Wang, Shulian MD*; Liao, Zhongxing MD‡; Chen, Yuan MD†; Chang, Joe Y. MD, PhD‡; Jeter, Melanda MD, MPH; Guerrero, Thomas MD, PhD‡; Ajani, Jaffer MD§; Phan, Alexandria MD§; Swisher, Stephen MD∥; Allen, Pamela MPH‡; Cox, James D. MD‡; Komaki, Ritsuko MD‡
Esophageal cancer located at the cervical and upper thoracic area is rare, representing less than 10% of esophageal cancer.1 Some believe that the biological behavior of esophageal cancer at this location differs from those at the mid and lower esophagus or gastroesophageal junction, because they are mostly squamous-cell histology with local invasiveness and less prone to distant metastasis, and that they should be treated like head and neck cancer. In one study, a more favorable cause-specific survival was observed in patients with upper-third tumors than in those with middle- or lower-third tumors when treated with definitive chemoradiotherapy.1
Treatment is challenging because of the high risk of adjacent anatomical structure invasion, which precludes radical resection of the tumor. Even in patients with resectable tumor, surgery is often an unacceptable option because a total laryngectomy is usually required.2 In four studies of patients with this type of cancer, those treated with surgery had morbidity rates of 60 to 70%, mortality rates of 7 to 11%, and a 5-year overall survival (OS) rate of 18 to 27%.3–6
Information on nonsurgical treatment of this tumor is sparse because of the rarity of its occurrence. In a few retrospective studies that focused on this subgroup of patients, concurrent chemoradiation was used as the treatment of choice.7–12 Concurrent chemoradiotherapy has become a standard treatment for surgically unresectable esophageal cancer since the results of an intergroup, randomized, phase III trial (RTOG 8501) were reported in 1999, which included tumors of the thoracic esophagus.13 There was no information on the location of the tumor or subgroup analysis on this group of patients. Because so little information is available on carcinoma at this location, we conducted a retrospective analysis of data from patients with cervical and upper thoracic esophageal cancer who had been treated with concurrent chemoradiotherapy, with or without induction chemotherapy, at our institution over a period of 11 years. To characterize the treatment and outcome for this group of patients, we analyzed the rates of clinical response, OS, cause-specific survival (CSS), disease-free survival (DFS), local relapse-free survival (LRFS), and distant metastasis-free survival (DMFS). We also investigated prognostic factors, particularly radiation dose, for the endpoints measured.
PATIENTS AND METHODS
From January of 1985 to December of 2001, a total of 703 patients with esophageal cancer were treated at the Department of Radiation Oncology at The University of Texas M. D. Anderson Cancer Center. Among them, 57 patients (8.1%) presented with a cervical or upper thoracic tumor. The medical records of these patients were retrospectively reviewed. To be included in our analysis, patients had to meet the following criteria: newly pathologically confirmed cervical and upper thoracic esophageal cancer (tumor located above the carina), no distant metastasis at presentation, and treatment with concurrent chemoradiotherapy but without surgery. We excluded patients if they had had distant metastases at presentation (n = 10), treatment by means other than surgery or radiotherapy (n= 4), treatment with surgery (n = 3), radiotherapy that had not been completed or followed up (n = 2), treatment with radiotherapy alone (n = 1), recurrent disease (n = 1), or two primary tumors both in the cervical and lower esophagus (n = 1). Thus, we included data on 35 patients in this analysis. We received approval for this retrospective review from our institutional review board, and we complied with all Health Insurance Portability and Accountability Act regulations.
The following pretreatment evaluations were performed on the 35 patients: all had a medical history interview, a physical examination, and laboratory studies, including complete blood count and biochemical survey; radiographic studies, including esophagograms (n = 30 [86%]), chest radiograph (n = 34 [97%]), computed tomographic (CT) scans of the chest (n = 35 [100%]), CT scans of the neck (n = 13 [37%]), and CT scans of the abdominal and pelvic scans (n = 30 [86%]); bone scan (n = 6 [17%]); and, when needed, CT or magnetic resonance imaging scans of the brain (n = 3 [9%]). Other specific studies included esophagogastroduodenoscopy with biopsy (n = 35 [100%]), endoscopic ultrasound of the esophagus (n = 6 [17%]), and bronchoscopic examination with or without brushing or biopsy (n = 15 [43%]).
