Past treatment efforts for head and neck squamous cell carcinomas have emphasized treatment intensification that increased local-regional control rates with an increased risk of late (swallowing) complications. With the improved survival demonstrated for human papillomavirus-related oropharyngeal carcinomas, strategies offering comparable outcomes but with fewer complications are needed. Radiotherapy dose reduction has been postulated to reduce the risk of late complications and is an active area of investigation. Alternative strategies may include the use of transoral surgery offering selective use of adjuvant therapy. This article summarizes the contributing risk factors of late swallowing complications and the strategies for risk reduction.
- •
Radiotherapy treatment intensification strategies can improve local-regional control of head and neck squamous cell carcinoma but with an increased risk of late swallowing dysfunction.
- •
Radiotherapy-related risk factors include the dose intensity of the radiotherapy, especially with accelerated radiotherapy schedules and schedules delivering a large dose per fraction to large volumes of the pharynx.
- •
Concurrent chemotherapy can also increase the risk of late swallowing dysfunction.
- •
Several deintensification strategies remain the subject of ongoing investigations.
- •
Selected clinical presentations in which transoral surgical approaches can be safely used offer the potential to evaluate the patient’s pathological risk with regard to the dose and volume of radiation that is administered along with the use of concurrent chemotherapy.
| 3D-CRT | 3D-conformal radiation therapy |
| EGFR | Epidermal growth factor receptor |
| EORTC | European Organisation for Research and Treatment of Cancer |
| HNSCC | Head and neck squamous cell carcinoma |
| IMRT | Intensity modulated radiotherapy |
| MDADI | MD Anderson Dysphagia Inventory |
| OPSCC | Oropharyngeal squamous cell carcinomas |
| PEG | Percutaneous endoscopic gastrostomy |
| PRO | Patient-reported outcome |
| PSS | Performance Status Scale |
| RTOG | Radiation therapy oncology group |
| SCM | Sternocleidomastoid |
| SIB | Simultaneous-in field boost |
| TORS | Transoral robotic surgery |
| TLM | Transoral laser microsurgery |
Introduction
Management approaches for oropharyngeal squamous cell carcinoma (OPSCC) historically used transcervical and mandibulotomy surgical techniques that emphasized the need to achieve local-regional disease control. This came at the expense of competing goals such as the preservation of swallow and laryngeal function. The impact of these functional deficits led many clinicians and investigators to question whether non-surgical treatment alternatives could reduce the functional impact and the overall morbidity involved with classic open en bloc resections while maintaining equivalent oncologic results. As such, the past 30 years has focused on the intensification of non-surgical management strategies for both resectable and unresectable clinical stage III/IV head and neck squamous cell carcinoma (HNSCC). While these efforts have improved the oncologic efficacy of radiotherapy, these efforts have largely been predicated on the assumption that preservation of anatomic “structure” would be sufficient for functional integrity. Moreover, the inclusion of heterogeneous stage III/IV cancers has raised questions regarding the generalizability of the survival benefits across various T-stage and N-stage presentations and across the different head and neck subsites.
It is now clear that the high-dose chemotherapy and altered radiotherapy fractionation strategies, which contributed to improvements in survival rates, are also associated with an increased risk of developing late swallowing complications. In recent years, the significance of this finding is underscored by the favorable prognosis that has been consistently observed in OPSCC associated with the human papillomavirus (HPV) (as defined by various techniques ). Such patients are typically younger with fewer competing co-morbidities and, hence, more likely to experience survivorship issues from current treatment approaches. With a diagnosis of head and neck cancer at a younger age and increased survival rates, the development of late swallowing complications becomes significant and is most likely to contribute to poor quality of life. For these reasons, it is of paramount importance to understand the current risk factors that contribute to late swallowing complications and to determine the best strategies for future investigation. These considerations become especially important because efforts are underway to change current treatment paradigms, especially in HPV-associated OPSCC, to reduce the risk of late swallowing complications.
What are the risk factors for late swallowing complications?
