eISSN: 2081-2841
ISSN: 1689-832X
Journal of Contemporary Brachytherapy
Current Issue Archive Supplements Articles in Press Journal Information Aims and Scope Editorial Office Editorial Board Register as Author Register as Reviewer Instructions for Authors Abstracting and indexing Subscription Advertising Information Links
SCImago Journal & Country Rank

vol. 12
Original paper

Image-guided interstitial brachytherapy boost for nasopharyngeal carcinoma: technical aspects

Naoya Murakami
Guanghui Cheng
Seiichi Yoshimoto
Jun Itami

Department of Radiation Oncology, National Cancer Center Hospital, Tokyo, Japan
Department of Radiation Oncology, China-Japan Union Hospital of Jilin University, Changchun, China
Department of Head and Neck Surgery, National Cancer Center Hospital, Japan
J Contemp Brachytherapy 2020; 12, 3: 294–302
Online publish date: 2020/06/30
Article file
Get citation
JabRef, Mendeley
Papers, Reference Manager, RefWorks, Zotero


Because of its anatomical complexity, it is generally difficult to safely insert interstitial brachytherapy applicators in the head and neck region compared to other anatomical sites such as prostate, breast, or uterine cervix. Therefore, interstitial brachytherapy (ISBT) is not widely performed in the head and neck region except for some expert institutions. In the National Comprehensive Cancer Network guidelines for head and neck cancers version 1.2020, it is stated that brachytherapy is now being used less often because of improved local control obtained with concurrent chemoradiation, and brachytherapy has a role for lip and oral cavity cancers [1]. Lack of efficient training system availability for head and neck ISBT, and the emergence of more sophisticated delivery of external radiotherapy such as intensity-modulated radiation therapy or volumetric modulated arc therapy, with the effectiveness of concurrent chemoradiotherapy also contributed to diminished usage of head and neck ISBT. There have been reports on the head and neck ISBT such as early-stage tongue cancer [2,3,4,5,6,7,8], oral cavity cancers [9,10,11,12], oropharyngeal cancers [5,13,14], or nasal vestibules [15], and its usefulness on local control and survival has been demonstrated. The vast majority of nasopharyngeal cancer (NPC) consists of non-keratinizing squamous cell carcinoma (NKSCC) and it is well-known that carcinogenesis of NKSCC is related to Epstein-Barr virus (EBV) infection [16,17]. While most NKSCC NPC responds well to radiation therapy or chemotherapy, it has been reported that unfortunately, approximately 7.5-14% of locally advanced NPC patients develop persistent or local relapse after definitive concurrent chemoradiotherapy (CCRT) [18,19,20]. Prognosis of such patients with local relapse is generally dismal due to lack of not safe and effective salvage treatment option for such patients. Another potentially promising approach can be adding boost dose after external beam radiation therapy (EBRT) by brachytherapy for tumors responded poorly against EBRT. Traditionally, intracavitary brachytherapy (ICBT) has been used for local boosting after EBRT for NPC [21,22,23,24,25]. However, because in ICBT, the radioactive source stays in the nasopharyngeal cavity, not directly in the tumor tissue. Therefore, it is usually difficult to deliver adequate dose deeper than 5 mm beneath the surface of cavity. Recently, a new technique of high-dose-rate interstitial brachytherapy (HDR-ISBT) for NPC as a boost has been reported to escalate local dose, while minimizing unnecessary irradiation to the surrounding normal tissues [26,27,28]. In the following, we describe technical aspects of boost HDR-ISBT as dose escalation for NPC patients who respond poorly to the CCRT with EBRT, not salvage HDR-ISBT for tumor persistence or local recurrence.


