The role of interstitial brachytherapy with salvage surgery for the management of recurrent head and neck cancers


  • Presented in part at the Annual Meeting of the American Radium Society, May 6–10, 2006, Maui, Hawaii.



The role of interstitial brachytherapy (IBT) in the salvage surgical management of regional lymphatic metastases for head and neck cancer has not been defined to date. To further explore its therapeutic benefit in the management of head and neck squamous cell carcinoma, the authors reviewed their experience utilizing IBT at the M. D. Anderson Cancer Center.


A retrospective chart review of all patients who were received IBT for cervical recurrences of head and neck cancer was undertaken. Twenty-two patients were identified; all patients underwent neck dissection and intraoperative placement of afterloading catheters as part of the salvage treatment. Soft tissue reconstruction was performed when necessary. The duration and dosage of salvage brachytherapy was reviewed, and the overall disease-free survival rates were determined.


All patients had been treated with external beam radiation therapy prior to recurrence, with an average dose of 65 grays (Gy), and 46% of patients had undergone prior neck dissection. The median time to regional recurrence after definitive treatment was 30 months. Soft tissue coverage of the surgical bed with a muscle flap, most commonly a pectoralis major flap, was performed in 19 of 22 patients. The median IBT dose was 60 Gy over a total duration of 4 days. Postoperative complications were few, and there were no perioperative deaths. Recurrences in the reirradiated necks occurred in 27% of patients.


Regional recurrences are common after multimodality treatment for head and neck cancer and are challenging to manage. Although its role has not been defined clearly, salvage neck dissection with IBT is beneficial in the management of recurrent lymphatic metastases of upper aerodigestive tract carcinomas. Cancer 2007. © 2007 American Cancer Society.

The contemporary management of head and neck cancer entails single-modality treatment for early-stage lesions and multimodality therapy for advanced-stage tumors. The mainstay of this treatment paradigm, in addition to surgical resection and chemotherapy, is external beam radiation therapy (EBRT). Numerous prospective, multiinstitutional, randomized trials conducted over the past 15 years have highlighted the central role of EBRT in treating squamous cell carcinomas of the nasopharynx,1 oropharynx,2 hypopharynx,3 and larynx.4 Despite the successes achieved, regional recurrences in the cervical lymphatics are common and pose difficult treatment challenges. Because these patients already have received EBRT, reirradiation with another course of EBRT incurs significant complications. Traditionally, treatment options under these conditions are limited to radical surgery and chemotherapy.

Often, patients who develop a regional recurrence after previous treatment are at high risk for distant metastasis and have an overall survival of from 7 months to 10 months.5, 6 The poor prognosis in these situations also is caused by extracapsular spread of tumor, carotid artery involvement, and unresectable disease. These tumors often involve multiple vascular and neuronal structures and frequently demonstrate extensive skin and soft tissue invasion pathologically. Patients often die of asphyxiation, hemorrhage, sepsis, or pneumonia. The aggressive biologic behavior of these tumors would make them amenable to surgical resection and adjuvant radiation, if it were not for previous treatment with EBRT. Single-modality surgical salvage has a low likelihood of success without additional treatment.

In an effort to achieve regional control of cervical lymphatic recurrences, we initiated a program of salvage surgery with interstitial brachytherapy (IBT) in the early 1990s. Although IBT has been a therapeutic modality for primary and recurrent tumors of the oral cavity,7 oropharynx,8 and nasopharynx,9 its role in the management of cervical lymphatic disease in the recurrent setting is unknown. The theoretical benefit of IBT, as an adjunct to the systematic total resection of involved lymphatics, is enhanced disease control in the tumor bed through precise delivery of continuous, low-dose-rate irradiation of limited range, which minimizes injury to overlying skin and surrounding neurovascular structures.

The potential benefit of surgical salvage with IBT in the setting of recurrent or persistent regional metastasis has not been well studied. The objectives of this analysis were to assess the outcomes and sequelae of IBT in conjunction with salvage neck surgery at the M. D. Anderson Cancer Center for the management of recurrent or residual neck metastases.


Between 1993 and 2005, 22 patients underwent cervical lymphadenectomy and IBT for the management of isolated regional recurrences of head and neck malignancies. An analysis of this patient population was performed after approval was obtained from the Institutional Review Board of the M. D. Anderson Cancer Center of the University of Texas.

All patients were evaluated preoperatively with a history and physical examination, routine hematology and chemistry surveys, axial imaging of the head and neck, and chest radiography. Pathologic diagnoses were confirmed with tissue biopsies. Neck dissection was classified as either 1) selective neck dissection, 2) modified radical neck dissection, or 3) radical neck dissection.