Various induction chemotherapy regimens had been used: two to three cycles of 5-fluorouracil (5-FU) and cisplatin had been given to three patients; six cycles of 5-FU, cisplatin, and paclitaxel had been given to one patient; two cycles of 5-FU, carboplatin, and paclitaxel had been given to one patient; and two cycles of paclitaxel had been given to one patient. Concurrent chemotherapy had consisted of four different regimens of 5-FU alone or in combination with cisplatin, irinotecan, or carboplatin. Ten patients had received continuous infusions of 5-FU alone (250 mg/m2) daily Monday through Friday for 5 weeks, and 10 patients had received continuous infusion of 5-FU (250 mg/m2) daily Monday through Friday for 2 weeks (the number of weeks depended on the duration of radiotherapy). 5-FU plus cisplatin had been administered to 12 patients in one of the following three different regimens: six patients had received continuous infusion of 5-FU (750 mg/m2) on days 1 through 5 and days 29 through 33, with cisplatin (75 mg/m2) given on days 1 and 29; four patients had received continuous infusion of 5-FU (1000 mg/m2) on days 1 through 4 and days 29 through 32 and cisplatin (75 mg/m2) given on days 1 and 29; and two patients had received continuous infusion of 5-FU (750 mg/m2) on days 1 through 5 and days 29 through 33 and cisplatin (15 mg/m2) on days 1 to 5 and days 29 to 33. Three patients had been treated with regimens that did not include cisplatin. One of these had received continuous infusion of 5-FU (250 mg/m2) daily Monday through Friday and carboplatin (area under the curve, 1.5) weekly for 5 weeks. The second of these three patients had received continuous infusion of 5-FU (300 mg/m2) daily Monday through Friday and irinotecan (30 mg/m2) weekly for 5 weeks. The third had received carboplatin alone (80 mg/m2) every 3 weeks for 9 weeks.
Before January of 2000, conventional radiation techniques were used. For patients who were irradiated with less than or equal to 30 Gy, anteroposterior (AP) and posteroanterior (PA) fields were used. For patients who were irradiated with greater than 40 Gy, AP and PA fields were used to deliver a total dose of up to 40 Gy, and then oblique fields were used to spare the spinal cord. Beginning in January of 2000, a three-dimensional conformal radiotherapy technique was used. The initial target volume encompassed the primary tumor, with 5-cm cephalad and caudal margins and a 2-cm radial margin, with a field arrangement of AP + PA + oblique. The supraclavicular and midcervical nodal region was treated for all patients.
Typically, two fractionation schedules were used: the “rapid-fractionation” (30 Gy given in 10 fractions within 2 weeks) and standard (≥46 Gy given at 1.8–2.0 Gy per fraction daily, with five fractions administered weekly). The rapid fractionation was used based on the principle that the total radiation dose required to obtain a given biological effect decreases as the dose per fraction increases. A 30-Gy total dose of radiation given in 10 fractions was considered radiobiologically equivalent to a standard 5.5-week (50.4 Gy/28 fractions) program with a shortened overall treatment time. It was designed as a definitive treatment, combined with concurrent chemotherapy for esophageal, pancreatic, and periampullary carcinomas and used for prospective clinical trials during the early 1990s in this institution. Radiation therapy was primarily delivered with 18-MV photons, 6-MV photons, or both. The total radiation dose varied from 24.5 to 64.8 Gy, with a median dose of 50.4 Gy at 1.8 Gy per fraction. Table 1 shows the distribution of the range of radiation doses.
Patients had undergone the following types of evaluations every 3 months for 2 years for the first 2 years and every 6 months thereafter: a physical examination, a complete blood cell count, blood chemistry tests, esophagograms, and chest radiography. Patients had also undergone endoscopic examinations; esophageal biopsies; and CT scans of the chest, neck, or abdomen, as indicated.
Complete remission (CR) of the primary tumor had been determined with endoscopy with or without biopsy, with esophagography plus CT, or with both. A CR was defined as the point at which all visible tumors had disappeared for 4 weeks or longer. The response of the metastatic lymph nodes had been assessed by CT scan, and a CR was defined as the complete disappearance of all measurable and assessable disease for 4 weeks or longer.
