Intensity-modulated radiotherapy of head and neck cancer aiming to reduce dysphagia: Early dose–effect relationships for the swallowing structures.

Feng FY, Kim HM, Lyden TH, et al.   Int J Radiat Oncol Biol Phys. 2007;68:1289–1298.

Purpose: To present initial results of a clinical trial of intensity-modulated radiotherapy (IMRT) aiming to spare the swallowing structures whose dysfunction after chemoradiation is a likely cause of dysphagia and aspiration, without compromising target doses.

Methods and Materials: This was a prospective, longitudinal study of 36 patients with Stage III–IV oropharyngeal (31) or nasopharyngeal (5) cancer. Definitive chemo-IMRT spared salivary glands and swallowing structures: pharyngeal constrictors (PC), glottic and supraglottic larynx (GSL), and esophagus. Lateral but not medial retropharyngeal nodes were considered at risk. Dysphagia endpoints included objective swallowing dysfunction (videofluoroscopy), and both patient-reported and observer-rated scores. Correlations between doses and changes in these endpoints from pre-therapy to 3 months after therapy were assessed.

 

Targets (clinical target volumes [CTVs] and their respective planning target volumes; yellow) in relationship to the retropharyngeal (RP) nodes and the pharyngeal constrictor (red; thick arrows marking its anterior edges). The CTVs encompass the lateral RP nodes (green; short arrows), outlined in the retropharyngeal space medial to the carotid arteries (marked by asterisks). The CTVs do not include the medial RP nodes (blue; long arrows).

 

Results: Significant correlations were observed between videofluoroscopy-based aspirations and the mean doses to the PC and GSL, as well as the partial volumes of these structures receiving 50–65 Gy; the highest correlations were associated with doses to the superior PC. All patients with aspirations received mean PC doses >60 Gy or PC V65 >50%, and GSL V50 >50%. Reduced laryngeal elevation and epiglottic inversion were correlated with mean PC and GSL doses. All 3 patients with strictures had PC V70 >50%. Worsening patient-reported liquid swallowing was correlated with mean PC (p = 0.05) and esophageal doses. Only mean PC doses were correlated with worsening patient-reported solid swallowing and observer-rated swallowing scores.

Conclusions: These dose–volume-effect relationships provide initial IMRT optimization goals and motivate further efforts to reduce swallowing structures doses to reduce dysphagia and aspiration.

 

Intensification of the therapy for head and neck cancer, by altered fractionated radiotherapy (RT) or the addition of concurrent chemotherapy, has resulted in improved tumor control rates. The main late sequela following treatment intensification has been increasing rates and severity of long-term dysphagia. For example, Radiation Therapy Oncology Group (RTOG) study 91-11 randomized patients between RT alone or RT concurrent with cisplatin and demonstrated improved tumor control rates in the chemo-RT arm. However, 1 year after therapy 23% of the patients in the chemo-RT arm could eat only soft/liquid food, compared with 9% in the RT-alone arm. Studies in which the chemo-RT regimens were intensified even further in an effort to improve tumor control rates reported 1-year rates of feeding-tube dependence of 20% in the experimental regimens. Evidence has recently emerged that aspiration pneumonia is associated with dysphagia after chemo-RT, constituting an underreported sequela of therapy

Improvements in target dose conformity may reduce the rate and severity of dysphagia following intensive therapy, if these improvements can sufficiently reduce the doses delivered to the anatomic structures whose malfunction after intensive chemo-RT causes dysphagia and aspiration. We have recently initiated studies to investigate this issue. The first step in these studies was the identification of the most important swallowing-related structures. The pharyngeal constrictors (PC) and the glottic and supraglottic larynx (GSL) were found to change anatomically after intensive chemo-RT, and their malfunction explained the posttherapy abnormalities observed in objective assessments of swallowing Certain intensity-modulated radiotherapy (IMRT) strategies (dysphagia/aspiration-specific IMRT) achieved improved sparing of these swallowing structures, without compromising target irradiation, compared with “standard” IMRT. We have subsequently initiated a prospective trial aiming to assess the clinical benefits gained by these strategies. Initial results of this trial are presented here, focusing on the relationships between the doses delivered to the swallowing structures that we aimed to spare (PC, GSL, and esophagus) and the changes in the objective and subjective measures of swallowing dysfunction and aspiration from before to 3 months after therapy.
 

In this clinical study of IMRT aiming at reducing dysphagia, we have found statistically significant, and potentially clinically important, dose–volume effect relationships for dysphagia and aspiration, which can serve as initial dosimetric goals for IMRT. These relationships support the hypothesis that reducing the doses to the swallowing structures may reduce the prevalence and severity of dysphagia; however, they do not yet prove this hypothesis because they do not establish a cause–effect association. In any case, our findings motivate efforts to further reduce these doses, without compromising target doses. The limiting factor in this regard is the percentage of the volume of each of the swallowing structures that is encompassed by the PTVs, found in our study to correlate highly with the mean doses to the whole structure. The first step in the efforts to improve the sparing of the swallowing (and other) structures in this series has been made by daily on-line imaging and correction of setup deviations, which facilitated reducing PTV margins to 3 mm. Future efforts at our institution include the elimination of PTV margins and the construction of IMRT plans that cover the CTVs and their known distribution of setup uncertainties . Additional potential strategies like proton beam IMRT or structure and target assessments and adaptation during therapy should be evaluated.

