic ratios. In the optimized T2 case, the soft palate D2cc and Dmean are 1.8 and 1.5 the PD, which is even more concerning if the equivalent dose in 2 Gy are to be considered (12 and 8.5 Gy, respectively). The BNA, with a four-catheter design and a deploy-and-lock mechanism, was developed to allow for a more tailorable geometry (each catheter may be deployed to varying lengths), for greater displacement of the catheters toward the nasopharynx and away from the soft palate, for more source dwelling paths and points for better dosimetric optimization, as well as for placement of a lead or tungsten shield to further reduce soft palate dose. To evaluate its dosimetry, we performed a simulation of the BNA using the RNA as a reference. Compared with the RNA, optimization for the HR-CTV, IR-CTV, and soft palate constraints were easier in both T1 and T2 cases. In the T1 case, the BNA, compared to the RNA, allowed for reduction of the soft palate D2cc and Dmean by 18% and 19%, respectively. In the T2 case, the BNA allowed for adequate dose delivery to the HR-CTV and IR-CTV while respecting the soft palate dose limit, which was not possible with the RNA. In the T2 case, reduction of the soft palate D2cc and Dmean by 41% and 38%, translates to a reduction of the corresponding equivalent dose in 2 Gy by 57% and 53%, respectively. Soft palate dose reduction is important as fistula or necrosis can lead to velopharyngeal insufficiency, especially in the recurrent setting, in which rates of up to 16e18% have been reported (25,26). Use of a soft palate shield during reirradiation with ICBT could further protect the soft palate. Osteoradionecrosis of the cervical vertebra, specifically the atlanto-axial joint, is a late toxicity that is particularly attributed to brachytherapy and occurs mostly after reirradiation with brachytherapy for recurrent disease (31,32). Moreover, in case of cervical vertebral osteoradionecrosis, brachytherapy reirradiation was found to be a significant predictor for the need for surgical intervention in both univariate and multivariate analyses (33). For this reason, we

-

(2018)

-

recommend routine monitoring of the atlanto-axial joint, as well as clival, D2cc and Dmean values. While the RNA was designed to remain in place for several days, there is no mechanism for lavage or transnasal delivery of medications (i.e., local or topical anesthesia or analgesic). The novel design allows for easier retrograde insertion (spatulate, thinner, and more flexible base) and has a system for lavage and delivery of medication, which can improve hygiene and patient comfort while the applicator remains in place for several days. The essential feature of this design is the integration of the medial and lateral plastic catheters that can be deployed superiorly and superolaterally, respectively, and of the pocket that can accommodate a shielding material for the soft palate. An optional feature is the integration of the lavage system. A possible incarnation is integration of the catheters alone. Preferably, the catheters are detachable to facilitate cleaning and replacement. Integration of both the catheters and the shield pocket is preferable for treating NPC recurrences after previous RT. A third embodiment is integration of the catheters, the shield pocket, and the lavage system. This is most useful when ICBT is employed as a monotherapy for NPC recurrence, in which case a more significant dose is to be delivered over several days. If it is to be used, the lead or tungsten shield is inserted into the base pocket before insertion of the applicator. Prototypes shall be necessary to evaluate clinical feasibility, safety, and utility, according to the guidelines defined by the American Association of Physicists in Medicine and the Groupe Europeen de Curietherapie-European Society for Radiotherapy and Oncology (34).

Conclusion The BNA, a novel four-catheter design with a deployand-lock mechanism and shield pocket is described. Simulated reconstruction of the BNA and dosimetry based on a proposed system for CTV/OAR delineation, prescription, and optimization for 3D-IGABT reveal more favorable profiles compared to the RNA. Clinical piloting using prototypes is necessary to evaluate its feasibility and utility.

Acknowledgment Dr. Warren Bacorro would like to thank Dr. Lam Duc Hoang of the Ho Chi Minh Oncology Center for his mentorship and Atty. Jude Amadeus Recio Marfil for his assistance in this project. References [1] Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin 2015;65:87e108.