Outcome Measures and Statistical Analyses
We defined relapse of the disease as any distant recurrence, regional lymph node recurrence, local recurrence, or persistent disease during the follow-up period. The definition of local relapse included local recurrence after tumor complete remission, or progression of persistent disease. Distant relapse included recurrence at any site other than those of the primary tumor and regional lymph nodes.
Death from any cause was considered the endpoint for OS, and death from esophageal cancer was the endpoint for CCS. The survival analysis was performed by using the Kaplan-Meier method, and the log-rank test was used to compare the survival curves. The Cox proportional hazards model was used for multivariate analyses. The time to an event was calculated using the date of diagnosis. Pearson’s chi-square test was used to assess measures of association in frequency. A value of p < 0.05 was considered statistically significant.
Patients’ Characteristics and Treatments
Among esophageal cancer patients who were treated with radiation therapy, 8.1% had cervical and upper esophageal tumor. Patients’ ages ranged from 54 to 86 years (median, 64 years). There were 31 cases of squamous-cell carcinoma and four cases of adenocarcinoma. The tumor stage was distributed as follows for the 35 patients: T1N0 (n = 1), T2N0 (n = 9), T3N0 (n = 7), T2N1 (n = 1), T4N0 (n = 4), T3N1 (n = 8), T4N1 (n = 3), and TxN0 (n = 2). Tumor stage for all patients was based on CT scans. Endoscopic ultrasound examinations were performed for tumor staging only when the ultrasound probe could be introduced through the lumen. The patients’ characteristics and treatments are summarized in Table 2. Eleven patients received a radiation dose of less than 50 Gy and 24 received greater than or equal to 50 Gy. There was a significantly higher percentage of patients with greater than 10% of weight loss and higher disease stage in the group that received less than 50 Gy.
Of the 35 patients, 22 (62.9%) experienced a CR of the primary tumor after concurrent chemoradiotherapy, three (8.6%) had achieved a partial remission, three had achieved a minor remission (8.6%), two had achieved stable disease (5.7%), two had developed progressive disease (5.7%), and three (8.6%) had nonassessable responses. Among the 22 patients with CR in the primary tumor, 14 were evaluated by both endoscopy and radiography, seven by radiography only, and one by endoscopy only. Of the 12 patients who had had nodal metastases at presentation, three (25%) had achieved a CR, one (8.3%) had developed progressive disease, and eight (66.7%) had responses that were not assessable.
Patients who had been given a radiation dose of greater than or equal to 50 Gy had a significantly higher primary tumor CR rate than did those who had been given a dose of less than 50 Gy (79.2% and 27.3%, respectively; p = 0.003). All 11 patients who had had T1 and T2 disease received a radiation dose of greater than or equal to 50 Gy, and their primary tumor CR rate was 81.8%. For patients who had had T3 and T4 disease, the primary tumor CR rates were 75% in those who had received a radiation dose of greater than or equal to 50 Gy and 30% for those who had received less than 50 Gy; this difference was statistically significant (p = 0.035).
The effect of a CR of the primary tumor on outcome was also analyzed. Patients who had experienced a CR had significantly higher OS (p = 0.008), CSS (p = 0.002), DFS (p < 0.001), and LRFS (p < 0.001) rates than did those who did not. However, CR was not associated with DMFS (p = 0.52). Figure 1 illustrates the OS, CSS, LRFS, and DMFS curves for patients who achieved a primary tumor CR and for those who did not.
Patterns of Relapse
The median follow-up in surviving patients was 39 months (range, 3–64 months). Sixteen patients had experienced a local relapse after an initial CR (n = 4) or after persistent disease in the esophagus (n = 12). The median time from diagnosis to local relapse or progression of the persistent disease was 6 months (range, 1–14 months). Two patients had developed regional lymph node relapse: one who had experienced relapse in the anterior mediastinum and also had developed a distant metastasis, and the other who had experienced relapse in the posterior neck and also had developed persistent disease in the esophagus. Twelve patients had developed distant metastases within 12 months of their diagnosis (range, 3–12 months; median, 6 months); six of those metastases had developed in the lung; two in the lung and bone; one in the lung, bone, and liver; one in the liver; and two in the bone. Three of these 12 patients had developed distant metastases with a local relapse or persistent disease, and nine had had distant metastases without a local relapse or persistent disease.
