Functional outcomes following chemoradiotherapy for head and neck cancer

Functional outcomes following chemoradiotherapy for head and neck cancer LEE M. AKST, MD, JAMES CHAN, MD, PAUL ELSON, DAVID ADELSTEIN, MD, Cleveland, Ohio

PHD,

JERROLD SAXTON,

MD,

MARSHALL STROME,

MD,

and

OBJECTIVE: To study functional outcomes following concurrent chemoradiotherapy of head and neck cancer. METHODS: A retrospective chart review assessed diet, feeding tube use, and tracheotomy pretreatment and 3, 6, 12, and 24 months posttreatment. RESULTS: Of 196 patients who underwent chemoradiotherapy between 1990 and 2002, 148 (76%) required feeding tubes immediately posttreatment; 150 (77%) required nutritional support at 3 months. By 12 months, 92% were tube-free and 83% resumed near-normal diet. Of 17 patients (9%) with tracheotomy posttreatment, 71% were decannulated by 6 months. Stage IV disease strongly predicted prolonged feeding tube use and slow recovery of diet; age >60 also predicted slow recovery. Primary site, gender, and radiotherapy schedule did not predict functional recovery after adjusting for stage and age. CONCLUSION: Few patients required tracheotomy before or during chemoradiotherapy. Many patients required feeding tubes and nutritional support for several months following therapy, but most were tube-free and eating near-normal diets within 1 year. EBM rating: C. (Otolaryngol Head Neck Surg 2004;131:950-7.)

limiting potential treatment morbidity. The data supporting the improved therapeutic effect of adding concomitant chemotherapy to radiation therapy of advanced HNSCC is well established, with regard to both locoregional recurrence4,5 and survival.6 The role of chemoradiation in preserving functions such as speech and swallowing, however, is much less clear. Some studies suggest that chemoradiation preserves quality-of-life (QOL) measures better than the combination of surgery and radiation therapy does.7,8 However, other studies suggest that chemoradiation may significantly impair swallowing,9,10 perhaps causing functional outcomes worse than radiation alone.11 Part of the discrepancy may accrue to different time points and types of measurements used in these various studies, and randomized data remains absent. Whatever the cause, the relative scarcity of functional data following chemoradiation therapy limits effective decision-making regarding treatment of advanced HNSCC. Knowledge of functional outcomes following chemoradiation is necessary for effective patient care, particularly if organ-preservation protocols are chosen in part to preserve function. This study aims to improve clinical decision-making by characterizing the functional outcomes after chemoradiation therapy for HNSCC over time.

D espite advances in multimodality therapy, survival

METHODS A retrospective chart review of all patients who received concomitant chemotherapy and radiation therapy for treatment of HNSCC at The Cleveland Clinic Foundation was performed. The study was performed with approval of the Institutional Review Board of the Cleveland Clinic Foundation. Many of the patients involved in this study had received concurrent chemoradiation as part of a clinical trial. All such clinical trials were also approved by the Institutional Review Board and appropriate informed consent was obtained from these patients. All patients had a histologically confirmed diagnosis of HNSCC, and no patient had evidence of distant metastases at the time of therapy. Radiation therapy was given either once or twice daily. Patients receiving once-daily treatments received standard fractions of 1.8 to 2 Gy, to a total dose of between 66 and 72 Gy. Patients receiving hyperfractionated radiotherapy received 1.2 Gy twice daily for 6 weeks, to a total dose of 72 Gy. Concurrent chemother-

rates for advanced head and neck squamous cell carcinoma (HNSCC) remain poor.1,2 Meanwhile, quality-oflife measures have become increasingly important determinants of the success or failure of cancer therapy.3 Recent clinical trials aim to improve results in both of these arenas, by improving treatment efficacy while From the Departments of Otolaryngology and Communicative Disorders (Drs Akst, Chan, and Strome), Biostatistics (Dr Elson), Radiation Oncology (Dr Saxton), and Hematology and Medical Oncology (Dr Adelstein), The Cleveland Clinic Foundation, Cleveland, OH. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, Orlando, FL, September 21-24, 2003. Reprint requests: Lee M. Akst, MD, Department of Otolaryngology and Communicative Disorders, The Cleveland Clinic Foundation, Desk A-71, 9500 Euclid Avenue, Cleveland, OH 44195; e-mail, [email protected] 0194-5998/$30.00 Copyright © 2004 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. doi:10.1016/j.otohns.2004.05.020

