Treatment of locally recurrent squamous cell carcinoma of the head and neck

INTRODUCTION — Recurrent squamous cell carcinoma of the head and neck is a major cause of morbidity and portends poor survival outcomes. Locoregional recurrence, which is seen in 15 to 50 percent of patients with squamous cell carcinoma of the head and neck, is a major factor contributing to mortality from head and neck cancer [1-3].

Recurrent head and neck cancer is difficult to treat for multiple reasons, including the effects of prior treatment on tumor cells and normal tissues, as well as the infiltrative and multifocal nature that typically characterizes recurrent disease in this area [4].

The approach to treatment for patients with locally recurrent squamous cell carcinoma of the head and neck is reviewed here. Recurrent nasopharyngeal cancer and recurrent salivary gland tumors are discussed separately:

●(See "Treatment of recurrent and metastatic nasopharyngeal carcinoma".)

●(See "Malignant salivary gland tumors: Treatment of recurrent and metastatic disease".)

GENERAL PRINCIPLES — Survival is poor for most patients with recurrent head and neck cancer, and therapeutic options are limited by previous treatment. This is especially true for those patients who have recurrent disease or second primary cancer in a previously irradiated field, due to the substantial risks associated with reirradiation. Thus, an important part of the evaluation of previously irradiated patients with recurrent head and neck cancer is distinguishing those who are candidates for aggressive, potentially curative therapies from those for whom palliative measures and supportive care are more appropriate. Whenever possible, patients should be referred to a center of excellence in the management of head and neck cancer to discuss treatment and clinical trials. (See "Reirradiation for locally recurrent head and neck cancer" and "Treatment of metastatic and recurrent head and neck cancer".)

Patients who have had prior radiation therapy comprise a heterogenous group and are substantially different from those who develop recurrences or second primary cancers after surgery alone. These previously irradiated patients should be evaluated for treatment based on the site and stage of disease. Isolated small-volume disease can be considered for surgery alone, while patients with bulkier and/or more extensive recurrences are appropriate candidates for concomitant chemotherapy and radiation therapy as definitive treatment or in the postoperative setting.

Patient selection — All patients with locally or regionally recurrent head and neck cancer should be evaluated for distant metastases. We obtain comprehensive imaging with either positron emission tomography (PET)-computed tomography (CT) from skull base to thigh or contrast-enhanced CT of the chest, abdomen, and pelvis. We also perform molecular testing on the biopsy specimen or on previously obtained tissue in order to evaluate for possible molecularly targeted therapies. Systemic therapy and/or palliative radiation therapy are options for those with distant metastatic disease. (See "Treatment of metastatic and recurrent head and neck cancer".)

Those whose disease is confined to the head and neck may benefit from definitive therapy (surgical salvage and/or reirradiation with chemotherapy). A number of factors may influence the likelihood of benefit from aggressive salvage therapy and should be considered in planning treatment:

●The stage of disease at recurrence may identify those patients who are most likely to benefit from aggressive treatment [5]. A combined analysis that included 32 reports with over 1000 patients, and a prospective series of 109 cases who underwent salvage surgery stratified outcomes and complication risks by stage of disease at recurrence [6]:

•For patients with stage I and II recurrent head and neck cancer, the two-year recurrence-free survival with salvage surgery was 70 percent, 60 to 85 percent achieved or exceeded presurgical quality of life (QOL), and the surgical complication rate was 6 percent.

•For patients with stage III recurrent cancer, the two-year recurrence-free survival dropped to 30 percent. Significant complications occurred in 30 percent, while only 40 percent of patients achieved or exceeded their baseline QOL.

•For patients with stage IV recurrent cancer, outcomes were poor. Fewer than one-fourth of patients lived recurrence free for two years; one-half of patients had a recurrence before five months and died within nine months. Complication rates and QOL were similar to those for stage III patients.

●The initial disease-free interval has an important role in determining the treatment approach. Patients with a disease recurrence within one year of initial treatment have a significantly worse prognosis compared with those who recur later [7].

●The site of recurrence has a considerable impact on outcome. Patients with a laryngeal recurrence, for example, have the most favorable outcome, and those with a hypopharyngeal or oral cavity recurrences have the poorest outcome [4,5].

●Patient-specific factors associated with a poor prognosis include poor cardiopulmonary reserve, poor cognitive functioning, other comorbidities, lack of social support, poor reported QOL, and continued tobacco/alcohol use [4].