Tumor response to EBRT should be monitored weekly by the usual fiber-optic endoscope. Typically, EBRT is delivered with conventional fractionation of 1.8-2 Gy by intensity-modulated radiation therapy (IMRT) or volumetric modulated arc therapy (VMAT), not to miss the target volume, which correlates to tumor recurrence. When a poor response is suspected, suggesting high probability of residual tumor even after the completion of planned irradiation, magnetic resonance imaging (MRI) should be performed after 50 Gy EBRT or more. If a definite residual tumor is pointed out, boost HDR-ISBT can be attempted after obtaining informed consent from a patient. Although optimal timing of assessing residual disease for boost HDR-ISBT for NPC is not yet decided, guidelines for head and neck cancers stated that in combination of EBRT and ISBT for oropharyngeal cancer, in which ISBT is used as a boost, ISBT is applied 1-2 weeks after EBRT completion and at most within 20 days [29,30]. Therefore, it can be said that judging the tumor response to EBRT in the latter part of the treatment and prepare for the ISBT boost is a reasonable approach to perform. Delivering high dose to large volume would directly connect to severe life-threatening late consequences, such as ulceration of the mucosa, abscess formation in the retropharyngeal space, and carotid blowout. The gross tumor volume (GTV) at the time of brachytherapy should be the clinical target volume (CTV) for the HDR-ISBT boost, not the initial GTV before treatment. Since it is generally considered to be dangerous to insert needle applicators into the intracranial residual tumor or skull base, we excluded residual tumors having extensive intracranial and/or skull base invasion. However, if supported by a system, which enabled us to insert applicator needles precisely such as the stereotactic neuro-navigation system, the indication of the HDR-ISBT boost for NPC can be further extended in the future. Organs at risk (OARs) in HDR-ISBT boost for NPC are carotid artery and surface of the nasopharynx. The location and the distance from the surface of nasopharynx to the carotid artery should be checked before needle insertion. Information about anti-coagulant or anti-platelet agents’ usage should be taken before the HDR-ISBT boost. The patient’s hemoglobin level should be checked beforehand and preparation for swift blood infusion should be performed.

Residual tumor in the lateral part of the nasopharynx

Most NPC arises from the lateral part of nasopharynx, namely from the Rosenmüller’s fossa. As mentioned before, when the residual tumor is in the Rosenmüller’s fossa, the organ at risk, to which crucial attention must paid is the carotid artery. Other OARs such as the brainstem, eyes, spinal cord, optic nerves, optic chiasm, etc. are typically further away from the applicators; therefore, brachytherapy dose contribution to these OARs would be minimal because of the inverse-square law. In this technique, CT is used both to determine depth and direction of needle insertion during preplanning, and to guide the actual needle implantation during operation. Before the interstitial needle insertion, the location of carotid artery and the distance from the surface of lateral wall of the nasopharynx and the carotid artery must be checked and measured by image modalities with contrast enhancement agents to avoid the injury of carotid artery. To facilitate the optimal needle insertion from the ipsilateral nasal cavity to the laterally located tumors, needle obturators are angled (Figure 1). Our group uses Pro-Guide® sharp plastic needles (Nucletron, an Elekta company, Elekta AB, Stockholm, Sweden) for HDR-ISBT, and Cheng et al. in their study used metallic needle (Figure 2C, D). Here, the angle should be kept under 30 degrees, not to obstruct the radioactive source transmission inside of the needle applicator, and a visual mark on the plastic needle within the length to the carotid artery will help to increase the safety of the procedure. Marking on the needle at a certain length from the needle tip can define the depth of the needle, preventing from not going further, which can be observed on a fibro-optic scope. Local anesthesia to both sides of nasal cavity is performed. Under general anesthesia or sedation, with continuous vital sign monitoring, interstitial needle insertion is performed. The fiber-optic endoscope is inserted to the contralateral side of the nasal cavity and the bent needle(s) is (are) inserted through the ipsilateral nasal cavity under fiber-optic endoscope guidance (Figure 2A, B). The location and depth of the needle(s) is (are) determined by the CT. Cheng et al. utilized a real-time image guidance system (Medtronic Navigation Inc., Louisville, USA) to ensure the safety of interstitial needle insertion [27]. Typically, less than 4 catheters are implanted because of limited space in the nasal cavity and nasopharynx, and because of the limited space, the distance between the catheters is less than 1 cm and optimally 5 mm apart. As with interstitial brachytherapy for other anatomical sites, an attempt to insert needles parallelly to achieve a favorable dose distribution should be performed. If needle(s) is (are) to be stayed in place for several days, the gauze packing using biodegradable synthetic polyurethane foams, the NasoPore (Polyganics BV, Groningen, The Netherlands), which is used after endoscopic sinus surgery and is packed until the nasopharyngeal cavity, followed by button fixation attached to the needle and sewed to the nasal wing can be used to improve needle fixation during the treatment period. On the other hand, if needle(s) is (are) removed after each brachytherapy session, generally, no fixation is needed because patients are laying down on the treatment table without movement. Certainly, as a measure of precaution, the above-mentioned fixation procedure can be applied for a single irradiation and removal case.

