Definitive radiotherapy alone or combined with a planned neck dissection for squamous cell carcinoma of the pharyngeal wall

Authors


Abstract

BACKGROUND

In the current study,the authors analyzed the results of definitive radiotherapy for squamous cell carcinoma of the pharyngeal wall.

METHODS

Between 1964 and 2000, 148 patients were treated with definitive radiotherapy. All patients had a 2-year minimum follow-up.

RESULTS

The following 5-year rates of local and ultimate local control were obtained: T1 disease, 93% and 93%; T2 disease, 82% and 87%; T3 disease, 59% and 61%; and T4 disease, 50% and 50%, respectively. Multivariate analysis revealed that twice-daily fractionation (P = 0.0009), American Joint Committee on Cancer Stage I–II disease (P = 0.0051), and oropharyngeal primary site (P = 0.0193) were associated with improved locoregional control. The following 5-year absolute and cause-specific survival rates were obtained: Stage I, 56% and 89%; Stage II, 52% and 88%; Stage III, 24% and 44%; Stage IV, 22% and 34%; and overall, 30% and 49%, respectively. Eight patients (5%) died of complications.

CONCLUSIONS

Locoregional control and survival were found to be related to site, extent of disease, and fractionation schedule. Although outcomes have improved in recent years, the morbidity of treatment was significant in the current study and a substantial proportion of patients died secondary to the malignancy. Cancer 2003. © 2003 American Cancer Society.

Squamous cell carcinoma of the pharyngeal wall is relatively uncommon, accounting for 20–25% of all hypopharyngeal malignancies.1 It may be treated with either surgery or radiotherapy.2–5 Patients usually present with advanced disease. Therefore, if surgery is the primary treatment, the majority of patients will receive adjuvant radiotherapy.6–8 Regardless of the treatment strategy, the likelihood of cure has been low, compared with squamous cell carcinomas arising elsewhere in the head and neck.9–15

Since 1964, the preferred treatment for essentially all previously unirradiated patients with pharyngeal wall carcinomas at the University of Florida has been radiotherapy alone or combined with a neck dissection.16–18 Planned split-course radiotherapy19 and interstitial implantation of the primary site,17 two techniques that have been used briefly in the past and found to be associated with a lower probability of cure, have long since been abandoned. Two modifications in the radiotherapy technique occurred in 1978. Both of them, i.e., an aggressive hyperfractionation schedule combined with placement of the posterior border of the “off-cord reduction” just anterior to the spinal cord to ensure adequate coverage of the tumor,16, 20 have been associated with an increased likelihood of cure.

The aim of the current study is to update the relatively unselected University of Florida experience with definitive radiotherapy in the treatment of patients with squamous cell carcinoma of the pharyngeal wall.

MATERIALS AND METHODS

Between October 1964 and December 2000, 148 patients with previously untreated squamous cell carcinoma of the pharyngeal wall were treated at the University of Florida with continuous-course definitive external-beam radiotherapy alone or combined with a planned neck dissection.18 Twenty-five patients were treated before 1980 and the remainder were treated between 1980 and 2000. Patients had a median follow-up period of 2.0 years (range, 0.1–18.6 years). One patient was lost to follow-up 1.6 years after treatment. All remaining living patients had a minimum follow-up period of 2 years. Characteristics of the patient population are depicted in Table 1. The median age of the patients was 64 years (range, 42–81 years). Patients were staged according to the 2002 American Joint Committee on Cancer21 staging system.

Table 1. Patient Population (n = 148)
ParametersNo. of patients (%)
Gender 
 Male124 (84)
 Female24 (16)
Race 
 White131 (89)
 African-American15 (10)
 Other2 (1)
Site 
 Lateral wall39 (26)
 Posterior wall109 (74)
 Oropharyngeal55 (37)
 Hypopharyngeal93 (63)
T classification 
 T115 (10)
 T245 (30)
 T376 (51)
 T4A6 (4)
 T4B6 (4)
N classification 
 N053 (36)
 N123 (16)
 N2A5 (2)
 N2B28 (19)
 N2C26 (18)
 N313 (9)
Overall stage 
 I9 (6)
 II27 (18)
 III36 (24)
 IVA57 (39)
 IVB19 (13)

Pretreatment computed tomography (CT) scans were available to determine the presence of radiographically positive retropharyngeal lymph nodes in 89 patients. Retropharyngeal lymph node positivity versus T classification revealed the following findings: T1 disease, 0 of 11 patients; T2 disease, 3 of 28 patients (11%); T3 disease, 16 of 43 patients (37%); T4 disease, 4 of 7 patients (57%); and overall, 23 of 89 patients (26%). Retropharyngeal lymph node positivity versus overall stage revealed the following: Stage I, 0 of 5 patients; Stage II, 0 of 16 patients; Stage III, 4 of 18 patients (22%); and Stage IV, 19 of 50 patients (38%).