●Recurrent disease should be distinguished from a second primary since patients with a second primary malignancy may have a better prognosis than those with a true recurrence [7,8]. However, this distinction is not always feasible. (See "Second primary malignancies in patients with head and neck cancers".)

●Consensus guidelines from the American Radium Society recommend against resection or reirradiation for recurrences after prior radiotherapy in those patients unlikely to achieve long-term (>2 years) survival [9].

Treatment options — Salvage surgery provides the best opportunity for long-term survival if the volume and location of the recurrent disease are amenable to resection and the patient is medically operable [6,10,11]. However, the morbidity associated with salvage surgery can be substantial.

A 2016 systematic review and meta-analysis of oropharyngeal cancer patients from 22 retrospective studies (with a total of 1105 patients) reported superior five-year overall survival (26 versus 16 percent, p<0.001) after surgical salvage compared with reirradiation; patient selection likely accounts for this difference [12].

For patients who are candidates for aggressive therapy but have cancer that is not amenable to surgical salvage, irradiation with or without concurrent chemotherapy is recommended and may offer long-term survival or palliation [13]. In some cases, treatment with immune checkpoint inhibitors may be an option. (See "General principles of radiation therapy for head and neck cancer" and "Locally advanced squamous cell carcinoma of the head and neck: Approaches combining chemotherapy and radiation therapy" and "Reirradiation for locally recurrent head and neck cancer".)

SALVAGE SURGERY — Surgical resection should be considered as a potential first-line treatment for recurrent head and neck cancer. For example, surgical treatment may be the most viable treatment option for many patients with oropharyngeal tumors that recur after definitive chemoradiotherapy. Surgical options include open techniques and more conservative approaches.

Much of the data on quality of life (QOL) and complications come from patients treated initially with primary surgery or radiation therapy alone. As more patients are treated with definitive chemoradiotherapy for functional organ preservation, the role of surgical salvage is evolving. However, prior treatment with chemoradiotherapy may complicate the management of patients who require salvage surgery [14-17].

Open surgical salvage — Reported five-year overall survival rates after salvage surgery generally range from 11 to 39 percent, although higher rates have been reported for highly selected patient populations [6,14,18,19].

Complications — Complication rates in patients undergoing open salvage surgery are higher than in previously untreated patients, secondary to the effects of prior radiation therapy and/or surgery. Furthermore, the increased use of chemoradiotherapy as a combined modality approach may further impair healing and lead to increased complication rates compared with patients managed with surgery or radiation therapy alone. Despite this, salvage surgery may result in an improvement in QOL for patients with recurrent disease. (See "Health-related quality of life in head and neck cancer".)

A meta-analysis that included data from 1633 patients reported rates of operative mortality, total complications, and major complications of 5 (range 0 to 18), 39 (range 10 to 88), and 27 (range 5 to 48) percent, respectively [6]. Major complications included large salivary fistula, partial or total flap necrosis, pneumonia, carotid rupture, and cerebrovascular accident [14]. Minor complications included wound infections, small salivary fistula, wound dehiscence, and minor flap necrosis.

Despite the potential complications, many patients report an improved QOL following salvage surgery. An observational study of 107 patients measured QOL before and after surgery using both the Functional Living Index for Cancer (FLIC), which is a general measure of QOL, and the Performance Status Scale for Head and Neck Cancer Patients (PSS-HN), which includes subscales that measure normalcy of diet, understandability of speech, and eating in public [6]. After surgery, 51 percent of patients showed improvement in FLIC score, with a 16 percent average increase in FLIC score over baseline achieved over a mean of 4.8 months. The percentage of patients who reached or exceeded their presurgery baseline PSS-HN score for normalcy of diet, understandability of speech, and public eating behavior was 47, 41, and 50 percent, with average improvement of 2 to 30 percent over baseline. (See "Health-related quality of life in head and neck cancer".)

Advances in reconstructive surgery — Advances in reconstructive surgery have allowed more patients to be candidates for salvage surgery and have decreased the incidence of some complications [15,20-23].

●Free flap reconstruction in previously irradiated patients decreases local wound complications by bringing nonirradiated, well-vascularized tissue into the wound; however, this technique prolongs surgeries and can create significant morbidity at the donor site [15,20]. As an example, prophylactic tissue flaps used in an "onlay technique" significantly reduce the incidence of pharyngocutaneous fistula in the setting of salvage total laryngectomy (28 to 10 percent) [24].