W.R. Bacorro et al. / Brachytherapy [2] Al-Sarraf M, LeBlanc M, Giri PG, et al. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized Intergroup study 0099. J Clin Oncol 1998;16:1310e1317. [3] Wee J, Tan EH, Tai BC, et al. Randomized trial of radiotherapy versus concur- rent chemoradiotherapy followed by adjuvant chemotherapy in patients with American Joint Committee on Cancer/International Union against Cancer stage III and IV nasopharyngeal cancer of the endemic variety. J Clin Oncol 2005;23:6730e6738. [4] Lee AW, Ma BYM, Ng WT, et al. Management of nasopharyngeal carcinoma: current practice and future perspective. J Clin Oncol 2015;33:3356e3364. [5] Yu KH, Leung SF, Tung SY, et al. Survival outcome of patients with nasopharyngeal carcinoma with first local failure: A study by the Hong Kong Nasopharyngeal Carcinoma Study Group. Head Neck 2005;27:397e405. [6] Koutcher L, Lee N, Zelefsky M, et al. Reirradiation of locally recurrent nasopharynx cancer with external beam radiotherapy with or without brachytherapy. Int J Radiat Oncol Biol Phys 2010;76:130e137. [7] Leung TW, Wong VYW, Sze WK, et al. High-dose-rate intracavitary brachytherapy boost for early T stage nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2008;70:361e367. [8] Yeo R, Fong KW, Hee SW, et al. Brachytherapy boost for T1/T2 nasopharyngeal carcinoma. Head Neck 2009;31:1610e1618. [9] Wu J, Guo Q, Lu JJ, et al. Addition of intracavitary brachytherapy to external beam radiation therapy for T1eT2 nasopharyngeal carcinoma. Brachytherapy 2013;12:479e486. [10] Chao H, Liu S, Chih-Cheng T, et al. Dose escalation via brachytherapy boost for nasopharyngeal carcinoma in the era of intensitymodulated radiation therapy and combined chemotherapy. J Radiat Res 2017;58:654e660. [11] Mazeron JJ, Limbergen E. Nasopharynx. In: Gerbaulet A, Potter R, Mazeron JJ, et al, editors. The GEC-ESTRO handbook of brachytherapy. Leuven, Belgium: ESTRO; 2002. p. 289e297. [12] Law SCK, Lam WK, Ng MF, et al. Reirradiation of nasopharyngeal carcinoma with intracavitary mold brachytherapy: An effective means of local salvage. Int J Radiat Oncol Biol Phys 2002;54: 1095e1113. [13] McLean M, Chow E, O’Sullivan B, et al. Re-irradiation for locally recurrent nasopharyngeal carcinoma. Radiother Oncol 1998;48: 209e211. [14] Levendag PC, Peters R, Meeuwis CA, et al. A new applicator design for endocavitary brachytherapy of cancer in the nasopharynx. Radiother Oncol 1997;45:95e98. [15] Ren YF, Zhao QC, Liu H, et al. 3D-image-guided HDRbrachytherapy versus 2D HDR - brachytherapy after external beam radiotherapy for early T-stage nasopharyngeal carcinoma. BMC Cancer 2014;14:894. [16] Mesic J, Fletcher G, Goepfert H, et al. Megavoltage irradiation of epithelial tumors of the nasopharynx. Int J Radiat Oncol Biol Phys 1981;7:447e453. [17] Perez CA, Devineni VR, Marcial-Vega V, et al. Carcinoma of the nasopharynx: Factors affecting prognosis. Int J Radiat Oncol Biol Phys 1992;23:271e280. [18] Levendag PC, Lagerwaard FJ, Noever I, et al. Role of endocavitary brachytherapy with or without chemotherapy in cancer of the nasopharynx. Int J Radiat Oncol Biol Phys 2002;52:755e768.