Fifty-four patients (36%) were treated once daily with approximately 1.8–2.0 grays (Gy) per fraction to the following median doses: T1 disease, 65.5 Gy (range, 54.7–70 Gy); T2 disease, 70 Gy (range, 65–75 Gy); T3 disease, 74 Gy (range, 65–76 Gy); and T4 disease, 70.3 Gy (range, 69.3–75 Gy). Since 1978, 94 patients (64%) were treated twice daily with 1.2 Gy per fraction to the following median doses: T1 disease, 74.4 Gy (range, 73.5–77.9 Gy); T2 disease, 76.8 Gy (range, 73.8–79.2 Gy); T3 disease, 76.8 (range, 52.8–82.8 Gy); and T4 disease, 76.8 Gy (range, 74.4–79.2 Gy). Eleven patients received adjuvant platinum-based induction chemotherapy (5 patients) or concomitant chemotherapy (6 patients). Thirty-six patients received a planned unilateral (31 patients) or bilateral (5 patients) neck dissection 4–6 weeks after the completion of radiotherapy. Patients who underwent a planned postradiotherapy neck dissection generally had N2–N3 disease and/or incomplete lymph node regression after irradiation. Residual tumor was identified in the neck dissection specimen in 19 of 36 patients (53%).

Complications were coded as severe if they necessitated hospitalization, surgical intervention, or resulted in death.22 The presence of a permanent tracheostomy and/or gastrostomy was also coded as a severe complication.

All statistical analyses were performed using SAS software.23 Univariate survival estimates were obtained by the Kaplan–Meier product-limit method.24 The log-rank statistic used to detect differences between strata and corresponding two-sided P values less than 0.05 was interpreted as being statistically significant. Cox regression was implemented for multivariate analyses.25 The following parameters were included in the multivariate analyses: T classification (T1–T2 vs. T3–T4); overall stage (I–II vs. III–IV), gender, oropharyngeal versus hypopharyngeal wall, lateral versus posterior pharyngeal wall, once-daily versus twice-daily fractionation, and location of the posterior border of the reduced field relative to the vertebral bodies and the anterior aspect of the spinal cord (the border between the posterior one-third of the vertebral body and the spinal cord vs. anterior to the posterior one-third of the vertebral body).

RESULTS

Local Control

The 5-year overall local control rate was 69%. The following 5-year rates for local and ultimate local control (including those patients who were successfully salvaged after local disease recurrence) were obtained: T1 disease, 93% and 93%; T2 disease, 82% and 87%; T3 disease, 59% and 61%; and T4 disease, 50% and 50%, respectively (Fig. 1).

Figure 1.

Local control after definitive radiotherapy.

Locoregional Control

The 5-year overall locoregional control rate was 59%. The following 5-year rates of locoregional and ultimate locoregional control were obtained: Stage I, 89% and 89%; Stage II, 83% and 86%; Stage III, 58% and 61%; and Stage IV, 47% and 50%, respectively (Fig. 2). Multivariate analysis revealed that locoregional control rates were significantly better for oropharyngeal carcinomas (P = 0.0193), Stage I–II lesions (P = 0.0051), and for patients treated with twice-daily radiotherapy (P = 0.0009). In contrast, T classification (P = 0.2095), gender (P = 0.9938), location of the posterior border of the reduced fields (P = 0.5920), and lateral versus the posterior pharyngeal wall (P = 0.2453) did not appear to significantly influence this endpoint in the multivariate analysis. The simultaneous implementation of hyperfractionation and placement of the reduced field borders just anterior to the spinal cord may have confounded the influence of the latter parameter in the multivariate analysis.

Figure 2.

Locoregional control after definitive radiotherapy alone or combined with a planned neck dissection.

Distant Metastasis-Free Survival

The following 5-year distant metastasis-free survival rates were obtained: Stage I, 100%; Stage II, 100%; Stage III, 80%; Stage IV, 63%; and overall, 77%.

Second Primary Carcinomas

Fifty-seven patients (38%) experienced 71 additional upper aerodigestive tract malignancies, which were fatal in 25 patients. Eight carcinomas were metachronous and occurred before the pharyngeal wall carcinoma, 32 carcinomas were synchronous, and 31 malignancies were metachronous and occurred after the pharyngeal wall carcinoma. The second malignancies were located in the head and neck (50 patients), lung (16 patients), and esophagus (5 patients).