950

Otolaryngology– Head and Neck Surgery Volume 131 Number 6

apy was provided as 4-day continuous intravenous infusions of fluorouracil (1000 mg/m2/d) and cisplatin (20 mg/m2/d). All patients were scheduled to receive 2 courses of chemotherapy, given during the first and fourth weeks of radiation therapy. Rarely, toxicity limited a patient to a single course of chemotherapy. Both during and after therapy, patients were encouraged to increase oral intake as much as possible, and feeding tubes (if present) were removed as soon as clinically indicated by a patient’s ability to maintain weight with oral intake alone. Following their initial chemoradiation therapy, patients received swallowing evaluations and interventions when clinically indicated. Routine speech pathology evaluation was not performed, nor were barium esophagrams ordered regularly on all patients. Dilation was performed on a case-by-case basis when dysphagia was thought to be secondary to hypopharyngeal/esophageal stricture. Unfortunately, data concerning the effects of swallowing evaluations and esophageal dilation were not readily available in systematic fashion, and therefore their impact is not measured in this study. In a retrospective fashion, pretreatment data was gathered concerning age, gender, primary site and stage of tumor, and neck involvement. Treatment variables recorded included the dose and fractionation of radiation received to the primary site and to the neck, the number of courses of chemotherapy received, and the need for feeding tube placement or tracheotomy during therapy. Functional data was recorded at 3 months, 6 months, 12 months, and 24 months following conclusion of therapy. The presence/absence of a feeding tube or tracheotomy tube were recorded as univariate variables, while quality of diet was graded on a four-point scale based upon the relationship of oral intake vs feeding tube use: normal diet, limited diet without need for a feeding tube, presence of a feeding tube with limited oral intake, and tube feeds only. Assessment of diet was made by retrospective analysis of physicians’ notes, combining information from Otolaryngology, Hematology/Oncology, and Radiation Oncology notes. Patients were censored from the study at death or if persistent/recurrent disease required surgical salvage at the primary site. Therefore, the functional results reported in this study reflect the outcomes achieved only in those patients who were successfully treated by concomitant chemoradiation therapy. At each of the above time points, the proportion of patients with feeding tubes, tracheotomy tubes, and each type of diet was calculated. Kaplan-Meier estimates provided the time to resolution of feeding tube or tracheotomy tube dependence, as well as the time to resumption of a normal diet. Fisher’s exact test was used to compare subgroups of patients and assess base-

AKST et al 951

Table 1. Pre-treatment characteristics and initial tracheotomy and feeding tube use (n ⫽ 196) Factor Gender Male Female Age (years) Mean ⫾ SD Median Range ⬍60 ⱖ60 Primary site Oral Cavity Base of Tongue Tonsils Other Oroph. Hypopharynx Larynx Unknown T Classification T1 T2 T3 T4 Unknown N Classification N0 N1 N2a N2b N2c N3 AJCC Stage (5th edition) II III IVA IVB

N (%)

148 (76%) 48 (24%) 57.3 ⫾ 10.5 58 14–77 113 (58%) 83 (42%) 12 (6%) 41 (21%) 41 (21%) 15 (8%) 34 (17%) 50 (26%) 3 (1%) 15 (8%) 42 (21%) 65 (33%) 70 (36%) 4 (2%) 41 (21%) 29 (15%) 24 (12%) 29 (15%) 42 (21%) 31 (16%) 6 (3%) 36 (18%) 123 (63%) 31 (16%)

line proportions of patients without feeding tubes and tolerating normal diets. After initial baseline variations were examined, the impact of different variables upon recovery of swallowing function (as measured by time to normal diet and percentage of patients without feeding tubes) was assessed using generalized Wilcoxon tests. The Cox proportional hazards model was used to assess the simultaneous effect of different variables on these outcomes. RESULTS A total of 202 patients were identified who had received concurrent chemoradiation therapy for HNSCC at The Cleveland Clinic Foundation. Of these patients, 196 had sufficient data to allow retrospective analysis. Treatment start dates ranged from November 1989 through June 2002; patients who began therapy after June 2002 were not included in the study given the

Otolaryngology– Head and Neck Surgery December 2004

952 AKST et al

Table 2. Tracheotomy tube and feeding tube use at completion of therapy Tracheotomy No Yes Feeding Tube No Yes Combined Data Feeding tube only Tracheotomy only Both Neither