In recent years, several large analyses have examined which factors independently contribute to an increased risk of developing late swallowing complications. In general, the endpoints reported for late swallowing complications or dysfunction have been heterogeneous with most of the reports incorporating some measure of percutaneous endoscopic gastrostomy (PEG)-tube dependency. Other measures have included the presence of aspiration (asymptomatic or symptomatic), assessment by speech language pathology, and/or various patient-reported quality of life instruments. At this time, there is little consensus on how late swallowing complications should be defined.
Fig. 1 summarizes risk factors for late swallowing complications and highlights those factors that may be amenable to potential therapeutic modification. Several factors are not amenable to risk reduction but should be recognized as factors contributing to the development of late swallowing complications, such as
- •
Patient age
- •
Pretreatment swallowing dysfunction due to the tumor
- •
Tumor location.
These studies reflect both the use of nonconformal radiotherapy techniques and in recent years, the use of modern conformal treatment approaches such as intensity-modulated radiotherapy (IMRT).
It is clear from these independent analyses that the very strategies that were used to intensify the treatment for HNSCC, such as radiotherapy dose intensification and concurrent chemotherapy, also deleteriously injure the surrounding normal tissues involved with swallowing. The volume of normal tissue that is irradiated also seems to be a consistent and important risk factor that has been observed in several analyses:
- •
Directly, when the length of the irradiated field is considered
- •
Indirectly, when the T-stage is considered
- •
Whether ipsilateral or bilateral necks are irradiated.
Several analyses did not have sufficient patients irradiated to only one side of the neck for the latter to be analyzed. In fact, Langendijk and colleagues observed that both advanced T-stage and the irradiation of bilateral necks were independent factors in multivariate analysis. This finding suggests that advanced T-stage may be increasing the risk of swallowing injury beyond its influence of increasing the irradiated volume. Increased T-stage definitions reflect more than the increasing size of the primary tumor, but deep tumor infiltration of the surrounding normal tissues which can result in destruction and altered function of the normal tissues.
Late swallowing complications are generally believed to result from
- •
Direct and fibroproliferative-mediated injury to the neuromuscular units that contribute to both sensation and motor functions involved with swallowing
- •
Development of chronic inflammation, a consistent observation, and likely to contribute to the underlying mechanism of injury that is an active area of investigation
- •
Injury to the secretory glands that provide lubrication of the food bolus that can contribute to complaints of dysphagia without any objective evidence of dysmobility.
Radiotherapy as a Risk Factor
Insight into the potential mechanism of late swallowing complications comes from several imaging and retrospective analyses of the dosimetry delivered to various swallowing structures including the pharyngeal constrictor muscles.
Concurrent radiation analysis
In a prospective pilot analysis of 12 patients with HNSCC of the pharyngeal axis treated with concurrent chemoradiation, Popovtzer and colleagues performed MRI of the pharyngeal constrictors at baseline and at 3 months after radiotherapy. These investigators demonstrated
- •
Significant increase in the T2-weighted signals and thickness of the pharyngeal constrictor muscles that received mean doses greater than 50 Gy compared with mean doses less than 50 Gy
- •
No significant differences in T1-weighted signals were noted, suggesting that the development of inflammation is dose related
- •
As a control, analysis of changes in the sternocleidomastoid (SCM) muscles demonstrated a modest increase in T2-weighted signals at 3 months that was not significantly different when partitioned at the mean dose of 50 Gy
- •
At 3 months, the thickness of the SCM muscles decreased significantly, whereas the thickness in the pharyngeal constrictors increased even for patients receiving less than 50 Gy.
As T2-weighted changes reflect tissue inflammation, these investigators hypothesized that the findings in the pharyngeal constrictors were a consequence of persistent treatment-related acute mucositis. Two patients were noted to be PEG-tube dependent at 3 months, both with elevated T2-weighted signals and thickness in the pharyngeal constrictors. Consistent with this interpretation, Dornfeld and colleagues demonstrated ongoing inflammation (as measured by fluorodeoxyglucose-PET activity) at 12 months postradiotherapy that correlated with late swallowing dysfunction and impaired quality-of-life measures.
