2499 The Evaluation and Treatment of Patients Receiving Radiation Therapy for Carcinoma of the Esophagus Results of the 1992–1994 Patterns of Care Study Lawrence R. Coia, M.D.1 Bruce D. Minsky, M.D.2 Madhu J. John, M.D.3 Daniel G. Haller, M.D.4 Jerome Landry, M.D.5 Thomas M. Pisansky, M.D.6 Christopher G. Willett, M.D.7 John P. Hoffman, M.D.8 Brian A. Berkey, M.S.9 Jean B. Owen, Ph.D.9 Gerald E. Hanks, M.D.8 BACKGROUND. For the first time, a Patterns of Care Study (PCS) was conducted in 1992–1994 to determine the national practice standards in evaluating and treating patients with esophageal carcinoma and to determine the degree to which clinical trials have been incorporated into national practice. METHODS. A national survey of 61 institutions using 2-stage cluster sampling was conducted, and specific information was collected on 400 patients with squamous cell carcinoma or adenocarcinoma of the thoracic esophagus who received radiation therapy (RT) as part of definitive or adjuvant management of their disease. Patients were staged according to a modified 1983 American Joint Committee on Cancer staging system. Chi-square tests for significant differences between academic and nonacademic institutions for a particular variable were performed. 1 Community Medical Center, Toms River, New Jersey. 2 RESULTS. The median age of patients was 66.7 years (range, 26 – 89 years); 76.5% were male and 23.5% were female. Karnofsky performance status was $80 for Memorial Sloan-Kettering Cancer Center, New York, New York. 88.3% of patients. Squamous cell carcinoma was diagnosed in 61.5% and adeno- 3 29.5% CS III. Evaluative procedures included endoscopy (.93%), computed to- Cancer Center at St. Agnes, Fresno, California. 4 University of Pennsylvania Medical Center, Philadelphia, Pennsylvania. 5 Emory University, Atlanta, Georgia. 6 carcinoma in 36.8%. Fifteen percent were Clinical Stage (CS) I, 39.5% CS II, and mography (CT) of the chest (86%), CT of the abdomen (75%), esophagography (68.5%), and endoscopic ultrasound (3.5%). Endoscopic ultrasound and CT of the chest were performed significantly more frequently at academic than nonacademic facilities (6.1% vs. 1.0% and 91.9% vs. 81.3%, respectively). Three-quarters of all Mayo Clinic and Mayo Foundation, Rochester, Minnesota. patients received chemotherapy and RT and 62.5% received concurrent chemo- 7 RT (54.0%), RT alone (20.3%), preoperative chemotherapy 1 RT (13.3%), postop- Cox Cancer Center, Boston, Massachusetts. 8 American College of Radiology, Philadelphia, Pennsylvania. 9 Fox Chase Cancer Center, Philadelphia, Pennsylvania. Presented at the 80th Annual Meeting of the American Radium Society, Monte Carlo, Monaco, May 2– 6, 1998. Supported by Grant No. CA94006 from the National Cancer Institute, National Institutes of Health. The authors thank Angela Katz, M.D., and Irene Mahon, R.N., M.P.H., for their technical assistance in data collection. therapy and RT as part of their treatment. Treatments included chemotherapy plus erative chemotherapy 1 RT (7.7%), postoperative RT (3.5%), and preoperative RT (1.2%). The chemotherapeutic agents most frequently used were 5-fluorouracil (84%), cisplatin (64%), and mitomycin (9%); academic instututions used cisplatin significantly more often and mitomycin significantly less often than nonacademic institutions. Brachytherapy was used in 8.5% of cases. The median total dose of external beam radiation was 50.4 gray and the median dose per fraction was 1.8 gray. CONCLUSIONS. This study establishes the national benchmarks for the evaluation and treatment of patients with esophageal carcinoma at radiation facilities in the U.S. It also indicates that the majority of patients given RT as a component of treatment for esophageal carcinoma receive chemoradiation rather than RT alone, as supported by clinical trials. Although some differences in the evaluation of esophageal carcinoma were noted between academic and nonacademic facilities, there was no difference in the frequency of use of chemoradiation versus RT by Address for reprints: Lawrence R. Coia, M.D., Department of Radiation Oncology, Community Medical Center, 99 Highway 37 West, Toms River, NJ 08755. facility type. Cancer 1999;85:2499 –505. © 1999 American Cancer Society. Received October 30, 1998; revision received March 1, 1999; accepted March 1, 1999. KEYWORDS: esophageal carcinoma, Patterns of Care Study, preoperative chemoradiation, chemoradiation alone. © 1999 American Cancer Society 2500 T CANCER June 15, 1999 / Volume 85 / Number 12 he Patterns of Care Study (PCS) periodically establishes national benchmarks for malignancies in which radiation therapy plays an important role. PCS is funded by grants and contracts from the National Cancer Institute (NCI) and administered by the American College of Radiology (ACR). PCS was begun in 1971 in an effort to improve the quality of and accessibility to radiation oncology throughout the U.S. Its operating premise was that differences in practice patterns exist, these difference can be measured, and deficiencies can be corrected. Major disease sites that PCS has periodically studied have included carcinomas of the uterine cervix, breast, prostate, and rectum, Hodgkin disease, and seminoma, among others.1 For the first time, PCS decided to study national practice patterns in the use of radiation to treat carcinoma of the esophagus, and the results are presented herein. Esophageal carcinoma is responsible for approximately 12,000 deaths annually in the U.S. Compared with rectal carcinoma, which occurs four times as frequently, there are more deaths annually from esophageal carcinoma.2 Radiation has been used in the treatment of esophageal carcinoma alone, preoperatively, or postoperatively. More recently, however, radiation has been combined with chemotherapy either as an adjuvant to surgery or as primary treatment. Because of the importance of esophageal carcinoma in the U.S. and the increasing role of radiation in its treatment, esophageal carcinoma was felt to be an appropriate disease for study by PCS. Along with the increased use of multimodal treatment for esophageal carcinoma there has been a change in the predominant histology. The incidence of adenocarcinoma of the esophagus has increased at a rate greater than that of nearly any other cancer over the past 2 decades.3 Although previously the ratio of squamous cell carcinoma of the esophagus to adenocarcinoma was approximately 9 to 1, the incidence rates of these histologies are currently approximately equal.4 Whether patients with adenocarcinoma of the esophagus should be treated similarly to patients with squamous cell carcinoma and whether the outcome of treatment varies according to histology are issues of interest to the practicing radiation oncologist. The potential benefits of concurrent chemotherapy and radiation (CRT) over radiation alone for esophageal carcinoma were suggested in several Phase II trials in the latter part of the 1980s.5–7 In 1992, Herskovic et al. published the results of a Phase III Intergroup clinical trial that demonstrated the superiority of CRT over radiation alone.8 Since the publication of those trial results, CRT has been considered one of the standard treatments for esophageal carcinoma. The objectives of this study were to determine the national practice standards for the evaluation and treatment of esophageal carcinoma with radiation. Furthermore, because significant improvement in the outcomes of patients treated with CRT over radiation alone had been demonstrated, this study sought to examine the penetrance of such clinical trial results into national practice. Specifically, this study selected and examined the records of patients with esophageal carcinoma treated with radiation between 1992 and 1994, when the results of the Intergroup became widely distributed. METHODS A national sample of 61 radiation facilities was selected from a master list of all facilities in the U.S. in 1992. A stratified two-stage cluster sample was used to select facilities and patients for review. In the first stage of sampling, facilities were selected. Nineteen academic and 42 nonacademic facilities were chosen. For the purposes of this study, academic facilities were defined as NCI-designated comprehensive cancer centers or the main teaching hospital of a medical school. All others were considered nonacademic. In the second stage, a randomized selection of records of patients from each facility treated with radiation for esophageal carcinoma was carried out. Data were collected between 1995 and 1997 by a PCS research associate who traveled to each facility and reviewed the records on site. All radiation therapy records were reviewed. In addition, hospital records were requested but not always available for review. The PCS Consensus Committee that dealt with esophageal carcinoma developed the survey forms used to collect the information. That committee had previously convened to discuss guidelines for the management of esophageal carcinoma and determined key areas of interest in the evaluation and management of the disease for this nationwide study. The