Survival

The following 5-year cause-specific and absolute survival rates were obtained: Stage I, 89% and 56%; Stage II, 88% and 52%; Stage III, 44% and 24%; Stage IV, 34% and 22%; and overall, 49% and 30%, respectively (Figs. 3, 4).

Figure 3.

Cause-specific survival versus overall American Joint Committee on Cancer stage.

Figure 4.

Absolute survival versus overall American Joint Committee on Cancer stage.

Multivariate analysis revealed that overall Stage I–II disease (P = 0.0003), oropharyngeal location (P = 0.0319), and treatment with twice-daily radiotherapy (P = 0.0004) were significantly associated with improved cause-specific survival. In contrast, T classification (P = 0.2510), gender (P = 0.3153), location of the posterior border of the reduced fields (P = 0.7184), and lateral versus posterior wall primary site (P = 0.8410) did not appear significantly influence this endpoint.

Severe Complications

Treatment complications resulted in the death of eight patients (5%). These included aspiration pneumonia (four patients), soft tissue and/or cartilage necrosis (three patients), and laryngeal edema (one patient). Twenty-three patients (16%) experienced 1 or more nonfatal severe complications, including permanent gastrostomy tube (15 patients), soft tissue and/or bone necrosis (7 patients), and carotid rupture, orocutaneous fistula, tracheostomy, and brachial plexopathy (1 patient each).

DISCUSSION

Squamous cell carcinoma of the pharyngeal wall may be treated with surgery and/or radiotherapy. Treatment results are influenced, to some extent, by selection bias and technique. Comparing the outcomes of various modalities is hampered by the relatively low incidence of pharyngeal wall carcinomas and the limited literature addressing this issue.

Local Control

Julieron et al.8 reported on 77 patients treated surgically at the Institut Gustave-Roussy (Villejuif, France) for squamous cell carcinoma for the posterior pharyngeal wall between 1984 and 1995. Fifty-four patients were previously untreated, 11 patients had received previous radiotherapy for a metachronous head and neck carcinoma, and 12 patients received surgery for recurrent disease after having already underwent definitive radiotherapy. Local control rates after surgery and postoperative radiotherapy for the 54 previously untreated patients were 92% (24 of 26 patients) for tumors measuring less than 4 cm, 86% (24 of 28 patients) for tumors larger than 4 cm, and 89% (48 of 54 patients) for the overall control rate. Forty-one patients underwent larynx-preserving surgery and 13 patients underwent surgery that included a total laryngectomy.

Collins7 reported on 11 previously untreated patients who underwent surgery and postoperative radiotherapy at Loyola University (Maywood, IL) for posterior pharyngeal wall carcinoma. The larynx was preserved in 8 of 11 patients. Patients had follow-up for a median of 22 months. Local control rates are 80% (4 of 5 patients) for T2 tumors, 100% (4 of 4 patients) for T3 tumors, 50% (1 of 2 patients) for T4 tumors, and 82% (9 of 11 patients) for the overall control rate.

Marks et al.26 reported on 89 patients treated for squamous cell carcinomas of the pharyngeal wall at the Washington University Mallinckrodt Institute of Radiology (St. Louis, MO) between 1964 and 1981. Patients were required to have a 1-year minimum follow-up to be evaluable for local control. The local control rate after low-dose preoperative radiotherapy and surgery in 36 evaluable patients was 61% compared with 24% in 25 evaluable patients who received definitive radiotherapy. Four patients who died of complications after preoperative radiotherapy and surgery were excluded from the analysis of local control.

Meoz-Mendez et al.27 reported on 164 patients treated with radiotherapy alone at The University of Texas M. D. Anderson Cancer Center (Houston, TX) between 1954 and 1974. The following local and ultimate local control rates were obtained: T1 disease, 91% and 100%; T2 disease, 73% and 78%; T3 disease, 61% and 71%; and T4 disease, 37% and 41%, respectively. Chang et al.28 reported on 74 patients treated at the Oregon Health Sciences University (Portland, OR) between 1971 and 1991 with continuous-course radiotherapy alone or combined with adjuvant chemotherapy. Patients had follow-up for 2 years or longer. The following local control rates were obtained: T1 disease, 100%; T2 disease, 55%; T3 disease, 31%; and T4 disease, 29%. Local control was improved for patients treated with twice-daily radiotherapy.