Residual tumor in the ceiling/posterior part of nasopharynx

When the residual tumor is in the ceiling/posterior part of nasopharynx, needle insertion is available by per-nasal approach with usual straight needles under fiber-optic endoscope guidance. Preparation and the actual procedure are similar to that of previous section, and needle insertion to the midline tumors is much easier than laterally located tumors, because there is no carotid artery or other critical normal structures in the midline.

Residual tumor in the inferior wall of nasopharynx

In rare cases, NPC can arise from the inferior wall of nasopharynx. When the main tumors are situated in the oral cavity side, they are categorized in soft palate cancer, namely, superior wall of oropharynx. But when the main tumors are in the nasopharyngeal side, they are classified into NPC, arising from the inferior wall of nasopharynx [31]. Similar to soft palate cancer, NPC in the inferior wall of nasopharynx can be treated with HDR-ISBT using trans-oral approach. Unless the needle goes to the lateral part of nasopharynx, there is no critical structure to cover the tumor located in the inferior wall of nasopharynx. Under general anesthesia or local anesthesia and sedation, needle insertion is performed. Needle insertion is guided by paranasal endoscope and CT. To assist the airway patency, the nasal airway can be inserted through the nasal cavity (Figure 3). Unlike the Pernot’s technique for soft palate cancer, the stability of needle in this per-oral approach is unreliable. Therefore, irradiation was performed only once, and the needles were removed after single treatment. However, because this technique is less invasive than the Pernot’s technique performed under local anesthesia, if multiple fractions is needed, multiple implantations can be achieved. Not for nasopharyngeal cancer, but safely performed trans-oral multiple fractions needle insertion for a recurrent oropharyngeal cancer patient with similar approach was previously reported by our group, in which 12 Gy in 2 fractions of HDR-ISBT was safely applied after 60 Gy of EBRT [32].

Treatment planning

HDR-ISBT boost for residual NPC is generally performed after delivering 60-70 Gy of EBRT [26,27,28]. Dose calculation should be completed on 3D CT-based image guidance (Figure 4). After needle(s) insertion, CT planning with slice thickness less than 3 mm is obtained. Our group uses Oncentra Brachy v. 4.5.1 (Nucletron, an ELEKTA company, ELEKTA AB, Stockholm, Sweden) for dose calculation and 192Ir remote afterloading system, Microselectron-V2, (Nucletron, an ELEKTA company, ELEKTA AB, Stockholm, Sweden) for dose delivery. Any optimizing method can be used such as geometrical optimization, dose-volume optimization, or inverse planning simulated annealing (IPSA); however, final manual modification checking slice by slice is mandatory. When a cold spot under the prescription dose is found on the CT, additional applicator needle(s) can be inserted to adequately cover the target volume higher than the prescription dose, using previously implanted needle(s) as a landmark. In institutions where MRI is available, MRI fusion is desirable with applicator(s) in place, because soft-tissue resolution of MRI is superior to that of CT. In several guidelines for head and neck brachytherapy, dose per fraction between 2.5 Gy to 6 Gy is recommended [30,33,34,35]. In the literature concerning HDR-ISBT for NPC, dose per fraction varied from 2 to 7 Gy, and total delivered dose also varied from 11 to 20 Gy [26,27,28]. However, the Chinese group deliver HDR-ISBT boost within 2-3 months after completion of EBRT, more than 10 Gy was required to control the tumor. Our group deliver HDR-ISBT boost shortly after completion of EBRT, and single fraction of 4 Gy could successfully control a residual tumor that responded poorly to CRT presumably because of shortened total treatment time (a case report currently under revision). Unlike vaginal mucosa, nasopharyngeal mucosa is more likely to develop mucosal ulceration. As mentioned before, the late consequence of such posterior wall mucosa ulceration of nasopharynx could be linked to abscess formation in retropharyngeal space and, in the worst-case scenario, could cause carotid artery blow-out. Therefore, after a full dose of EBRT, the dose per fraction is recommended to be kept under 4 Gy. Additionally, a high dose volume greater than 200% of the prescribed dose on the surface of mucosa should be securely monitored. If the target volume is large, lowering the dose per fraction should be considered.