The indication for surgical salvage with IBT was neck disease that could be treated postoperatively by a 1-plane implant, preferably ≥5 months after previous EBRT. An assessment was made intraoperatively to determine the likelihood that all microscopic residual disease was within 5 mm from the plane of the planned implant. At surgical resection, afterloading catheters were placed at 1-cm intervals along the tumor bed and were sutured in place with absorbable sutures. The objective of catheter placement was to have at least 1.5-cm to 2-cm margins around the clinical tumor bed. If the tumor was resected off of the carotid artery, then the catheters were placed directly across the carotid (Fig. 1A). All implants were single plane. Soft tissue reconstruction with a pedicled myocutaneous or myofascial flap was performed at the discretion of the attending surgeon (Fig. 1B). The objectives of flap placement were to displace the skin and surgical wound away from regions of high-dose radiation and to replace resected skin. The location of the lymph node recurrence was marked with surgical clips to facilitate postoperative dosimetry planning (Fig. 1C).

Figure 1.

(A) Placement of afterloading catheters directly over the carotid artery after radical neck dissection. (B) Right neck recurrence 12 years after multimodality treatment for a Stage III tonsillar squamous cell carcinoma. After radial neck dissection, a pectoralis myofascial flap is raised and tunneled into the soft tissue defect. At the time of surgical resection, afterloading catheters are placed at 1-cm intervals along the tumor bed and sutured in place with absorbable sutures. Surgical clips are placed for delineation of the tumor bed, facilitating treatment planning. (C) Treatment planning computed tomography scan obtained 2 days postoperatively. Catheters are highlighted in green. Isodose lines are demonstrated. The total dose delivered was 60 grays, specified at 5 mm from the plane of the implant.

Within 48 hours after surgery, patients underwent planning radiography, and a brachytherapy treatment plan was generated. Five days postoperatively, the catheters were loaded with 192Ir wire. Dose prescriptions were individualized and were dependent on the clinical scenario. The dose rate varied and was dependent on the activity of the radioactive wires at the time of implantation. The implants were dosed at 5 mm from the plane of the implant. Patients were monitored in the hospital postoperatively and during brachytherapy for any surgical or radiation-related complications. All patients were prescribed a regimen of physical therapy for postoperative rehabilitation.

Patient outcomes were assessed, and the regional recurrence rate, disease-free survival, and overall survival were calculated using the Kaplan-Meier method. Adverse events were defined as either acute (occurring within the first 30 days after surgery) or late (occurring >30 days after surgery) and were graded using the Common Toxicity Criteria for Adverse Events (version 3.0).10 Pathology reports were reviewed for adverse pathologic features, including extracapsular spread, perineural invasion, and soft tissue involvement. Statistical analysis was performed using the Stata 9 software package (StataCorp, College Station, Tex).


Patient Characteristics

We identified 22 patients who developed neck recurrences, then underwent salvage neck dissection, and received IBT. The pathologic diagnoses are shown in Table 1. The patient with acinic cell carcinoma had distant metastases at the time of implantation. The median follow-up was 30 months (range, 2.5–157 months).

Table 1. Demographic Data on Patients in the Study
CharacteristicNo. of patients (%)
  1. SCCA indicates squamous cell carcinoma; CA, carcinoma; UADT, upper aerodigestive tract; XRT, external beam radiation therapy; Gy, grays.

Median age [range], y59.1 [41–79]
 Acinic cell CA1
 Adenoid cystic CA1
Prior neck dissection10/22 (45.5)
Median initial XRT dose [range],Gy65.2 [50–72]
Median time to recurrence [range], mo30 [3–363]

Initial Treatments

The initial treatments for all patients are shown in Table 1. EBRT was used for all patients in the initial treatment of their malignancies up to an average dose of 65 grays (Gy) (range, 50–72 Gy). Fifty percent of the patients received either induction or concomitant chemotherapy as part of a protocol. Ten of 22 patients had undergone previous neck dissection as part of their initial treatment. The median time to recurrence in the neck after initial treatment was 30 months (range, 3–363 months).

Salvage Treatment

All patients underwent salvage neck dissection and placement and loading of the brachytherapy catheters, and all patients completed the treatment (Table 2). Soft tissue reconstruction was performed in 19 of 22 patients, with a pectoralis major myocutaneous flap used most commonly. The tumor bed was treated to a median dose of 60 Gy (range, 20–60 Gy) specified at 5 mm from the implantation plane over a median duration of 4 days. The median dose rate was 62 centigrays (cGy) per hour (range, 36–80 cGy per hour). The median number of catheters used was 8 (range, 5–14 catheters). The median total activity of the implants was 35 milligram radium equivalents (mgRaeq) (range, 12–59 mgRaeq).