Survival and Prognostic Factors
The median survival time for the 35 patients was 13 months, and the actuarial 5-year rates for OS, CSS, DFS, LRFS, and DMFS were 18.6%, 27.6%, 22.4%, 47.7%, and 57.0%, respectively. Twenty-six of the 35 patients died, 22 as a result of esophageal cancer, three as a result of disease unrelated to cancer, and one as a result of a second primary tumor in the lung. Of the nine patients who had survived, six had had no evidence of disease and three had had local or distant disease or both.
We used 11 factors for the univariate analysis of survival rates: sex, age, weight loss, Karnofsky performance status score, tumor location, tumor histologic type, tumor stage, nodal stage, clinical stage, induction chemotherapy, and radiation dose. We found that the radiation dose was the only factor that was significantly related to OS (p = 0.002) and CSS (p = 0.0009). The 5-year OS rates were 0% and 29% for patients who had received a radiation dose of less than 50 Gy and greater than or equal to 50 Gy, respectively. The 5-year CSS rates were 0% and 44% for patients who had received a radiation dose of less than 50 Gy and greater than or equal to 50 Gy, respectively. Radiation dose (p = 0.03), tumor stage (p = 0.009), and clinical stage (p = 0.020) were significant factors in predicting DFS. Patients who had received a radiation dose of greater than or equal to 50 Gy and who had T1 or T2 disease and clinical stage I or II disease had higher DFS rates than other patients. Radiation dose (p = 0.0001), weight loss (p = 0.029), tumor stage (p = 0.034), and clinical stage (p = 0.039) were significant factors in predicting LRFS rate. Patients who had received a radiation dose of greater than or equal to 50 Gy and had had a weight loss of less than or equal to 10%, T1 or T2 disease, and clinical stage I or II disease had a better LRFS rate than those of the other patients. No factor had a significant effect on the DMFS rate. Only one of the six patients who had received induction chemotherapy developed a distant metastasis, compared with 11 of the 29 patients who had not received it, but the difference was not statistically significant (p = 0.263).
In the multivariate analysis, radiation dose, weight loss, tumor stage, and clinical stage were included as variables. Radiation dose was the only factor associated with OS (p = 0.006), CSS (p = 0.003), and LRFS (p = 0.001). Tumor stage was the only factor found to be associated with DFS (p = 0.007). Figure 2 shows the OS, CSS, LRFS, and DMFS curves for patients who received radiation doses of greater than or equal to 50 Gy and less than 50 Gy.
In this study, we found that the survival rates of patients with cervical and upper thoracic esophageal cancer who had received concurrent chemoradiotherapy were comparable to the results reported in the literature (Table 3). We also found that the OS, CSS, and LDFS rates were significantly higher in patients who had received a radiation dose of greater than or equal to 50 Gy than in those who had received a dose of less than 50 Gy.
In our study, radiation dose was the only independent factor associated with improved local control and overall survival; however, because of the small number of patients and the retrospective nature of this study, conclusions should be drawn with caution.
The data to support the existence of a dose-dependent response in esophageal cancer are not strong. Three studies using high-dose, three-dimensional, conformal radiation therapy concurrently with chemotherapy have yielded promising results. In one study, patients with a large tumor burden (T3 or T4 primary) in the upper thoracic and midthoracic esophagus were given a higher dose of up to 65 Gy, and the long-term survival was 31% at 3 years.11 In another study, The 5-year actuarial survival rate was 55% when high-dose radiotherapy (median, 61.2 Gy; range, 50.4 Gy/20 fractions–65 Gy/33 fractions) was given for patients with cervical esophageal cancers, although the majority of patients in this group had early-stage tumors.8 The 5-year survival of 50% was also observed in another study in which a total radiation dose of 60 Gy was delivered to patients with upper-third esophageal cancers.6
The only randomized dose-escalation study (Intergroup Trial 0123) shows that increasing the radiation dose from 50.4 Gy to 64.8 Gy does not result in better local control or survival rate among patients with esophageal cancer who were treated with concurrent chemoradiotherapy.14 However, seven of the 11 treatment-related deaths in the high-dose arm of the RTOG 94-05 trial occurred in patients who had received 50.4 Gy or less. There was a significant prolongation of treatment time because of breaks required for recovery from side effects after correction for the number of radiation treatments and a significantly lower dose of 5-FU given to patients on the high-dose arm. The authors believed that these factors might have contributed, at least in part, to the lack of benefit for patients who received high-dose versus standard-dose radiotherapy.14 The findings in the RTOG 94-05 study warrant extensive research on methods to reduce treatment toxicity to allow radiation dose intensification for patients who are not considered candidates for surgery and for whom chemoradiation is the only treatment alternative.