The findings by Popovtzer and colleagues suggest that mean doses greater than 50 Gy may be associated with an increased risk of late swallowing complications mediated in part by persistent inflammation. This dose threshold is also suggested in several retrospective studies that have correlated the radiotherapy dose administered to the pharyngeal constrictor muscles and to the supraglottic-endolaryngeal structures and the subsequent development of late swallowing complications. These retrospective studies have demonstrated that exceeding mean doses of 50 to 60 Gy delivered to these structures is associated with an increased risk of swallowing complications.
3D-conformal radiation therapy and IMRT
Characterizing the dose-effect relationship becomes important if, in fact, a dose threshold does exist. The report by Levendag and colleagues offers the greatest insight into the nature of this dose-effect relationship because a wide spectrum of dose delivered to the swallowing muscles was analyzed due to the incorporation of a brachytherapy implant in the management plan. A total of 81 subjects with oropharyngeal carcinomas were treated with either 3D-conformal radiation therapy (3D-CRT) or IMRT, with 53% of subjects also receiving a planned brachytherapy boost. These investigators analyzed
- •
For RTOG grade 3 toxicities: severe dysphagia requiring enteral support
- •
For ROTG grade 4 toxicities: complete obstruction, ulceration, perforation, or fistula formation of the esophagus
- •
And in 64 subjects who were alive and free of disease recurrence: various patient-reported outcome (PRO) measures. These investigators sought to determine the relationship between these endpoints and the dose delivered to the swallowing organs.
The mean doses to various swallowing muscles and structures were calculated based on a summation of dose delivered from brachytherapy and external beam radiotherapy ( Fig. 2 ).
With a mean follow-up of 18 months (2–34 months) for IMRT-treated subjects and 46 months (2–72 months) for 3D-CRT–treated subjects, Levendag and colleagues reported a 23% rate of late RTOG grade 3 or 4 dysphagia with a significant association between this late toxicity and the mean dose delivered to the superior ( P = .002), middle ( P = .003), and inferior ( P = .006) constrictor muscles. A threshold dose of 55 Gy (mean) to the superior constrictor muscles for developing late RTOG grade 3 or 4 dysphagia was demonstrated. This risk of dysphagia appeared to increase linearly with increasing dose beyond 55 Gy (see Fig. 2 ). Supporting the notion that a threshold dose effect for the pathogenesis of late swallowing complications may exist.
Gokhale and colleagues observed a similar and significant difference in the risk of late swallowing complications (as defined by the use of PEG tubes for more than 6 months) when comparing patients who received 70 Gy versus 60 Gy to the pharyngeal constrictors ( Fig. 3 ).
Levendag and colleagues further reported on the results of PROs obtained posttreatment in 88% of subjects without disease relapse. PROs analyzed included the European Organisation for Research and Treatment of Cancer (EORTC) core Quality of Life Questionnaire (QLQ) Core 30 (C30) and the EORTC QLQ-Head and Neck 35 (HN35) swallowing scale. Functional assessment was also analyzed using the observer-reported List Performance Status Scale (PSS) and the patient-reported M.D. Anderson Dysphagia Inventory (MDADI). Although not specified, it appeared that the PROs were obtained at a minimum of 12 months following completion of radiotherapy. Levendag and colleagues clustered the scores in these PROs to derive surrogate grade 3 or 4 toxicity rates and found similar complication rates to the retrospective chart review assessments for RTOG grade 3 or 4 dysphagia:
- •
HN35: 7–18%
- •
PSS: 2–30%
- •
MDADI: 21–32%.
More significantly, dose correlations were also observed with these patient-reported measures of dysphagia offering additional validation of a dose-effect relationship ( Fig. 4 ).
Nonconformal radiation therapy
Further validation of a threshold dose for developing dysphagia has also been suggested in the setting of a randomized phase III study of postoperative radiotherapy dose intensification. All subjects were irradiated with a nonconformal technique and would have resulted in most of the pharyngeal constrictor muscles being irradiated. With a median follow-up of 59 months (22–83 months), Ang and colleagues observed
- •
A late grade 3 or 4 dysphagia rate of 13%–16% in patients receiving 63 Gy postoperatively (without concurrent chemotherapy)
compared with..