Table 2. Distribution of Surgical Procedures, Types of Reconstruction, and Radiation Doses Utilized for Patients Treated With Salvage Neck Dissection and Brachytherapy
CharacteristicNo. of patients (%)
  1. ND indicates neck dissection; MRND, modified radical neck dissection; RND, radical neck dissection; SCM, sternocleidomastoid; IBT, interstitial brachytherapy; Gy, grays.

Type of salvage ND
 MRND12/22 (55.5)
 RND10/22 (45.5)
Type of reconstruction
 Pectoralis16/22 (73)
 Median dose [range], Gy60 [20–60]
 Average dose, Gy51.6
 Average duration [range], d4 [15]

The total radiation dose varied in direct proportion to the time interval between the end of the previous course of EBRT and the implantation. Two patients (Group 1) who had persistent disease within 3 months after radiotherapy were treated to 20 Gy and 23 Gy. Because this was not considered an adequate dose to treat viable disease, only 2 patients received implantations to this dose. An additional 2 patients (Group 2) who received postoperative brachytherapy between 3 months and 5 months after EBRT were treated to 40 Gy and 45 Gy. The other 18 patients (Group 3) were implanted ≥5 months after EBRT, and received a median dose of 60 Gy (range, 50–70 Gy).


Early adverse events occurred in 5 patients (23%) and were limited to local infection, donor site hematoma, and cardiopulmonary sequelae, 3 of which were considered grade 3 or 4 events. There were no perioperative deaths. Late adverse events occurred in 5 patients (23%) and were related to local wound complications, 3 of which were categorized as grade 3 or 4 and required operative intervention. Patients with wound breakdown required secondary regional myocutaneous flap coverage for reconstruction and had no ensuing complications (Table 3). A pathologic fracture of the mandible occurred in a patient whose radiation sources were placed directly under the mandible. The osteoradionecrosis was managed successfully with a partial mandibulectomy. Late carotid hemorrhage did not occur, even though several implants were situated directly on the carotid.

Table 3. The Incidence of Early (<30 Days) and Late (>30 Days) Adverse Events After Salvage Neck Dissection and Interstitial Brachytherapy
 Acute toxicityLate toxicity
GradeNo. of patients%GradeNo. of Patients%
Grade 4 4.5Grade 4 9
 Edema, larynx1  Soft tissue necrosis: Neck2 
Grade 3 9Grade 3 4.5
 Hematoma1  Osteonecrosis1 
 Infection: Skin, cellulitis1    
Grade 2  Grade 2  
 None   None  
Grade 1 9Grade 1 4.5
 Muscle weakness: Facial1  Neuropathy: Motor1 
 Wound complication: Noninfectious1    


A review of the pathologic data revealed that 19 of 22 patients had extracapsular spread of disease. Only 3 patients had lymph node recurrences that were confined exclusively to the lymph nodes, whereas 19 patients had disease in soft tissues (n = 14 patients) or in lymph nodes and soft tissue (n = 5 patients).

Of the patients who underwent salvage neck dissection and received IBT, 6 patients developed recurrences in the reirradiated neck. One of the 4 patients who was treated for persistent/progressive disease <5 months after his initial radiation developed a recurrence, and 5 of 18 patients who were treated >5 months after their EBRT developed a regional recurrence.

The median time to regional recurrence after salvage neck dissection and brachytherapy was 19 months. The 2-year actuarial regional control rate was 67%. The median time to death was 21 months. Overall actuarial 1-year, 2-year, and 5-year survival rates were 82%, 57%, and 46%, respectively (Fig. 2).

Figure 2.

(A) Overall survival and (B) disease-free survival of patients who underwent salvage neck dissection and received interstitial brachytherapy after cervical recurrence. The overall actuarial 1-year, 2-year, and 5-year survival rates were 82% and 57%, and 46%, respectively.


The management of recurrent neck metastasis after EBRT for head and neck malignancies remains a difficult clinical problem. Because of prior treatments, therapeutic options are limited. Reirradiation has the potential sequelae of cerebrovascular accidents, carotid rupture, and skin and spinal cord necrosis. Chemotherapy is of limited benefit when administered as a single modality. Extracapsular spread is an adverse pathologic characteristic that is observed commonly in this population, and surgical resection alone usually is not curative. Our data suggest that salvage neck dissection with IBT is a feasible therapeutic option for the regional treatment of selected cervical recurrences.

