The evolving landscape of head and neck brachytherapy: A scoping review

Systematic search

A systematic literature search was performed in PubMed, EBSCOhost, Europe PMC, and Google Scholar databases for articles on HN brachytherapy. Search strategies included the following: “SU brachytherapy AND SU (head and neck cancer)” for EBSCOhost, “(brachytherapy[MeSH Major Topic]) AND (head and neck cancer[MeSH Major Topic])” for PubMed, “(KW:“brachytherapy” AND KW:“head and neck cancer”)” for Europe PMC, and “allintitle: brachytherapy AND “head and neck” OR nasopharynx OR nose OR nas OR oropharynx OR oral OR orbit OR lip OR buccal OR tongue OR lingual OR neck OR sinus OR sino OR maxilla” for Google Scholar. A filter was applied for articles published from the year 2017 onwards. The last search was done on June 29, 2023.

Study selection

Clinical trials, prospective and retrospective cohort studies, cross-sectional studies, case series, case reports, and dosimetric and simulation studies were included. Dosimetric studies were those reporting on doses from treatment plans that were actually delivered to patients; simulation studies were those that used image datasets from brachytherapy cases or phantoms, and reported on simulated doses based on hypothetical dose regimens that were not delivered to patients. Letters, editorials, and commentaries were excluded. Bibliographies of guidelines and systematic reviews were scanned for other relevant titles. Only studies in English were included. Each study was screened by one of four reviewers, and checked by a second. Any disagreement was resolved by discussion with other reviewers (Figure 1).

Fig. 1

PRISMA diagram of the articles screened and included in the review

Data extraction and charting

Four reviewers performed data extraction. The final list of screened articles was equally divided between two sets of independent reviewers, with each pair working on a designated subset. In case of a conflict or discrepancy, a third reviewer would arbitrate.

Study details were extracted from articles using a standardized template. An iterative team approach to data charting (categorization, extent of detail) was employed, beginning with five pilot charts for each reviewer. In the last iteration, the following data were tabulated: HN subsite, other topics discussed (e.g., re-irradiation, nursing, physics, training), publication year, treatment period (if applicable), origin of publication, study design, number of patients, dose-rate (i.e., LDR, pulsed-dose-rate [PDR], high-dose-rate [HDR]), implant technique (i.e., intracavitary, interstitial, mold, combined), implant approach (i.e., free-hand, template-guided), implant guidance (i.e., visual-guided, palpation-guided, imaging-guided), treatment setting (i.e., definitive, post-operative, perioperative), method details reported, outcomes reported, and analyses performed. Given the objective, no formal risk of bias appraisal was done [5].

Data analysis

Quantitative analysis was completed with descriptive frequency counts of the tabulated entries for each category of extracted information. Qualitative content analysis was performed using a method described by Hancock [6]. A narrative synthesis on the identified emerging themes was formulated and discussed in the context of topics in the GEC-ESTRO 2017 recommendations.

Temporal and geographical trends

In this period, there was no noticeable upward or downward trend in the number of publications on HN brachytherapy, with an average of 21 publications per year from 2017 to 2022. More than two-thirds of the publications came from China, the United States of America (USA), India, and Japan (Table 1). China consistently produced the most research output per year from 2017 to 2022. Since 2020, an increasing proportion of published studies originated from Japan.

Study types

A total of 112 studies reported clinical outcomes, including four clinical trials, 17 prospective cohorts, 72 retrospective cohorts, four case series, and 15 case reports. Of these, 89 studies reported outcomes of cohorts treated in the 2000s and 2010s. The largest number of patients included in a single report was a retrospective cohort from the USA, with brachytherapy outcomes for base of tongue carcinoma using national registry data of 15,934 patients treated from 2004 to 2012 [7]. A total of 22 articles included simulation (n = 11), dosimetric (n = 7), and physics (n = 4) studies.

Clinical studies

Among clinical research, the most studied sites were the oral cavity (n = 84), oropharynx (n = 37), and salivary glands (n = 20). The subsites studied are detailed in Table 2. Furthermore, 21 studies investigated re-irradiation using brachytherapy, 5 pediatric brachytherapy, and 10 combination with other modalities, such as external beam irradiation, chemotherapy, chemoradiation, or pre-operative trans-arterial chemoembolization.