- •
Late grade 3 or 4 dysphagia rate of 4% when 57.6 Gy was delivered.
This study is noteworthy, not only for the randomized prospective nature of the study design, but that patients were treated without concurrent chemotherapy.
Radiotherapy intensification
Radiotherapy intensification using accelerated fractionation schedules with nonconformal radiation techniques has been shown to be a significant risk factor for late swallowing complications in multivariate and univariate analyses. A similar inverse correlation between the overall treatment time and the severity of acute mucositis and the risk of late swallowing complications have been reported. This risk seems to be also related to the daily dose of radiotherapy, total radiotherapy dose delivered at the end of treatment, and the severity of acute mucositis. These observations offer a cautionary lesson on more modern strategies to accelerate the radiotherapy using simultaneous-in field boost (SIB) IMRT strategies.
In fact, similar correlations between SIB-IMRT large daily dose per fractions administered to a large mucosal volume have been observed to cause:
- •
Unacceptable acute mucositis
- •
Increased acute dysphagia
- •
Delayed mucositis healing.
In a phase I dose-escalation study of SIB-IMRT alone, Lauve and colleagues demonstrated acute dose-limiting mucosal toxicities with daily fractions of 2.46 Gy, leading these investigators to conclude that the maximum-tolerated dose was 2.36 Gy. Late toxicities were also reported with grade 3 dysphagia and grade 4 mucosal toxicities seen at 2.36 Gy, suggesting that, in fact, the optimal dose per fraction may be lower.
Bhide and colleagues also performed a phase I dose-escalation SIB-IMRT with concurrent chemotherapy and demonstrated:
- •
A significant increase in the rate of grade 3 dysphagia with daily fractions of 2.4 Gy compared with 2.25 Gy or less
- •
A correlation between the length of the radiation field treated to doses of 50 Gy or greater with the risk of grade 3 dysphagia
- •
A correlation between the duration of grade 3 mucositis (longer than 12 weeks) and the risk of developing late dysphagia at 6 months.
Radiation to ipsilateral neck
Most OPSCC patients typically receive radiation to the primary site and bilateral cervical necks due to the risks of bilateral lymphatic metastases. However, when patients receive radiation to the primary and only the ipsilateral neck, Langendijk and colleagues noted that the risk of late swallowing complications is significantly reduced, even when T-stage was introduced into the multivariate analysis. Similar observations were reported by Frowen and colleagues who demonstrated that irradiation of only the ipsilateral neck significantly reduced the risk of late dysphagia with liquids as evaluated by videofluoroscopy. Consistent with the adverse influence of the high dose irradiated tumor volume was the observation that the length of the nonconformal fields also directly correlated with the risk of swallowing dysfunction at 3 months and a trend at 6 months. Not surprising are similar direct correlations between the length of nonconformal radiation fields and the severity of the acute mucositis and the requirement for PEG support, reaffirming that late swallowing complications can arise as a consequence of severe acute mucositis.
In summary, several lines of evidence support the conclusion that the radiotherapy dose and the volume of the pharynx that is irradiated can influence the risk of developing late swallowing complications—increasing when a large volume of the pharynx is irradiated with mean pharyngeal constrictor doses greater than 50 to 60 Gy, as is the case with definitive bilateral neck radiation for oropharyngeal carcinomas. Whether or not specific swallowing regions of the pharynx (such as the inferior constrictor muscles) are more sensitive to radiation-induced injury is not clear at this time. What is clear is that the severity of the acute mucosal injury is correlated with the risk of persistent mucosal inflammation and a potential consequential risk of late swallowing complications. At this time, the exact nature of the dose-volume relationship for late swallowing complications has not been clearly delineated. When considering the mean dose to the irradiated constrictor muscles, there may indeed be a threshold dose for the risk of developing late swallowing complications. Beyond this threshold dose, the risk may increase significantly.