179 (91%) 17 (9%) 48 (24%) 148 (76%) 138 (70%) 7 (4%) 10 (5%) 41 (21%)

short period of available follow-up. Table 1 summarizes the patient and disease characteristics of the study population. Patients were predominantly male, with an average age in the sixth decade of life. Oropharyngeal tumors were most common, followed by laryngeal and pharyngeal carcinomas. The vast majority of patients treated with chemoradiotherapy had advanced (Stage III or IV) disease. Effect of Treatment Upon Baseline Function Table 2 shows the proportion of patients with feeding tubes, tracheotomy tubes, or both, at the completion of chemoradiotherapy (taken as “baseline” posttreatment function). Of the 17 patients who were tracheotomy tube– dependent at the completion of therapy, 13 (76%) had required tracheotomy before chemotherapy or radiation was started. The remaining 4 patients developed progressive airway obstruction necessitating tracheotomy during their course of therapy. Unsurprisingly, most of the patients requiring tracheotomy at baseline (10 of 17, 59%) had laryngeal tumors; the remaining patients were split evenly between base-oftongue and piriform sinus cancers. As compared to the relatively few patients who finished chemoradiotherapy with a tracheotomy tube in place, more than three-fourths of patients had either a percutaneous endoscopic gastrostomy (PEG) tube or a nasogastric tube (NGT). Almost one-third (31%) of these tubes were placed before therapy was initiated; the remainder were placed during the course of chemoradiotherapy. Combining both tracheotomy and feeding tube data, only 21% of patients finished therapy without a tube of either sort. Outcomes over Time Table 3 reflects changes in the need for tracheotomy tubes and feeding tubes over time. Initial values shown in the left-hand column are the baseline values from Table 2. Also shown is change in diet over time, as categorized into one of four groups: feeding tube only,

feeding tube with limited oral intake, limited oral intake without a feeding tube, and normal oral intake. No baseline data was available for this variable at the conclusion of therapy, as the diet was likely not normal for any patient immediately following treatment; the type of diet at 3 months is used as baseline instead. As patients recovered from therapy, data was recorded for 6-month, 12-month, and 24-month posttreatment time points. Using the baseline values established above, the cumulative percentage of patients with functional deficits who then recover function can be calculated. These values are expressed as Kaplan-Meier estimates of the cumulative probability of resolution of a functional deficit, ie, patients were able to be decannulated, have their feeding tube removed, or resume a normal diet. In this analysis, patients are censored (and appear as “N/A”) if they did not have follow-up data after a particular time point. For instance, patients who have been followed for only 11 months since completion of their therapy were censored at the 12-month and all later time points. Patients were also censored from the study at death or if persistent/recurrent disease required salvage surgery at the primary site. Tracheotomy tube use. With regard to tracheotomy tube use, it is seen that 35% of patients with initial tracheotomy tube dependence were able to be decannulated by 3 months, and 71% were without tracheotomy tubes by 6 months. Unfortunately, none of the 5 patients who were tracheotomy-dependent at both 3 and 6 months posttreatment had available 12-month follow-up data: one patient was temporarily lost to follow-up, one died of an unrelated colon cancer, one underwent salvage total laryngectomy for recurrent disease, and the remaining two died of distant metastatic disease from their original HNSCC. Feeding tube use and diet. Use of feeding tubes was much more common than use of tracheotomy tubes, as shown in Table 3. At the conclusion of chemoradiation, 76% of patients required use of a feeding tube. Among these patients, Kaplan-Meier analysis reveals steady improvement from one time point to the next. The greatest interval improvement in removal of feeding tubes occurred between 3 and 6 months posttreatment; conversely, little interval improvement was seen between 12 and 24 months. Overall, almost all patients (94%) who initially required use of a feeding tube were able to have it removed within 2 years. As expected, improvements in diet parallel the feeding tube data. Retrospective assessment of diet was not performed at time points immediately following completion of chemoradiotherapy, so 3-month data was taken instead as baseline. Kaplan-Meier analysis reveals that among patients who did not have a normal

Otolaryngology– Head and Neck Surgery Volume 131 Number 6

AKST et al 953

Table 3. Outcomes over time* Outcome Trach No Yes N/A Cumulative % with FT No Yes N/A Cumulative % with Trach or FT No Yes N/A Cumulative % with Diet FT only FT ⫹ limited PO Limited PO, no FT Normal N/A Cumulative % with