Where dose-rate was specified, HDR (n = 57) or LDR (n = 50) brachytherapy were predominantly investigated. Four studies reported PDR brachytherapy, and six a combination of two or all the above. Most LDR studies explored permanent seed implants (PSI) using ¹²⁵I (n = 39). Majority of HDR or PDR studies used ¹⁹²Ir (n = 34). Others reported the use of ¹⁹⁸Au (n = 8), ¹³¹Cs (n = 5), ⁶⁰Co (n = 2), ¹⁰³Pd (n = 4), ²⁵²Cf (n = 1), and ²²⁴Ra (n = 2).

The treatment setting as well as implant technique, approach, and guidance described in the studies are summarized in Table 3. Brachytherapy was applied in definitive, post-operative, or perioperative settings in 78, 33, and 17 studies, respectively. Interstitial, mold, and intra-cavitary techniques were described in 105, 13, and 5 studies respectively. Five studies also utilized a combination of these techniques. Free-hand approach was used in 88 studies, and template-based approach in 27. Nearly all studies used image-guided methods (n = 52), mostly PSI studies. Of these, 50 were CT-guided and 2 ultrasound-guided. Other studies described visual (n = 57) or palpation (n = 25) guidance.

Pre-clinical studies

Among pre-clinical studies, 11 discussed 3D-printing and 6 reported applicator design. Two physics papers discussed new approaches to dose calculation and dose optimization algorithms. Two studies investigated the use of novel radioactive sources, such as ²⁵²Cf (n = 1), ⁷⁵Se (n = 1), ¹⁶⁹Yb (n = 1), and ¹⁵³Gd (n = 1). Other pre-clinical studies explored ¹⁹⁸Au (n = 1), ¹³¹Cs (n = 2), and ⁶⁰Co (n = 1).

Low-dose-rate permanent seed brachytherapy

The latest update of the 2017 GEC-ESTRO recommendations discussed general aspects of treatment planning, including target volume definition, treatment planning parameters with current stepping source systems, and fractionation schedules for HDR and PDR brachytherapy [1]. Recent publications suggest a resurging interest in LDR brachytherapy, particularly in the form of PSI, as shown by the growing number of studies on this technique (n = 50). It was applied in the definitive or post-operative treatment of locally advanced or inoperable cancers [8-10], early tongue cancers [11], parotid malignancies [12-17], and minor salivary gland carcinomas of the lip and buccal mucosa [18]. Details on the methods of implantation, dosimetric planning, and treatment delivery are described in these studies. Iodine-125 seeds were mostly used, with prescribed doses ranging from 60 Gy to 160 Gy in the mentioned articles. Various reports also described the utility of LDR seed brachytherapy in the setting of re-irradiation or recurrent tumors. Doses applied in these studies ranged from 90 Gy to 160 Gy using ¹²⁵I seeds [19-24], and from 40 Gy to 70 Gy using ¹³¹Cs seeds [25-27]. To our knowledge, there are no current specific guidelines for this modality. Further analysis of data from methodologies and outcomes of these studies may provide evidence for future recommendations on implantation, dosimetry, and treatment planning.

Subsites

The GEC-ESTRO guidelines included discussions regarding the role of primary brachytherapy in malignancies, such as lip, oral cavity, oropharynx, nasopharyngeal, and superficial cancers [1].

Several retrospective studies in this review reported the application of brachytherapy in salivary gland malignancies, particularly parotid cancers. As mentioned previously, several studies on LDR PSIs were performed on this site, and patients were treated both in definitive and post-operative setting. There are articles reporting LDR as an effective primary treatment in the definitive setting without causing severe complications [12, 17]. These findings indicate that there may be emerging data on utilization of LDR seed brachytherapy for parotid cancers, which warrant further investigations.

Perioperative brachytherapy

In addition to discussion on brachytherapy as a primary modality of treatment, the 2017 GEC-ESTRO recommendations also tackled the role of adjuvant brachytherapy. Most of the discussion focused on the use of post-operative brachytherapy performed 1-2 months after surgery [1].