Chemotherapy as a Risk Factor
Concurrent chemotherapy can significantly and independently increase the risk for late swallowing complications. Most of the evidence to date comes in the setting in which high doses such as 70 Gy are planned. Multivariate analyses in which a sufficient number of subjects were treated with and without concurrent chemotherapy have demonstrated hazard ratios ranging from 2.6 to 9. Whether or not the influence of concurrent chemotherapy is different at lower planned radiotherapy doses to the pharynx is not clear at this time. However, the analysis by Caudell and colleagues demonstrated that when concurrent chemotherapy was modeled with radiotherapy doses less than or greater than 70.4 Gy, the administration of concurrent chemotherapy was not only independently significant but was associated with a ninefold risk in contrast to a nonsignificant influence of the radiotherapy dose partitioned at 70.4 Gy.
The risk of late swallowing complications has not been found to be significantly different between platinum alone and platinum combination chemotherapy regimens, though this observation may be limited by a lack of statistical power in its analysis. It is also not clear if alternative schedules of cisplatin administration, such as weekly low-dose cisplatin, can reduce this risk. In a randomized trial of low-dose weekly cisplatin (20 mg/m 2 ) with radiotherapy compared with radiotherapy alone, the administration of weekly cisplatin increased the risk of late esophageal (9% vs 3%, P = .03) and laryngeal (11% vs 4%, P = .05) toxicities when compared with radiotherapy alone. Interestingly, no evidence of tumor radiosensitization was seen.
An alternative radiosensitizing option has been the weekly administration of the epidermal growth factor receptor–binding monoclonal antibody, cetuximab. The results of a large randomized study of concurrent cetuximab and radiotherapy demonstrated that acute mucositis rates may not be significantly increased to that of fractionated radiotherapy alone. This has been interpreted as evidence of potential selective targeting of HNSCC cells. This observation along with the modest spectrum and severity of systemic toxicities has lead to the perception that cetuximab may also have a favorable late toxicity profile. However, the rates of long-term toxicities, especially measures of late swallowing complications, have not been reported, even with long-term follow-up. Hence, the ability to use cetuximab as a strategy to potentiate the effects of radiotherapy while reducing the risk of late swallowing complications is unclear at this time.
Surgery as a Risk Factor
The practice of postradiotherapy prophylactic neck dissections began in response to unsalvageable neck relapses, especially for advanced nodal disease, even when a complete clinical response in the neck was achieved. Although this practice continued even in the era of radiotherapy intensification, the high neck control rates observed may have potentially been confounded by the unrecognized and evolving epidemiology of HPV-associated OPSCC. Moreover, post-chemoradiotherapy neck dissections have recently been associated with an increased risk of late swallowing complications in a recent large analysis of the RTOG database, complicating the risk-benefit assessment of this practice.
Late RTOG grade 3 or 4 laryngeal or pharyngeal dysfunction
Machtay and colleagues performed a case-control analysis comparing 99 subjects with late RTOG grade 3 or 4 laryngeal or pharyngeal dysfunction matched to 131 controls from three previously completed RTOG head and neck phase trials (RTOG 9111, 9703, and 9914). Multivariate analysis demonstrated several factors associated with late laryngeal/pharyngeal dysfunction, including:
- •
Age (odds ratio [OR] 1.05, P = .001)
- •
Advanced T-stage (OR 3.07, P = .0036)
- •
Larynx or hypopharynx primary site (OR 4.17, P = .0041)
- •
Post-chemoradiation neck dissection (OR 2.39, P = .018).
The strength of this analysis is the large study cohort that also included the use of negative matched controls. These results are consistent with consistent with other retrospective reviews.
In contrast, Caudell and colleagues did not observe postradiotherapy neck dissections to be significant in their multivariate analysis. Although 69% of the 122 study subjects received concurrent chemotherapy, this analysis may have been underpowered as the number of individuals with late swallowing dysfunction (which was defined as a composite clinical endpoint including the presence of PEG-tube dependency or the presence of aspiration or strictures) who underwent a neck dissection was 12 compared with 35 patients who did not receive a neck dissection. Similarly, the recent retrospective review by Chapuy and colleagues analyzed only patients who received a neck dissection without a cohort of patients without a neck dissection while evaluating for factors that correlated with impaired swallowing.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