Initial

3 Months

6 Months

12 Months

24 Months

179 (91%) 17 (9%) -0-

185 (94%) 11 (6%) -035%

179 (97%) 6 (3%) 11 71%

154 (100%) -042 –

115 (99%) 1 (1%) 80 –

48 (24%) 148 (76%) -0-

67 (34%) 129 (66%) -014%

129 (69%) 57 (31%) 10 61%

141 (92%) 13 (8%) 42 92%

109 (94%) 7 (6%) 80 94%

41 (21%) 155 (79%) -0Neither†

63 (32%) 133 (68%) -015%

125 (68%) 60 (32%) 11 61%

141 (92%) 13 (8%) 42 92%

108 (93%) 8 (7%) 80 94%

normal diet†

24 (12%) 105 (54%) 21 (11%) 46 (23%) -0–

10 (5%) 47 (25%) 26 (14%) 103 (55%) 10 46%

7 (5%) 6 (4%) 14 (9%) 127 (82%) 42 83%

3 (3%) 4 (3%) 3 (3%) 106 (91%) 80 92%

no trach†

no FT†

Trach, tracheotomy tube; FT, feeding tube; N/A, not applicable. *% are based on the number of patients with data. †Cumulative % are Kaplan-Meier estimates of the time to the first resolution of each outcome based on patients with trach/FT, or less than normal diet at start (at 3 months for diet).

diet at 3 months, recovery of normal diet occurred steadily. By 24 months, only 7 of 116 evaluable patients (6%) continued to require feeding tubes for nutritional support. Of patients with continued feeding tube use, some reasons for failure to resume normal diet include aspiration and esophageal stricture. Unfortunately, even as many patients did receive therapeutic esophageal dilations throughout the study period, complete data regarding esophageal stricture and dilation was not available. Therefore, the impact of esophageal dilation on diet and feeding tube use was not assessed. Factors Predictive of Outcome In univariate analysis, several pretreatment variables had impact on both initial outcomes and recovery of function for both use of feeding tubes (Table 4) and normalcy of diet (Table 5). In each of these tables, the P-value documented beneath the percentage of patients without feeding tubes and tolerating normal diets at baseline assesses the impact of a variable at completion of therapy, while the Wilcoxon P-value shown to the right of the variables assesses the impact of a variable on recovery of function. The estimated probability of recovery by 6 months is provided as a convenient summary measure; however, the statistical test compares the entire distribution. For example, Table 4 shows that while 27% of men and 17% of women did

not require feeding tubes at the end of therapy, this difference was not significant (Fisher’s exact test, P ⫽ 0.18). However, over time men tended to recover more quickly than women (generalized Wilcoxon test, P ⫽ 0.05). These gender differences are also true for normalcy of diet, with men enjoying better initial function and better recovery of function than women. Dichotomous analysis of age, split into groups ⱖ60 and ⬍60 years old, is mixed. Younger patients had a much lower incidence of feeding tube use at the conclusion of therapy than did older patients (percent without tubes: 34% vs 12%, P ⬍ 0.0001). However, at 6 months posttherapy, the proportion of patients without feeding tubes was essentially equal among both age distributions (71% vs 70%). Similarly, patient age did not impact upon normalcy of diet at the conclusion of therapy. However, younger patients did recover oral intake better than did older patients. Primary tumor site did not affect either initial outcome or recovery of function, for either feeding tube use or diet. Tumor stage, however, was highly significant in all of these areas. For purposes of analysis, AJCC Stage II (n ⫽ 6) and III (n ⫽ 36) tumors were considered together and compared to Stage IV tumors (n ⫽ 154). Overall, patients with less advanced disease had improved function, with regard to both feeding tube use and normalcy of diet, at the conclusion of

Otolaryngology– Head and Neck Surgery December 2004

954 AKST et al

Table 4. Time to no FT by patient and disease characteristics Factor Sex Male Female P value† Age ⬍60 ⱖ60 P value† Primary Oral Cav. BOT Tonsil Other Oro. Hypop. Larynx P value‡ Stage II/III IV P value† RT Schedule BID QD P value†

Cumulative % No FT at 6 months

Wilcoxon P value*

N

No FT During Tx

148 48

40 (27%) 8 (17%) 0.18

74% 58%

0.05

113 83

38 (34%) 10 (12%) ⬍0.0001

71% 70%

0.07

12 41 41 15 34 50

2 (17%) 8 (20%) 14 (34%) 2 (13%) 4 (12%) 16 (32%) 0.12

56% 63% 66% 80% 76% 74%

0.48

42 154

16 (38%) 32 (21%) 0.03

90% 65%

0.0002

67 129

11 (16%) 37 (29%) 0.08

63% 74%

0.02

*Generalized Wilcoxon test; all patients included. †Fisher’s exact test comparing baseline proportions. ‡Chi-square test comparing baseline proportions.