Needle removal

At the time of needle removal, crucial attention must be paid to bleeding from the puncture site. To facilitate the compression hemostasis, a small portion of gauze soaked with epinephrine can be prepared. When active bleeding is observed after needle removal, prompt compression hemostasis with the epinephrine-soaked gauze is performed under careful continuous vital sign monitoring.


Standard therapy for locally advanced NPC is CCRT, followed by adjuvant chemotherapy or induction chemotherapy, followed by CCRT [36]. NPC responds well to the chemoradiation therapy and local residual/persistent or local recurrences are uncommon. There are, however, some patients who respond poorly to standard chemoradiation therapy, and it has been reported that about 7.5-14% of locally advanced NPC patients develop persistent or local relapse after definitive chemoradiation therapy [18,19,20]. When local control cannot be achieved, the prognosis of such patients is quite poor, because of limited options to salvage such diseases. Several attempts have been made to salvage local recurrent disease with various modalities; outcomes of salvage surgical resection were reported, but this approach resulted in high complication rates with high positive resection margin (29.0%) and 5-year disease-free survival was 42% [37]. Re-irradiation with EBRT has been explored in the form of conventional fractionation using IMRT [38,39] or stereotactic body ablative radiotherapy using the CyberKnife (Accuray, Inc., Sunnyvale, CA, USA) [40], but such treatments were associated with severe late toxicities due to accumulated radiation. It was reported by Han et al. that along with recurrent T stage, tumor volume larger than 38 cm3 and fractional dose larger than 2.3 Gy were associated with late severe toxicities after re-irradiation with IMRT for locally recurrent NPC patients [38]. In a systematic review of literature involving 1,554 patients with re-irradiation for recurrent head and neck cancer, rate of carotid blow-out was reported to be 2.6%, and 76% of these carotid blow-outs were fatal [41]. Therefore, achieving local control in the initial treatment is of paramount importance in the management of NPC. Intracavitary brachytherapy (ICBT) has been used for local boosting after external beam radiation therapy (EBRT) for NPC [21,22,23,24,25]. Because in ICBT, the radioactive source stays in the nasopharyngeal cavity and not directly placed inside of the tumor tissue, it is usually difficult to deliver adequate dose deeper than 5 mm beneath the surface of the nasopharyngeal cavity without delivering dangerously enormous dose to the mucosal surface. If the thickness of the residual tumor is less than 5 mm, the ICBT boost for NPC is an adequate boost modality. However, not all the residual tumors fall into less than 5 mm thickness. Accordingly, a phase III prospective clinical trial by the International Atomic Energy Agency (IAEA) to determine whether ICBT boost after EBRT improves clinical outcomes in NPC patients has failed to show its efficacy, supposedly because of inadequate irradiation dose coverage of the lateral tumor extension with ICBT [22]. Cumulative knowledge of interstitial brachytherapy (ISBT) in other anatomical sites [42,43,44,45,46,47] proved that if the thickness of the tumor is larger than 5 mm, it is unreachable with conventional ICBT, and if interstitial needle application is possible, a boost with HDR-ISBT would be a better option to escalate local dose, while minimizing unnecessary irradiation to surrounding normal tissues [26,27,28]. A group from the Thomas Jefferson University reported the usefulness of salvage surgery combined with re-irradiation with intraoperative 131Cs permanent brachytherapy implant in recurrent head and neck cancer [48,49]. Although, the number of patients investigated in the study was as small as eleven and because the dose fall-off gradient adjacent to the brachytherapy source was extremely steep, only limited volume of carotid artery (less than 1 mm3) received 19 ±10 Gy, which was acceptable [48], suggesting that brachytherapy is a suitable modality to deliver high dose close to critical structures. Because of a paucity of reports on HDR-ISBT boost for NPC, only limited clinical outcomes were reported so far. However, a group from the Sun Yat-sen University compared clinical results between EBRT alone (n = 101) and combination of EBRT and HDR-ISBT boost (n = 40) for T2b NPC patients, and they reported better 5-year local failure-free survival in favor of combination group (80.2% vs. 97.5%; p = 