●Primary tracheoesophageal puncture has been shown to have a low rate of complications. Although pharyngocutaneous fistula and wound dehiscence are observed in this group, this does not seem to be directly related to the tracheoesophageal puncture site [25]. (See "Alaryngeal speech rehabilitation", section on 'Tracheoesophageal voice restoration'.)

Conservative approaches — More conservative surgical procedures may offer an alternative for carefully selected patients with localized recurrences.

Partial laryngectomy — Carefully selected patients with local recurrence of laryngeal and hypopharyngeal cancers may be candidates for conservative resection with partial laryngectomy rather than total laryngectomy. Although there are no randomized trials comparing partial laryngectomy with complete laryngectomy, evidence from observational series suggests that a conservative approach results in similar survival and at least some degree of serviceable speech [26-29].

●In one series, partial laryngectomy was used to manage patients with subglottic extension not greater than 5 mm, cartilage not invaded (as determined by CT scan), extension to contralateral cord not greater than 3 mm, arytenoids (except the vocal process) free of tumor, mobility of vocal cord preserved, and supraglottic extension no further than the lateral sinus of Morgagni [26]. For these patients, partial laryngectomy offered greater five-year survival compared with total laryngectomy (40 versus 26 percent) and at least some degree of serviceable speech (88 percent).

●A contemporary series demonstrates the utility of partial laryngectomy in patients with an original diagnosis of early stage glottic cancer. Multivariate analysis showed that the stage of recurrent disease, rather than the extent of surgery, determined prognosis when comparing patients treated with a partial laryngectomy rather than a total laryngectomy [28].

●Similarly, another series compared five-year disease-specific survivals after partial and total laryngectomy in patients with both early and advanced laryngeal cancer. There was no significant difference in survival between groups [29].

Transoral robotic surgery — Transoral robotic surgery (TORS) approaches are useful alternatives to open surgical resection for recurrent oropharyngeal cancers. These procedures are limited to carefully selected patients with adequate visualization of tumor using robotic retractors, palpable tumor margins, and tumors that do not cross the midline of the tongue base. In this group of patients, TORS is feasible and, according to retrospective data, appears to be oncologically sound.

A contemporary, retrospective, multi-institutional case-control study compared 64 patients treated with a TORS approach with an equal number of patients managed with an open salvage resection [30]. The TORS approach was associated with significantly lower rates of tracheostomy and feeding tube requirements, less blood loss, and shorter hospital stays. The two-year recurrence-free survival rate was 74 percent for this carefully selected group of patients managed with TORS.

Transoral laser microsurgery — Transoral laser microsurgery (TLM) provides an alternative to open salvage surgery for appropriately selected patients with recurrent laryngeal and oropharyngeal cancer. Although there may be slightly lower rates of locoregional disease control, TLM is a reasonable treatment option given its lower morbidity and improved functional outcomes.

●As an example, in a series of 114 patients treated with TLM for recurrent laryngeal or pharyngeal cancer, resection with negative margins was achieved in 97 percent of patients, which resulted in five-year locoregional control and overall survival rates of 67 and 58 percent [31]. Although temporary tracheostomy and gastrostomy tubes were required in 20 and 28 percent of cases, respectively, only one patient required a permanent tracheostomy, and two needed feeding tubes after treatment. The treatment-related mortality rate was 2 percent, and the average hospital stay was two days. After treatment, the majority of patients reported normal swallowing and normal communication with minor dysphonia.

●A 2014 meta-analysis that included data from 286 patients assessed the oncologic and functional outcomes of TLM in recurrent laryngeal cancer following initial radiation therapy [32]. The local control rate was 57 percent at 24 months after the first TLM procedure. The local control rate at 24 months, including the results from a repeat TLM, was 64 percent. The laryngeal preservation rate was 72 percent, and overall survival was 75 percent. This study demonstrates a trend of inferior local control when compared with open partial laryngectomy.