-

(2018)

-

9

[19] Rosenblatt E, Abdel-Wahab M, El-Gantiry M, et al. Brachytherapy boost in loco-regionally advanced nasopharyngeal carcinoma: A prospective randomized trial of the International Atomic Energy Agency. Radiat Oncol 2014;9:67. [20] Au KH, Ngan RKC, Ng AWY, et al. Treatment outcomes of nasopharyngeal carcinoma in modern era after intensity modulated radiotherapy (IMRT) in Hong Kong: A report of 3328 patients (HKNPCSG 1301 study). Oral Oncol 2018;77:16e21. [21] Teo PML, Leung SF, Chan ATC, et al. Final report of a randomized trial on altered-fractionated radiotherapy in nasopharyngeal carcinoma prematurely terminated by significant increase in neurological complications. Int J Radiat Oncol Biol Phys 2000;48:1311e1322. 21. [22] Lin JC, Liang WM, Jan JS, et al. Another way to estimate outcome of advanced nasopharyngeal carcinomadIs concurrent chemoradiotherapy adequate. Int J Radiat Oncol Biol Phys 2004;60:156e164. 22. [23] Wolden SL, Zelefsky MJ, Kraus DH, et al. Accelerated concomitant boost radiotherapy and chemotherapy for advanced nasopharyngeal carcinoma. J Clin Oncol 2001;19:1105e1110. [24] Lee AW, Tung SY, Chan AT, et al. Preliminary results of a randomized study (NPC- 9902 Trial) on therapeutic gain by concurrent chemotherapy and/or accelerated fractionation for locally advanced nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2006;66: 142e151. [25] Kwong D, Wel W, Cheng A, et al. Long term results of radioactive gold grain implantation for the treatment of persistent and recurrent nasopharyngeal carcinoma. Cancer 2001;15:1105e1113. [26] Choy D, Sham J, Wei W, et al. Transpalatal insertion of radioactive gold grain for treatment of persistent and recurrent nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 1993;25:505e512. [27] Syed A, Puthawala A, Damore S, et al. Brachytherapy for primary and recurrent nasopharyngeal carcinoma: 20 Years’ experience at long beach memorial. Int J Radiat Oncol Biol Phys 2000;47: 1311e1321. [28] Cheah S, Lau F, Yusof M, et al. Treatment outcome with brachytherapy for recurrent nasopharyngeal carcinoma. Asian Pac J Cancer Prev 2014;14:6513e6518. [29] Chang Y. Device for applying media in cavum nasopharyngeum. Chinese Patent No. CN 2426743Y. China: SIPO; 2001. [30] Potter R, Haie-Meder C, Limbergen E, et al. Recommendations from Gynaecological (GYN) GEC ESTRO Working Group (II): Concepts and terms in 3D image-based treatment planning in cervix cancer brachytherapyd3D dose volume parameters and aspects of 3D image-based anatomy, radiation physics, radiobiology. Radiother Oncol 2006;78:67e77. [31] Tiruchelvarayan R, Lee K, Ng I, et al. Surgery for atlanto-axial C1-2 involvement or instability in nasopharyngeal carcinoma patients. Review Article. Singapore Med J 2012;53:416e421. [32] King A, Griffith J, Abrigo J, et al. Osteoradionecrosis of the upper cervical spine: MR imaging following radiotherapy for nasopharyngeal carcinoma. Eur J Radiol 2010;73:629e635. [33] Huang W, Wong S, Chan J, et al. Role of surgery in the treatment of osteoradionecrosis and its complications after radiotherapy for nasopharyngeal carcinoma. Head Neck 2018;40:369e376. [34] Nath R, Rivard MJ, DeWerd LA, et al. Guidelines by the AAPM and GEC-ESTRO on the use of innovative brachytherapy devices and applications: report of Task Group 167. Med Phys 2016;43:3178e3205.