A pathologic CR was found in 24 to 43% of patients treated with neoadjuvant concurrent chemoradiotherapy with radiation doses ranging from 37 to 45.6 Gy given in various fractions.15–17 The CR rate in our study was 63%, which is comparable to those in other studies (47–91%) for cervical and upper thoracic esophageal cancer treated with concurrent chemoradiotherapy.7–9 In the 35 patients in our analyses, a higher radiation dose was associated with a higher CR rate, and a higher CR rate was associated with higher 5-year OS and CSS. Because patients with an early clinical tumor stage had all received high-dose radiation, we also performed a dose-response analysis on patients who had had advanced tumors and found similar results. Thus, as indicated by the results of those previous studies and ours, it appears that the radiation dose of 50 to 65 Gy should be used as a definitive treatment for this patient population.
The development of distant metastasis is an obstacle to improving the survival rate in patients with this disease. Because 75% (nine of 12) of the distant metastases developed in our patients who had not experienced a local relapse, a higher radiation dose and CR in the primary tumor improved local tumor control but failed to improve DFS because of the higher rate of distant metastases. Theoretically, because all the distant metastases in our patients developed within 12 months of diagnosis, induction chemotherapy or intensified concurrent chemotherapy might be beneficial in the early treatment of occult micrometastases. Unfortunately, however, induction chemotherapy did not significantly reduce the number of distant metastases in our small number of patients. In an intergroup trial (INT 0122) reported in 1999, treatment with three cycles of induction chemotherapy (5-FU and cisplatin) plus concurrent chemoradiotherapy (5-FU and cisplatin plus 64.8 Gy radiation) did not appear to improve the 5-year OS rate (20%) or the distant metastasis rate (24%), and 9% of the patients died as a result of complications of treatment.18
Newer agents than 5-FU and cisplatin that are being used clinically or are still investigational for treating esophageal cancer are showing promise for increasing local control and decreasing distant metastasis. Preliminary results of treatments that have included paclitaxel or irinotecan are encouraging and have shown a pathologic CR rate of approximately 60%.19,20 The use of molecular targeting agents such as cetuximab in combination with radiotherapy, which yielded positive results in treating head and neck cancer,21 is being investigated for treating esophageal cancer in combination with cisplatin, paclitaxel, and radiotherapy in the RTOG 0436 trial. Whether these investigational approaches will result in better responses than are achieved with the conventional chemoradiation regimens based on 5-FU and cisplatin remains to be seen.
In summary, we believe that concurrent chemoradiotherapy is a good treatment option for patients with cervical and upper thoracic esophageal cancer. Our results suggest that a total radiation dose of 50 to 65 Gy with a concurrent chemotherapy regimen may improve local control and the OS rate in this rare type of esophageal cancer.
This project was supported by the Radiology Society of North America Radiology Research and Education Program grant to “Teach the Teachers” from Emerging Nations, 2001. The authors thank Barbara E. Lewis for her excellent assistance in the preparation of this article.
1. Denham JW, Burmeister BH, Lamb DS, et al. Factors influencing outcome following radio-chemotherapy for oesophageal cancer. The Trans Tasman Radiation Oncology Group (TROG). Radiother Oncol 1996;40:31–43.
2. Triboulet JP, Mariette C, Chevalier D, et al. Surgical management of carcinoma of the hypopharynx and cervical esophagus: analysis of 209 cases. Arch Surg 2001;136:1164–1170.