0.012). Optimal timing of delivering HDR-ISBT is not yet determined. While Chinese groups deliver HDR-ISBT boost within 2-3 months after completion of EBRT, our group tries to perform it shortly after EBRT to reduce the total treatment time, which may facilitate to control the tumor with a smaller additional dose. Obviously, because HDR-ISBT boost to NPC is a relatively new technique, patients should be followed up as long as possible to report late complications related to this treatment. Incidentally, ISBT for NPC itself is not a new technique. In 1984, Vikram et al. reported trans-nasal permanent 125I interstitial implantation for NPC patients [50]. They employed a similar technique, however, while we inserted needles under trans-nasal endoscopic guidance, they inserted 125I seeds trans-nasally under trans-oral endoscopic guidance, presumably because only the rigid scope was available at that time. Because neither ultrasonography nor the conventional X-ray fluoroscopy can visualize the relationship between the needle tip and the carotid artery, and once the carotid artery is injured, devastating consequences would follow. Therefore, it is highly emphasized that needle insertion should be performed under CT guidance. There are so many things needed to be explored with adding HDR-ISBT boost in patients with NPC. Appropriate timing for judging the response of EBRT and HDR-ISBT boost application, appropriate fractionation, correct total dose, applicable target volume definition, and dose constraints for surrounding normal OARs are unknown. In the section of technique, guidelines about the combination of EBRT and HDR-ISBT for oropharyngeal cancer were cited [29,30]. Certainly, there is a difference between the combination of EBRT and HDR-ISBT for oropharyngeal cancer and NPC, which is presented in this article. In oropharyngeal cancer, HDR-ISBT boost is used to complement lower EBRT dose and HDR-ISBT is intended to be applied from the starting point, while in NPC, the decision of adding HDR-ISBT boost is based on the tumor response. In the authors’ opinion, a dose of 70 Gy was just determined historically by surrounding normal structures’ tolerance dose for EBRT, not by the tumor response itself. Therefore, while most NPC respond well to conventional 70 Gy of CCRT, some tumors respond poorly to 70 Gy of EBRT, and such tumors would need more than 70 Gy to be controlled. It was demonstrated that in different anatomic sites such as prostate cancer or uterine cervical cancer, brachytherapy boost can safely deliver a much higher dose than EBRT due to its specific feature of steep dose fall-off. Regarding the response of NPC, if the weekly response by the fiberscope can be carefully monitored, it is possible to distinguish between good and poor responders as well as to recognize intermediate responders. For intermediate responders, it is recommended to wait for the final response 8-12 weeks after the completion of 70 Gy, because there are patients who respond slowly. HDR-ISBT boost should initially be applied to poor responding patients. However, after an accumulation of knowledge, it could be possible that intermediate responders should also receive HDR-ISBT boost to obtain favorable local control. But until substantial knowledge is accumulated, the boost should be spared. Finally, its efficacy should be validated in a future prospective clinical trial with NPC patients who respond poorly to the conventional CCRT with or without HDR-ISBT boost application. Since in South China and Southeast Asian countries, many NPC patients can be found, it is highly expected that further important studies on HDR-ISBT boost for NPC will be performed by researchers (brachytherapists) from those countries.


This study was partially supported by the Japan Agency for Medical Research and Development (AMED, 19ck0106305h0003), the National Cancer Center Research and Development Fund (26-A-18 and 26-A-28), and the Project of Social Development Division in Department of Science and Technology of Jilin Province (20190303151SF).


Dr. Jun Itami receives personal fees from HekaBio and from kay J outside the submitted work. The rest of the authors report no conflict of interest.
Copyright: © 2020 Termedia Sp. z o. o. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License (http://creativecommons.org/licenses/by-nc-sa/4.0/), allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material, provided the original work is properly cited and states its license.
Quick links
© 2020 Termedia Sp. z o.o. All rights reserved.
Developed by Bentus.
PayU - płatności internetowe