Endoscopic nasopharyngectomy — Endoscopic nasopharyngectomy is a minimally invasive approach for recurrent nasopharyngeal cancers. Endoscopic nasopharyngectomy is associated with less morbidity than open surgical approaches [33] and improves overall survival over intensity-modulated radiation therapy with or without concurrent cisplatin-based chemotherapy [34]. Further details on the use of endoscopic nasopharyngectomy in recurrent nasopharyngeal cancer are discussed separately. (See "Treatment of recurrent and metastatic nasopharyngeal carcinoma", section on 'Salvage surgery (cT1 to T2 nasopharyngeal tumors)'.)

Prophylactic neck dissection — The role of prophylactic neck dissection for patients undergoing surgical salvage is unclear. Some retrospective series of patients who underwent surgical salvage and simultaneous neck dissection report that the risk of occult nodal metastases is low [35-37]. However, others have demonstrated higher rates of occult nodal metastases in some patient groups.

As an example, in patients with laryngeal cancer, occult metastases were detected in 20 percent of patients with a supraglottic recurrence, as well as in those with a T3 or higher glottic recurrence (table 1) [38]. A contemporary retrospective study reports an overall 28.3 percent incidence of pathologic nodal disease in patients with recurrent laryngeal cancer. The rate of recurrence was higher in supraglottic and transglottic carcinomas [39]. In addition, patients with neck metastasis at initial treatment and a recurrence within one year have a significantly higher regional control rate after elective salvage neck dissection compared with observation [40].

Bilateral neck dissection is usually performed for patients with recurrent disease in sites associated with a high risk of occult nodal metastases (for example, recurrent supraglottic, glottic, or pharyngeal cancer) and with extensive recurrences from other sites [6,26,38]. Other patients with clinically negative neck nodes undergoing surgical salvage are treated with ipsilateral selective neck dissection, or in the case of paranasal sinus or nasal cavity cancer, they may forego neck dissection completely.

Isolated neck recurrence — Isolated neck recurrences are uncommon, occurring in 2 to 7 percent of patients who complete initial therapy. The incidence of such recurrences in patients who have been treated with radiation therapy or chemoradiotherapy may be decreased by the appropriate use of imaging studies during the first months following initial treatment. The authors' approach is summarized in the algorithm (algorithm 1). (See "Management of the neck following definitive radiotherapy with or without chemoradiotherapy in head and neck squamous cell carcinoma".)

Surgical salvage of an isolated neck recurrence is feasible in 30 to 50 percent of patients with isolated neck recurrences, particularly if the relapse develops in a previously irradiated area. Five-year rates of both disease control and survival are approximately 10 to 30 percent [41-45]. Results are worse for patients with recurrent disease in a previously dissected neck because surgical salvage is technically difficult, associated with an increased risk of complications, and more likely to result in recurrent disease [41,46,47].

Postoperative treatment with radiation therapy is suggested for patients who have not been previously irradiated [44]. Salvage neck dissection followed by interstitial brachytherapy or intraoperative radiation therapy may also be considered, especially for patients previously irradiated [48,49].

Radiation combined with concurrent chemotherapy in a previously untreated neck or reirradiation with or without concurrent chemotherapy is preferable to surgery if multiple lymph nodes are involved or if there is evidence of extracapsular extension [41,44,50]. (See "Reirradiation for locally recurrent head and neck cancer".)

Postoperative radiation therapy — Among patients with no prior head and neck radiation, postoperative radiation with or without chemotherapy is considered the standard of care for those with the commonly accepted adverse risk features of advanced tumor stage (T3/T4), positive or close resection margins, nodal extracapsular extension, two or more positive lymph nodes (N2/N3), perineural invasion, and/or lymphovascular space invasion. (See "Adjuvant radiation therapy or chemoradiation in the management of head and neck cancer".)

For patients who undergo salvage resection with minimally invasive, tissue-sparing approaches (such as TORS or TLM), it remains unknown if there is a greater need for postoperative radiation therapy. These patients are highly selected, with the expectation of attaining negative surgical margins.

For those patients previously irradiated to the head and neck, our approach is to limit the use of postoperative radiation and chemotherapy for surgically resected patients to those with high-risk pathologic features (positive margins, perineural invasion, lymphovascular invasion, extranodal extension).

The addition of postoperative reirradiation and chemotherapy to surgical salvage may improve locoregional control but does not significantly improve overall survival. In a randomized trial, 130 patients who had undergone salvage surgery were randomly assigned to full-dose reirradiation combined with chemotherapy (fluorouracil plus hydroxyurea) [51]. Disease-free survival was significantly improved. However, there was no improvement in overall survival, and there was a significant increase in both acute and late toxicity. (See "Reirradiation for locally recurrent head and neck cancer", section on 'Postoperative Reirradiation'.)