3. Kelley DJ, Wolf R, Shaha AR, et al. Impact of clinicopathologic parameters on patient survival in carcinoma of the cervical esophagus. Am J Surg 1995;170:427–431.
4. Kakegawa T, Yamana H, Ando N. Analysis of surgical treatment for carcinoma situated in the cervical esophagus. Surgery 1985;97:150–157.
5. Vigneswaran WT, Trastek VF, Pairolero PC, et al. Extended esophagectomy in the management of carcinoma of the upper thoracic esophagus. J Thorac Cardiovasc Surg 1994; 107: 901–906.
6. Wang HW, Kuo KT, Wu YC, et al. Surgical results of upper thoracic esophageal carcinoma. J Chin Med Assoc 2004;67:447–457.
7. Ampil FL, Mills GM, Burton GV. Induction chemotherapy followed by concomitant chemoradiation for cervical esophageal cancer. Am J Gastroenterol 1999;94:2325–2326.
8. Burmeister BH, Dickie G, Smithers BM, et al. Thirty-four patients with carcinoma of the cervical esophagus treated with chemoradiation therapy. Arch Otolaryngol Head Neck Surg 2000;126:205–208.
9. Bidoli P, Bajetta E, Stani SC, et al. Ten-year survival with chemotherapy and radiotherapy in patients with squamous cell carcinoma of the esophagus. Cancer 2002;94:352–361.
10. Stuschke M, Stahl M, Wilke H, et al. Induction chemotherapy followed by concurrent chemotherapy and high-dose radiotherapy for locally advanced squamous cell carcinoma of the cervical oesophagus. Oncology 1999;57:99–105.
11. Stuschke M, Stahl M, Wilke H, et al. Induction chemotherapy followed by concurrent chemotherapy and high-dose radiotherapy for locally advanced squamous cell carcinoma of the upper-thoracic and midthoracic esophagus. Am J Clin Oncol 2000;23:233–238.
12. Seung SK, Smith JW, Molendyk J, et al. Selective dose escalation of chemoradiotherapy for esophageal cancer: role of treatment intensification. Semin Oncol 2004;31:13–19.
13. Cooper JS, Guo MD, Herskovic A, et al. Chemoradiotherapy of locally advanced esophageal cancer: long-term follow-up of a prospective randomized trial (RTOG 85-01). Radiation Therapy Oncology Group. JAMA 1999;281:1623–1627.
14. Minsky BD, Pajak TF, Ginsberg RJ, et al. INT 0123 (Radiation Therapy Oncology Group 94-05) phase III trial of combined-modality therapy for esophageal cancer: high-dose versus standard-dose radiation therapy. J Clin Oncol 2002;20:1167–1174.
15. Urba SG, Orringer MB, Turrisi A, et al. Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 2001;19:305–313.
16. Bosset JF, Gignoux M, Triboulet JP, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 1997;337:161–167.
17. Lee JL, Park SI, Kim SB, et al. A single institutional phase III trial of preoperative chemotherapy with hyperfractionation radiotherapy plus surgery versus surgery alone for resectable esophageal squamous cell carcinoma. Ann Oncol 2004;15:947–954.
18. Minsky BD, Neuberg D, Kelsen DP, et al. Final report of Intergroup Trial 0122 (ECOG PE-289, RTOG 90-12): phase II trial of neoadjuvant chemotherapy plus concurrent chemotherapy and high-dose radiation for squamous cell carcinoma of the esophagus. Int J Radiat Oncol Biol Phys 1999;43:517–523.
19. Meluch AA, Greco FA, Gray JR, et al. Preoperative therapy with concurrent paclitaxel/carboplatin/infusional 5-FU and radiation therapy in locoregional esophageal cancer: final results of a Minnie Pearl Cancer Research Network phase II trial. Cancer J 2003;9:251–260.
20. Ilson DH, Minsky B. Irinotecan in esophageal cancer. Oncology (Williston Park) 2003;17:32–36.
21. Bonner JA, Harari PM, Giralt J, et al. Cetuximab prolongs survival in patients with locoregionally advanced squamous cell carcinoma of head and neck: a phase 3 study of high dose radiotherapy with or without cetuximab (Abstract 5507). Proc Am Soc Clin Oncol 2004; 22: 489a.