Table 5. Time to normal nutrition by patient and disease characteristics Factor Sex Male Female P value† Age ⬍60 ⱖ60 P value† Primary Oral Cav. BOT Tonsil Other Oro. Hypop. Larynx P value‡ Stage II, III IV P value† RT Schedule BID QD P value†

Normal Diet at 3 months

Cumulative % with Normal Diet at 6 months

148 48

40 (27%) 6 (13%) 0.05

62% 48%

0.03

113 83

31 (28%) 15 (18%) 0.13

64% 53%

0.05

12 41 41 15 34 50

2 (17%) 7 (17%) 11 (27%) 1 (7%) 6 (18%) 17 (34%) 0.17

48% 55% 63% 60% 59% 57%

0.71

42 154

16 (38%) 30 (19%) 0.02

83% 52%

0.0002

67 129

11 (16%) 35 (27%) 0.11

53% 62%

0.08

N

*Generalized Wilcoxon test; all patients included. †Fisher’s exact test comparing baseline proportions. ‡Chi-square test comparing baseline proportions.

Wilcoxon P value*

Otolaryngology– Head and Neck Surgery Volume 131 Number 6

therapy. From this improved baseline function, patients with less advanced disease also had statistically significantly better recovery than patients with Stage IV disease (P ⫽ 0.0002 for both feeding tube use and return to normal diet). The combined effects of improved function at baseline as well as improved recovery led 90% of Stage II/III patients to be free of feeding tubes at 6 months compared to 65% of Stage IV patients; similar results are seen with the estimated probability of tolerating a normal diet at 6 months (83% vs 52%). Lastly, in univariate analysis, there was a suggestion that patients who received daily radiation therapy may have enjoyed better function than patients who received hyperfractionated therapy with respect to initial use of feeding tubes (P ⫽ 0.08), initial normalcy of diet (P ⫽ 0.11), and recovery of diet (P ⫽ 0.08). Recovery of function relative to feeding tube use was significant between the groups (P ⫽ 0.02), such that 74% of patients with daily radiation were free of feeding tubes by 6 months, compared to only 63% of patients who received hyperfractionation. As discussed above, stage, gender, age, and radiation schedule all appear to have some impact upon functional outcomes following chemoradiation. In order to identify independent predictors of recovery (both feeding tube use and return to a normal diet), multivariable analyses were performed to simultaneously assess the impact of the different factors on these outcomes. In these analyses, stage was the single most important predictor of both time to feeding tube discontinuation (P ⫽ 0.002) and return to a normal diet (P ⫽ 0.001). Adjusting for stage, there were no other statistically significant predictors of time to feeding tube discontinuation. Age, however, was found to be an independent predictor of time to reestablishing a normal diet even after adjusting for the effects of stage (P ⫽ 0.04). Neither gender nor radiation schedule remained significant for either feeding tube use or diet in a multivariate model. The effects of stage and age can be combined and patients stratified into prognostic groups by counting the number of “poor prognostic” factors present (ie, stage IV or age ⱖ60). Of patients with no poor prognostic factors (ie, patients ⬍60 years old with stage II/III disease, n ⫽ 20), 55% had resumed a normal diet at 3 months and an estimated 90% had returned to a normal diet by 6 months. In contrast, of the 113 patients with one poor prognostic factor present (ie, either stage IV disease or age ⱖ60), 21% had resumed a normal diet by 3 months and an estimated 60% had done so by 6 months. Only 15% of the 60 patients with both poor prognostic factors present had resumed a normal diet at 3 months, and the estimated probability of normal nutrition at 6 months was only 44% (P ⬍ 0.001).