We have found significant dose–volume effect relationships regarding aspiration for the PCs as a whole and also for each of their parts: the superior, middle, and inferior constrictors. These relationships were statistically strongest for the superior constrictor. The importance of the superior PC doses may be explained by the details of the swallowing mechanism. Elevation of the larynx and pharynx during the swallow is essential for airway protection and bolus propulsion. This elevation is facilitated by the contraction of longitudinal muscles (glossopharyngeus, stylopharyngeus, salpingopharyngeus, and palatopharyngeus), which blend with the circular fibers of the superior constrictor As the larynx and pharynx are pulled up and forward by these muscles, they are pulled away from the lower posterior pharyngeal wall and facilitate opening of the upper esophageal sphincter at the cricopharyngeus level. These mechanisms of swallowing and protection from aspiration, as well as our VF-based results, suggest that the benefits from efforts to spare the swallowing structures are likely to be maximized if they include the superior constrictors rather than being confined to the esophagus and its upper inlet. Our findings that patient-reported dysphagia was also highly correlated with the doses to the superior PC serve as an independent validation of the importance of sparing this structure. In addition, a recently presented study in which brachytherapy was found to reduce dysphagia, concluded that the doses to the upper and middle constrictors were the most significant predictors of patient-reported dysphagia.

We have also found significant correlations between the dose–volume parameters in the GSL and dysphagia. Several recently presented studies examined various dysphagia endpoints after conventional radiotherapy and found significant correlations with the doses to the supraglottic or glottic larynx. In general, these correlations were similar to those reached by our longitudinal study, in which the endpoints were the differences between the pre- and the postradiation dysphagia measures (rather than the postradiation dysphagia alone). In aggregate, these studies affirm the potential benefits in reducing the doses to both glottic and supraglottic larynx.

The dose–volume effect relationships for the swallowing structures may depend on the intensity of the chemo-RT regimen. In the present study, no strictures were observed in patients receiving mean PC dose <66 Gy. In comparison, we have previously found that after an intensive gemcitabine-RT regimen, the minimal dose associated with strictures was 50 Gy. The differences are likely related to the severity of acute mucositis and its consequential effect on pharyngeal tissue. Chemo-RT regimens that do not differ markedly in the rate and severity of the acute mucositis seem to cause similar types and rates of swallowing abnormalities. We therefore anticipate that the dose–volume effect relationships found in the present study, which used a moderate-intensity chemo-RT regimen, will be reproduced after other commonly used regimens of chemo-RT. The site of the primary tumor also affects dose–response relationships, because different primary tumor sites were found to be associated with different rates of both pre- and posttherapy swallowing abnormalities. The relative homogeneity of the patient population in our study, most of whom had oropharyngeal cancer, may have facilitated identifying the dose–response relationships for the swallowing structures. The 3-months posttherapy swallowing results reported here, as well as the dose–volume effect relationships, may change over longer observation time. Swallowing seems to reach a steady state after approximately 12 months, as edema subsides and long-term fibrosis develops  This issue will be addressed as we continue to collect swallowing endpoints at 12 and 24 months.

Swallowing-related laryngeal and pharyngeal motion during treatment may change dose distributions in these structures compared with those observed in the simulation CT. A detailed study of these effects found that the incidence and duration of swallowing during RT is very low, averaging 0.45% (range, 0–1.5%) of the total irradiation time. Also, the mean doses to the swallowing structures were found in our study to be highly correlated with the percentages of the structure volumes inside the PTVs. In a previous study, we found that these percentages did not change significantly when expansion of the swallowing structures to produce planning organ-at-risk volumes was made, compared with the non-expanded structures (5). These data suggest that expanding the swallowing structures to obtain their respective planning organ-at-risk volumes would not alter substantially the planning, optimization, or results of our study. This issue deserves further investigation.

In conclusion, this study has demonstrated that IMRT aiming at sparing the swallowing structures is feasible. Significant relationships were found between dose–volume parameters for these structures and objective and subjective measures of swallowing dysfunction and dysphagia. These relationships can now serve to define optimization goals, and they motivate efforts to reduce these doses as much as possible. Longer follow-up is clearly necessary. Most importantly, care in the outlining of targets in the vicinity of these structures, avoiding target underdosing, and determining and reporting the locations of locoregional recurrences, are essential to ensure that the rates of local recurrences do not increase compared with the rates observed previously after IMRT.