Wang29 reported on the following 5-year local control rates after definitive radiotherapy in 105 patients treated at the Massachusetts General Hospital (Boston, MA) between 1970 and 1994: T1 disease, 88%; T2 disease, 55%; and T3–T4 disease, 49%. The local control rates 5 years after radiotherapy at the University of Florida were T1 disease, 93%; T2 disease, 82%; T3 disease, 59%; and T4, 50%.

Local control rates after surgery and adjuvant radiotherapy are comparable to those observed after definitive radiotherapy for patients with T1–T2 disease. Although local control may be improved after surgery for T3–T4 disease, there are too few data to definitively make this determination.

Locoregional Control

Julieron et al.8 observed that 42 of 54 patients (78%) who underwent surgery and postoperative radiotherapy for previously untreated pharyngeal wall carcinomas achieved locoregional control. Nakatsuka et al.30 (University of Tokyo, Tokyo, Japan) reported on eight previously untreated patients who underwent surgery alone or combined with adjuvant radiotherapy for T1–T2 posterior pharyngeal wall carcinomas. Six of the eight patients achieved locoregional control.

Marks et al.26 reported locoregional control rates of 42% after low-dose preoperative radiotherapy and surgery in 36 evaluable patients compared with 24% after definitive radiotherapy in 25 evaluable patients. Pene et al.15 reported on 122 patients treated with radiotherapy at the Institut Gustave Roussy (Villejuif, France) between 1958 and 1973 and observed the following 5-year locoregional control rates: primary tumor less than 4 cm, 16 of 23 patients (70%); primary tumor larger than 4 cm, 10 of 99 patients (10%); and overall, 26 of 122 patients (21%). In contrast, the 5-year locoregional control rates after definitive radiotherapy at the University of Florida were I, 89%; II, 83%; III, 58%; IV, 47%; and overall, 59%.

Locoregional control rates after surgery and radiotherapy or definitive radiotherapy appear to be comparable for early-stage lesions. Data pertaining to advanced-stage lesions are conflicting, with particularly poor outcomes for some of the older definitive radiotherapy series and improved outcomes from the University of Florida experience. The locoregional control rates in the series reported by Meoz-Mendez et al.27 from the M. D. Andersen Cancer Center appear to be comparable to the current series.

Survival

Julieron et al.8 reported a 5-year survival rate of 35% for 54 previously untreated patients who underwent surgery and postoperative radiotherapy for posterior pharyngeal wall carcinomas. Meoz-Mendez et al.27 reported on 25 patients treated with surgery and adjuvant radiotherapy at the M. D. Anderson Cancer Center. Of the 25 patients, 4 (16%) were alive at 5 years. Collins7 described 11 patients treated with surgery and adjuvant radiotherapy at Loyola University. They had a median follow-up of 22 months. Four patients were alive and disease free, three died of cancer, and four died of intercurrent disease. Lydiatt et al.31 reported on nine patients treated with larynx-preserving surgery at the Memorial Sloan-Kettering Cancer Center (New York) for T2 (five patients) and T3 (four patients) carcinomas of the posterior pharyngeal wall. Seven patients received adjuvant radiotherapy. Three patients were alive and disease free at 11 months, 13 months, and 49 months; 4 patients were alive or died of recurrent disease; and 2 patients died of intercurrent disease at 4 months and 9 months, respectively. Marks et al.26 reported a 5-year survival rate of 22% for 41 patients treated with low-dose preoperative radiotherapy and surgery compared with 6% for 41 patients treated with radiotherapy. Of the 41 patients in the latter group, 7 received less than 50 Gy.

Pene et al.15 reported a 5-year survival rate of 3% for 122 patients treated with radiotherapy alone at the Institut Gustave Roussy between 1958 and 1973. Talton et al.32 reported a 5-year survival rate of 35% for 24 patients treated with radiotherapy alone at the University of Virginia Hospital (Charlottesville, VA) between 1965 and 1975. Chang et al.28 reported the following 3-year overall and cause-specific survival rates for 74 patients treated with radiotherapy: Stage I, 75% and 100%; Stage II, 67% and 85%; Stage III, 33% and 58%; and Stage IV, 30% and 40%, respectively. Wang29 observed the following 5-year cause-specific survival rates in 105 patients treated with definitive radiotherapy: T1 disease, 83%; T2 disease, 57%; and T3–T4 disease, 42%. The 5-year cause-specific and absolute survival rates were 49% and 30%, respectively, for 148 patients treated with definitive radiotherapy at the University of Florida.

Although there are limitations inherent in comparing the results of retrospective, single-institution studies, there does not appear to be a survival advantage associated with primary surgery for patients with pharyngeal wall carcinomas.