AKST et al 955

DISCUSSION In this cohort of 196 patients who received concomitant chemoradiation therapy for treatment of HNSCC, functional results over time (as measured by the need for tracheotomy and feeding tubes, as well as quality of diet) were generally good. Only 4 tracheotomy tubes were placed for airway obstruction during the course of chemoradiotherapy; the remaining 13 patients with tracheotomy tubes at the completion of therapy had these placed secondary to tumor effect before therapy was begun. Among patients with tracheotomy tubes, most were able to be decannulated by 6 months posttherapy. Unfortunately, for those 5 patients in whom tracheotomy tubes were necessary from conclusion of therapy through the 6-month time point without decannulation, outcomes seem poor—3 of 5 experienced either recurrence of disease at the primary site or distant metastases within the following 6 months. The small numbers of patients in this situation, however, preclude a generalization concerning the prognosis of patients with continued tracheotomy dependence. With regard to feeding tube use and quality of diet, as expected, patients seemed to have significant dysfunction early after treatment completion with steady improvement over time. At one year posttherapy, 92% of patients were without feeding tubes, and 83% tolerated a normal diet. Few patients with feeding tubes at 12 months had these removed by 24 months (increment change in percentage without feeding tubes: 92% to 94%). However, diet continues to improve, even from 12 to 24 months posttherapy. Overall, half of the 17% of patients with limited diets at 12 months will improve to normal diets at the 24-month time point. Pretherapy characteristics such as gender and primary site seem to have little impact upon these functional results, particularly at later time points. Stage, however, was significantly predictive of feeding tube use and quality of diet following chemoradiotherapy, with patients with earlier disease performing better than patients with more advanced disease. Additionally, in multivariate analysis, age also had a small impact, with patients younger than 60 years old recovering a normal diet better than older patients. The effect of radiation schedule approached significance in a univariate analysis but had no impact in a multivariate model. This is likely due to the fact that stage and radiation schedule are associated; 39% of advanced-stage (IV) patients were treated with hyperfractionation compared to 17% of patients with earlierstage (II, III) disease, P ⫽ 0.01. A number of other studies have evaluated swallowing function following chemoradiation.9,10,12-18 Unfortunately, direct comparisons between studies are difficult, as treatment protocols, outcome measures, and

956 AKST et al

length of follow-up are often different. In particular, studies that use videoesophagrams to document aspiration may report higher degrees of dysfunction than studies that depend upon quality of diet and use of feeding tubes. This difference occurs as many aspiration events in this population of patients may be clinically silent.16 That this study utilizes quality of diet and use of feeding tubes as measures of dysphagia is a limitation imposed by its retrospective nature. However, if aspiration is indeed often clinically silent, quality of diet and use of feeding tubes may prove to be more important indicators of quality of life than is aspiration. Indeed, Graner et al used both videoesophagrams and quality-of-life indices and found that the two measures did not correlate.18 Meanwhile, similar studies in the past have used normalcy of diet and use of feeding tubes as endpoints,15,17,19 and other studies have linked swallowing with quality of life.3,14 An unfortunate consequence of this retrospective study is that data concerning the incidence and treatment of pharyngoesophageal stricture was only intermittently available, and thus was not included. At present, it is unknown how many of the patients in this study received pharyngeal or esophageal dilation in order to obtain the level of function that they enjoyed. Attempts to collect this data are underway. Meanwhile, one benefit of the retrospective nature of this study is its ability to accrue large numbers of patients. The 196 patients included in this study comprise a group several times larger than the cohorts described in other studies.9,10,12-18 Additionally, the large database of chemoradiation patients at The Cleveland Clinic Foundation allows for longer follow-up than that seen in most comparable studies. Several small studies have examined the mechanisms of the significant dysphagia that may result from chemoradiation. Koch et al,13 for instance, used videoesophagrams to study 22 patients who received radiation with concurrent carboplatin and cisplatin and found patients experienced dysphagia secondary to worsening oropharyngeal motor function and decreased pharyngeal sensation. Lazarus et al9 compare 9 HNSCC patients treated with radiotherapy and adjuvant chemotherapy with 9 age-matched control subjects. They find swallowing abnormalities in both the oral and pharyngeal stages of swallowing among the treated patients, including reduced posterior movement of the tongue base and reduced laryngeal elevation as documented on videofluorography within 6 months of treatment. However, it should be noted that the patients chosen for this study all experienced swallowing difficulties, and all were hospitalized with weight loss, dehydration, or aspiration pneumonia as a criterion for study inclusion. Lastly, Robbins points out that even if