Several randomized, multiinstitutional trials have demonstrated successful organ-preservation strategies using chemotherapy and EBRT for laryngeal squamous cell carcinomas.4, 11 Furthermore, several nonsurgical approaches for oropharyngeal and hypopharyngeal carcinoma have been implemented at many institutions. In some series, the rates of regional recurrence from hypopharyngeal and oropharyngeal squamous cell carcinoma were approximately 20% to 35%.2, 3, 12 Among patients with cutaneous squamous cell carcinoma who have undergone treatment for regional lymph node metastases, 5% will develop a cervical recurrence.13, 14 Similarly, for patients who have undergone treatment for laryngeal squamous cell cancer,11, 15 there is a 16% risk of developing recurrent disease in the cervical lymphatics. In a study by Johansen et al., the rate of cervical recurrence after either definitive surgical resection or primary radiotherapy for supraglottic carcinoma was 19%.16 In an extensive retrospective review by those authors on outcomes after radiotherapy for laryngeal and pharyngeal carcinomas, only 19% of patients who developed cervical recurrences achieved control of their disease.17

Although lower rates of cervical recurrences after nonsurgical modalities have been reported, adverse overall outcomes are common. Adelstein et al. reported that, of 225 patients who received concurrent chemotherapy and radiation for advanced squamous cell malignancies and developed regional recurrences, only 3 of 18 patients were salvaged successfully.18 Among patients with oral cavity carcinoma who develop a regional recurrence, salvage surgery alone afforded an average survival of 18 months.19 For patients with disease encasing the carotid artery, surgical options are limited to a radical resection, which harbors the exceedingly high risk of neurologic injury and death.20 These and other studies demonstrate that cervical recurrences portend a poor prognosis.

Few centers use the aggressive treatment paradigm we have implemented in the recurrent setting. The experience with reirradiation protocols generally has been unsuccessful and is associated with unacceptable rates of morbidity and mortality.21–23 Puthawala et al. reviewed their experience with low-dose-rate salvage brachytherapy for recurrent head and neck squamous cell cancer. Many of those patients received perioperative chemotherapy and interstitial hyperthermia concurrent with their interstitial implant, and macroscopic resection of disease was not performed routinely.24 Those authors reported that, of the patients who were treated for neck recurrences, regional control was achieved in 74% of patients with a minimum follow-up of 6 months. However, disease-free survival and overall survival at 2 years were 40%. The significant rate of long-term complications (27%), including carotid hemorrhage and soft tissue necrosis, generally was not observed in our patient cohort and may have been related to our standard approach of surgical resection and soft tissue reconstruction. To avoid significant wound complications among patients with early recurrences after EBRT, we limited the IBT dose. The interval of between 3 months and 5 months after EBRT was not considered to be of sufficient duration to allow for adequate normal tissue repair to receive our standard dose of 60 Gy.

Although we used 192Ir-based therapy, permanent 125I implants have shown some benefit as an adjunct to surgery in the past. Park et al. reported on the use of I125 as an adjunct to surgical salvage in patients with recurrent head and neck carcinomas.25 Although their survival results were comparable to ours, their major complication rate was significantly higher, raising the issue of implantation dose and reconstructive techniques. Pellizzon et al. used 192Ir IBT for patients with isolated cervical recurrences and reported a 52% regional control rate.26 In that series, however, only 70% of patients had undergone prior EBRT, and all patients received EBRT after salvage surgery and IBT. The unique treatment paradigm described by those authors makes comparisons with our cohort difficult.

Although our cohort represented a heterogeneous group of patients, we report a high rate of regional control utilizing salvage neck dissection with IBT compared with to published historic data. Among this group of patients, aggressive combination therapy resulted in impressive disease-free intervals and overall survival in patients who had surgically resectable cervical metastasis, with minimal perioperative and long-term morbidity. Compared with published historic control rates, with overall survival rates from 10% to 25% at 2 years for reirradiation, these data appear to represent a significant improvement in survival.21

Although they may be limited in nature, these data support the use of this aggressive multimodality approach in these patients. Future directions include the implementation of adjuvant chemotherapy potentially to improve regional and distant control. We may be able to identify those patients who are suited best for salvage neck dissection in conjunction with IBT as more patients are managed in this fashion. Surgery with IBT has become our standard approach for managing patients who present with isolated, recurrent, regional disease after previous radiotherapy.