While post-operative brachytherapy remains the more common adjuvant procedure in recent literature (n = 33), studies on perioperative brachytherapy are also growing. With some publications reporting longer follow-up, these additional data can add to the current evidence on its oncologic outcomes and toxicity. For example, acceptable six-year loco-regional outcomes of its use in early mobile tongue cancer was reported [28]. Also, in Khan et al. study, ten-year data on recurrences and overall survival rates were described on its application in the salvage setting for neck recurrences, showing encouraging results and relatively low toxicity rates [29].

Salvage brachytherapy and re-irradiation

The 2017 recommendations recognized salvage brachytherapy as a treatment option in previously irradiated patients. A continued interest in salvage brachytherapy in the setting of re-irradiation was seen, as more studies in recent years (n = 21) further investigated its role. Procedures employed different techniques using HDR, PDR, or LDR brachytherapy, and were performed in different HN sites. In addition to several case reports and retrospective studies, recent prospective data were delivered on this topic. In a study by Martínez-Fernández et al. on 63 patients, perioperative HDR brachytherapy in addition to surgery resulted in long-term loco-regional control, with a 5-year loco-regional control rate of 55% [30]. However, the authors observed significant rates of toxicities, with 50.8% of patients experiencing at least grade 3 adverse effects. Luginbuhl et al. enrolled 49 patients in a prospective study using intra-operative ¹³¹Cs seed brachytherapy. They demonstrated comparable outcomes in comparison with historical cohorts and acceptable safety profile. Two-year disease-free survival was reported in 49%, and rates of osteo-radionecrosis and percutaneous endoscopic gastrostomy (PEG) tube placement were low [26].

Physics

The discussion on general quality assurance and physical aspects in the 2017 GEC-ESTRO recommendations described practical guidance on the procedure of implant checking, treatment planning, dose calculation, and treatment delivery. Recent studies on the physical aspect of brachytherapy included some dosimetric or simulation research. One study compared the dosimetric results, total dwell time, and number of active positions between plans generated by inverse planning simulated annealing (IPSA) and hybrid inverse planning and optimization (HIPO). They observed comparable dosimetric results between the two algorithms, and a benefit of shorter dwell time using HIPO [31]. Another study determined the performance of advanced collapsed-cone engine (a model-based dose calculation algorithm) in treatment planning for scalp brachytherapy. This was compared with the Task Group 43 (TG-43) dose calculations, in which similar doses were found above the skull layer of the phantom, while an underestimation of the dose through the bone was observed [32].

Applicator design and 3D printing

Applicator design and 3D printing were also among the major themes in simulation studies found in the current review [21, 33-36]. These studies described in detail the process of applicator design and fabrication. No specific recommendations have been provided on this topic yet, but emerging interest in this practice may deliver insights on its value.

For general HN cancer sites, brachytherapy using collagen matrix tiles with ¹³¹Cs was investigated in a cadaveric study, and reported feasibility, ease of use, and less carotid dose [33]. For nasopharyngeal cancers, a novel applicator design for intra-cavitary brachytherapy was proposed. A simulation study compared the dosimetric outcomes with the Rotterdam nasopharyngeal applicator, and demonstrated significantly lower soft palate doses with the new design [34].

Various studies also investigated the application of 3D printing in the fabrication of templates and applicator guides. A 3D-printed patient-specific applicator guide for oral tongue cancers was used in a phantom study. Insertion time, geometric accuracy, and dose-volumetric analysis were reported, showing improvement in the treatment process, catheter positioning, and dose homogeneity [35]. There were also retrospective studies on 3D-printed templates in patients who underwent ¹²⁵I seed brachytherapy for recurrent tumors [21, 36]. The utilization of 3D-printed templates resulted in improvements in dosimetry and positioning, with no obvious adverse reactions. Similarly, 3D-printed templates were used for seed implant brachytherapy in cervical node metastases, and resulted in accurate positioning without complications [37].

Another study on personalized brachytherapy involved the use of a 3D-printed anthropometric phantom and lead shielding for the eyes in facial surface brachytherapy procedures. The study aimed to verify the doses to critical organs by measuring the calculated and measured doses, and reported using lead shield as a method for protection of organs at risk [38].