Otolaryngology– Head and Neck Surgery December 2004

neuromuscular swallowing function remains intact, xerostomia alone may limit oral intake of solids.20 Some studies that explore late outcomes show recovery of good swallowing function with time, while others do not. Smith et al10 examined 10 patients more than 1 year after therapy, with findings of pharyngeal dysmotility in all 10 patients, late aspiration in 3 patients, and hypopharyngeal stricture in 3 patients. In contrast to the findings of the present study, Smith et al found that 6/10 (60%) of their patients required gastrostomy tube supplementation. List et al21 agree with this estimate of persistent dysfunction and report that, secondary to xerostomia and dysphagia, only 50% of patients are able to resume normal diets after chemoradiation. These results are in contrast with those of the current study, which finds 83% of patients eating normal diets and only 8% with feeding tubes at 1 year. Also in contrast to the present study, which observed progressive recovery of swallowing function at later time points, Eisbruch et al16 compare early (1-3 months posttherapy) to late (6-12 months posttherapy) videoesophagrams and find little difference in aspiration rates (65% early, 62% late). There are studies, however, that generally agree with the findings of the current investigation. Staton et al17 choose an early time point of 6 months and document 36% incidence of a feeding tube and/or a tracheotomy tube; the current study reports a cumulative incidence of either tube at 39% for this same time point. One of the earliest investigations into chemoradiation, by Hirsch et al,12 utilized a regimen of concomitant fluorouracil, cisplatin, and radiation and found that with a median follow-up of 5 years, 86% of patients preserved speech and/or swallowing function. Meanwhile, Newman et al15 report 87% of patients with normal diet and only a 13% incidence of feeding tubes 18 months following chemoradiation. Like the current study, in which only stage of disease and, to a lesser degree, patient age (but not radiation schedule, gender, or primary site) were predictive of swallowing and feeding tube use, few other studies are able to identify factors that influence function. This inability may reflect the relatively small number of patients in many of the studies that examine swallowing after chemoradiotherapy. Surprisingly, primary site does not predict outcomes in this cohort of 196 patients. This finding agrees with that of Smith et al,10 who also find primary site fails to predict severity of dysfunction. Eisbruch et al16 conclude that neither tumor T classification, the presence of a tracheotomy tube, nor the dose of chemotherapy affected posttreatment aspiration. Newman et al15 also find that tumor T classification does not predict outcome.

Otolaryngology– Head and Neck Surgery Volume 131 Number 6

Even as no pretherapy factors aside from stage have been found to consistently predict outcome, there are some interesting interventions during and after therapy that might improve results. Studies of amifostine suggest that it might serve to protect tissue against radiation damage, and therefore limit the toxicity of chemoradiation protocols.22 Use of a nasogastric tube rather than a gastrostomy tube may also improve posttherapy function, perhaps by motivating the patient to more aggressively rehabilitate swallowing or by limiting pharyngoesophageal stenosis.23 Additionally, swallowing therapies might improve overall function, and frequent dilation might lessen the morbidity imposed by hypopharyngeal/esophageal stenosis. Further research on functional outcomes following chemoradiation should incorporate these factors, so that attempts can be made to improve as well as assess function. CONCLUSION For patients in whom concurrent chemoradiotherapy successfully treats head and neck cancer, functional deficits following therapy are largely reversible. In multivariate analysis, only advanced stage and age ⱖ60 years predicted decreased function. In this cohort, few patients who did not require tracheotomy before therapy developed a new need for tracheotomy during or after therapy. Although more than three-quarters of patients had limitations in the early posttreatment time period, function steadily improved as patients recovered. By 1 year following therapy, most patients (⬎90%) required neither a tracheotomy tube nor a feeding tube, and many (⬎80%) were tolerating a normal diet. REFERENCES 1. Dimery IW, Hong WK. Overview of combined modality therapies for head and neck cancer. J Natl Cancer Inst 1993;85:95111. 2. Adelstein DJ, Tan EH, Lavertu P. Treatment of head and neck cancer: The role of chemotherapy. Crit Rev Oncol Hematol 1996;24:97-116. 3. Nguyen NP, Sallah S, Karlsson U, et al. Combined chemotherapy and radiation therapy for head and neck malignanices: Quality of life issues. Cancer 2002;94:1131-41. 4. Wendt TG, Grabenbauer GG, Rodel CM, et al. Simultaneous radiochemotherapy versus radiotherapy alone in advanced head and neck cancer: A randomized multicenter study. J Clin Oncol 1998;16:1318-24. 5. Lavertu P, Adelstein DJ, Saxton JP, et al. Aggressive concurrent chemoradiotherapy for squamous cell head and neck cancer. Arch Otolaryngol Head Neck Surg 1999;125:142-8. 6. Pignon JP, Bourhis J, Domenge C, et al. Chemotherapy added

AKST et al 957

7. 8.