REIRRADIATION — For patients who are candidates for aggressive therapy but have tumors not amenable to surgical salvage, reirradiation with or without concurrent chemotherapy may offer long-term survival or palliation [13].

Reirradiation for locally recurrent head and neck cancer is discussed separately. (See "Reirradiation for locally recurrent head and neck cancer".)

Complications associated with reirradiation, with or without chemotherapy, are significant. (See "Management and prevention of complications during initial treatment of head and neck cancer" and "Management of late complications of head and neck cancer and its treatment".)

PALLIATIVE SYSTEMIC THERAPY — Systemic therapy may be an option for patients with locally recurrent head and neck cancer who are ineligible for (or choose to forego) definitive therapy or those with rapidly progressive, likely incurable disease. The use of systemic therapy for patients with unresectable locoregional disease or as adjuvant therapy after surgical salvage in patients with recurrent disease and aggressive features remains under investigation. Patients interested in this approach should be encouraged to enroll in clinical trials, where available.

Chemotherapy and/or immune checkpoint inhibitors are effective in patients with metastatic squamous cell head and neck cancer, and may be most appropriate for those with rapidly progressive recurrent disease that is thought to be incurable [9]. As an example, one international consensus guideline for locally recurrent nasopharyngeal carcinoma recommends against resection or reirradiation for recurrences that occur less than 6 to 12 months after completing radiation therapy [52]. Systemic therapy may be more appropriate for such patients.

Further details on the role of systemic therapy for recurrent and metastatic head and neck cancer are discussed separately. (See "Treatment of metastatic and recurrent head and neck cancer".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Head and neck cancer".)

SUMMARY AND RECOMMENDATIONS

●General principles – Patients presenting with locally recurrent squamous cell carcinoma of the head and neck present a therapeutic challenge. Treatment options are limited because of prior therapy (especially radiation therapy) and toxicity, and morbidity can be substantial. Nevertheless, selected patients with recurrent disease may achieve long-term survival. Whenever possible, patients should be referred to a center of excellence in the management of head and neck cancer. (See 'General principles' above.)

●Resectable disease – We recommend salvage surgery for patients with recurrent head and neck cancer if there is no evidence of distant metastases, if the volume and location of the recurrent disease are amenable to resection, and if the patient is medically operable (Grade 1B). (See 'Salvage surgery' above and 'Patient selection' above.)

•Management of the neck – We suggest elective bilateral neck dissection only for patients at high risk of occult nodal disease (eg, recurrent supraglottic, glottic, and pharyngeal cancer) (Grade 2C). Other patients with clinically negative neck nodes undergoing surgical salvage are treated with ipsilateral selective neck dissection. Patients with paranasal sinus and nasal cavity cancer with no clinical evidence of lymph node involvement may forego neck dissection all together. (See 'Prophylactic neck dissection' above.)

•Postoperative radiation – We suggest postoperative reirradiation for patients with high-risk pathologic features after surgical resection (Grade 2C). (See 'Postoperative radiation therapy' above and "Reirradiation for locally recurrent head and neck cancer", section on 'Postoperative Reirradiation'.)

●Unresectable disease – We suggest reirradiation with concurrent chemotherapy for selected patients who are candidates for definitive therapy but have unresectable locally recurrent disease (Grade 2C). (See "Reirradiation for locally recurrent head and neck cancer", section on 'Reirradiation with concurrent chemotherapy'.)

●Isolated neck recurrences – We suggest surgical salvage, if feasible, for an isolated neck recurrence (Grade 2C). Radiation therapy with concurrent chemotherapy is preferable for multiple neck nodes and extracapsular extension. (See 'Isolated neck recurrence' above.)

●Indications for systemic therapy – Chemotherapy and/or immune checkpoint inhibitors may be treatment options for patients with recurrent disease who are ineligible for (or choose to forego) definitive therapy or those with rapidly progressive, likely incurable disease. The use of systemic therapy for patients with unresectable locoregional disease or as adjuvant therapy after surgical salvage for patients with recurrent disease and aggressive features remains under investigation. (See "Treatment of metastatic and recurrent head and neck cancer".)