Complications

Julieron et al.8 reported the complications and functional results for 55 patients who underwent larynx-preserving surgery between 1984 and 1995. Fourteen patients had received previous radiotherapy. Complications occurred in 12 of 55 patients (22%), including pneumonia (8 patients), cardiovascular events (3 patients), sepsis (4 patients), and/or hemiplegia (3 patients). None of the complications resulted in death. Twenty-five of 55 patients (45%) experienced “local” complications that included hematoma (2 patients), cervical abscess (18 patients), salivary fistula (9 patients), flap loss (6 patients), and “other” (7 patients). One patient required a completion total laryngectomy and another patient required a permanent tracheostomy. Six patients (11%) required a permanent gastrostomy tube and 17 patients (31%) had 1 or more episodes of aspiration pneumonia. Functional results were analyzed at 1 year in 42 evaluable patients: normal diet, 28 patients (67%); soft diet, 4 patients (10%); liquid diet, 2 patients (4%); and tube-dependent diet, 8 patients (19%).

Collins7 reported on 11 patients who underwent surgery and postoperative radiotherapy for previously untreated pharyngeal wall carcinomas. Of the 11 patients, 8 underwent a larynx-preserving procedure. After treatment, one patient had an open tracheostomy, three patients had a plugged tracheostomy, four patients had no tracheostomy, and three patients had a stoma. Three patients were gastrostomy tube dependent, two patients could take food orally but required gastrostomy tube supplementation, and six patients did not require a feeding tube. Nakatsuka et al.30 reported on eight patients who underwent surgery alone or combined with adjuvant radiotherapy for T1–T2 posterior pharyngeal wall carcinoma. Four patients underwent a forearm flap reconstruction and four patients received a jejunal flap reconstruction. Complications included infection (one patient) and loss of a jejunal flap necessitating a deltopectoral flap (one patient). Lydiatt et al.31 reported on nine patients who underwent resection with laryngeal preservation and radial foreman flap reconstruction for T2–T3 pharyngeal wall carcinomas. Complications included adult respiratory distress syndrome (two patients), wound infection (two patients), aspiration pneumonia (one patient), and flap loss (one patient). One patient died of a complication. Marks et al.26 reported on 38 patients treated with low-dose preoperative radiotherapy and surgery at Washington University Mallinckrodt Institute of Radiology. Complications included pharyngocutaneous fistula in 11 patients (29%), carotid rupture in 4 patients (11%), airway obstruction in 3 patients (8%), and pharyngeal hemorrhage in 1 patient (3%). Four patients (11%) died of complications. The incidence of complications decreased during the latter years of the study.

Marks et al.26 reported on 34 patients treated with definitive radiotherapy at Washington University Mallinckrodt Institute of Radiology. Of 34 patients, 7 (21%) required tracheostomy for airway obstruction. Meoz-Mendez et al.27 observed severe complications after definitive radiotherapy and/or salvage surgery for recurrent disease in 20 of 164 patients (12%) treated at the M. D. Anderson Cancer Center. Complications included soft tissue necrosis (three patients), carotid rupture (eight patients), bone necrosis (three patients), myelitis (two patients), severe laryngeal edema (two patients), and severe neck fibrosis (two patients). Eight patients (5%) died of complications. Eight of 148 patients (5%) treated at the University of Florida died of a complication.

Patients with T1–T2 pharyngeal wall carcinomas appear to have a similar locoregional control rate after surgery alone or combined with adjuvant radiotherapy compared with definitive radiotherapy. Locoregional control may be somewhat better after surgery for T3–T4 tumors, but at the expense of more complications and poorer function. Regardless of the initial treatment approach, survival is about the same. For these reasons, our bias is to treat essentially all previously unirradiated patients with definitive radiotherapy.

Our current approach is to employ altered fractionation using hyperfractionation for patients undergoing three-dimensional conformal radiotherapy. Patients receive 74.4–76.8 Gy at 1.2 Gy per fraction twice daily with a minimum 6-hour interfraction interval. Patients who undergo intensity-modulated radiotherapy receive 72 Gy in 42 fractions over 6 weeks using the M. D. Anderson Cancer Center concomitant-boost technique. Patients with T3–T4 and/or bulky N2/N3 disease receive concomitant weekly cisplatin or weekly carboplatin and paclitaxel. Patients with clinically positive neck lymph nodes undergo a CT scan 4 weeks after chemoradiation and a neck dissection is performed if the likelihood of residual neck disease is believed to be greater than 5%.18 Patients receive aggressive swallowing therapy during and after chemoradiation to reduce the risk of long-term gastrostomy tube dependence.

Ancillary