9.

10.

11. 12.

13. 14.

15.

16.

17. 18. 19.

20. 21. 22. 23.

to locoregional treatment for head and neck squamouscell carcinoma: three meta-analyses of updated individual data. MACH-NC Collaborative Group. Meta-Analysis of Chemotherapy on Head and Neck Cancer. Lancet 2000; 355:949-55. Terrell JE, Fisher SG, Wolf GT, et al. Long-term quality of life after treatment of laryngeal cancer. Arch Otolaryngol Head Neck Surg 1998;124:964-71. McDonough EM, Varvares MA, Dunphy FR, et al. Changes in quality of life scores in a population of patients treated for squamous cell carcinoma of the head and neck. Head Neck 1996;18:487-93. Lazarus CL, Logemann JA, Pauloski BR, et al. Swallowing disorders in head and neck cancer patients treated with radiotherapy and adjuvant chemotherapy. Laryngoscope 1996;106: 1157-66. Smith RV, Kotz T, Beitler JJ, et al. Long-term swallowing problems after organ preservation therapy with concomitant radiation therapy and intravenous hydroxyurea: Initial results. Arch Otolaryngol Head Neck Surg 2000;126:384-9. Beaver ME, Matheny KE, Roberts DB, et al. Predictors of weight loss during radiation therapy. Otolaryngol Head Neck Surg 2001; 125:645-8. Hirsch SM, Caldarelli DD, Hutchison JC Jr, et al. Concomitant chemotherapy and split-course radiation for cure and preservation of speech and swallowing in head and neck cancer. Laryngoscope 1991;101:583-6. Koch WM, Lee DJ, Eisele DW, et al. Chemoradiotherapy for organ preservation in oral and pharyngeal carcinoma. Arch Otolaryngol Head Neck Surg 1995;121:974-80. Murry T, Madasu R, Martin R, et al. Acute and chronic changes in swallowing and quality of life following intraarterial chemoradiation for organ preservation in patients with advanced head and neck cancer. Head Neck 1998;20:31-7. Newman LA, Vieira F, Schwiezer V, et al. Eating and weight changes following chemoradiation therapy for advanced head and neck cancer. Arch Otolaryngol Head Neck Surg 1998;124: 589-92. Eisbruch A, Lyden T, Bradford CR, et al. Objective assessment of swallowing dysfunction and aspiration after radiation concurrent with chemotherapy for head-and-neck cancer. Int J Radiat Oncol Biol Phys 2002;53:22-8. Staton J, Robbins KT, Newman L, et al. Factors predictive of poor functional outcome after chemoradiation for advanced laryngeal cancer. Otolaryngol Head Neck Surg 2002;127:43-7. Graner DE, Foote RL, Kasperbauer JL, et al. Swallow function in patients before and after intra-arterial chemoradiation. Laryngoscope 2003;113:573-9. Skoner JM, Andersen PE, Cohen JI, et al. Swallowing function and tracheotomy dependence after combined-modality treatment including free tissue transfer for advanced-stage oropharyngeal cancer. Laryngoscope 2003;113:1294-8. Robbins KT. Barriers to winning the battle with head-and-neck cancer. Int J Radiat Oncol Biol Phys 2002;53:4-5. List MA, Mumby P, Haraf D, et al. Performance and quality of life outcome in patients completing chemoradiotherapy for head and neck cancer. Qual Life Res 1997;6:274-84. Buntzel J, Kuttner K, Frohlich D, et al. Selective cytoprotection with amifostine in concurrent radiochemotherapy for head and neck cancer. Ann Oncol 1998;9:505-9. Mekhail TM, Adelstein DJ, Rybicki LA, et al. Enteral nutrition during treatment of head and neck cancer: Is a percutaneous endoscopic gastrostomy tube preferable to a nasogastric tube? Cancer 2001;91:1785-90.