results of their management guidelines have been published.9 To be eligible for this study, patients had to meet several criteria. First, patients were required to have been treated with radiation for esophageal carcinoma between 1992 and 1994. The decision was made to study patients with either squamous cell carcinoma or adenocarcinoma of the esophagus, including the thoracic esophagus and the gastroesophageal (GE) junction. Those with GE junction cancers were not studied if there was gastric extension. Patients with 1983 American Joint Commission on Cancer (AJCC) Stage I, II, or III esophageal carcinoma who had no evidence of distant metastasis were eligible for review. It should be noted that the staging system was a modification of the 1983 AJCC staging system, as shown in Table 1. Radiation for Esophageal Carcinoma/Coia et al. 2501 TABLE 1 Comparison between the 1983 and 1997 AJCC TNM Classification and Stage Groupings for Esophageal Carcinomaa Modified 1983 classification 1997 classification Primary Tumor (T) clinical classification Tis Carcinoma in situ T1 Tumor involves 5 cm or less of esophageal length, produces no obstruction, and has no circumferential involvement T2 Tumor involves more than 5 cm of esophageal length, causes obstruction, or involves the circumference of the esophagus T3 Extraesophageal spreadb T4 Lymph node involvement (N) pathologic classification N0 No positive lymph nodes (surgical evaluation) N1 Positive lymph nodes (surgical evaluation) NX Regional lymph nodes for the upper, midthoracic esophagus that are not ordinarily accessible for clinical evaluation (clinical evaluation) Distant metastasis (M) clinical classification M0 No distant metastasis M1 Distant metastasis Stage grouping Stage I T1, N0 or NX, M0 Stage II T2, N0 or NX, M0 Stage IIA Stage IIB Stage III T3, any N, M0 Stage IV Any T, any N, M1 Noninvasive Tumor invades lamina propria or submucosa Tumor invades muscularis propria Tumor invades adventitia Tumor invades adjacent structures No regional lymph node metastasis Regional lymph node metastasis No distant metastasis Distant metastasisc T1, N0, M0 T2–T3, N0, M0 T1–T2, N1, M0 T3, N1, M0 or T4, any N, M0 Any T, any N, M1 AJCC: American Joint Committee on Cancer; CT: computed tomography. a Data from 1983 and 1997 editions of the AJCC Manual for Staging of Cancer. b Extraesophageal spread is defined by recurrent laryngeal nerve involvement, positive pleural effusion, tracheoesophageal fistula, sympathetic nerve involvement, phrenic nerve involvement, or widened mediastinum on chest radiograph. In this modified staging system, CT scan evidence of local extraesophageal extension was allowed to indicate T3. c Distant metastases include cervical or celiac lymph node involvement for tumors of the thoracic esophagus. The 1983 AJCC staging system was used because it was a clinically based system that did not require the use of an initial surgical resection to determine stage. The 1997 AJCC system is also shown in Table 1 for comparison. Additional eligibility requirements were that patients had to have a Karnofsky performance status greater than 50 and could not have had another type of malignancy within the 5 years prior to treatment. National averages from sampling were developed by Sedransk et al.,10 and methodology for PCS development of national averages have recently been reviewed by Owen et al.11 Tests for significant differences between parameters under study were determined by the chi-square test. Consistency checks were performed on all tabulations. RESULTS Four hundred patients whose records were reviewed met the eligibility criteria for this study. One hundred ninety-seven were from academic facilities and 203 were from nonacademic facilities. Patient and tumor characteristics are shown in Table 2. The median age was 66.7 years. More than three-quarters of the patients were male. Eighty-eight percent had a Karnofsky performance status greater than or equal to 80. Sixtytwo percent of the patients had squamous cell carcinoma, and only 15% had clinical Stage I malignancies. Methods of tumor evaluation are shown in Table 3. Over two-thirds of patients had an esophagogram and nearly all had an endoscopy. Endoscopy was not performed for 6% of patients, probably due to lack of documentation, because all patients had histologic proof of malignancy.) On the other hand, endoscopic ultrasound was performed for only 3.5% of patients. There was a significant difference in the rates of performance of endoscopic ultrasound at academic and nonacademic institutions (6% vs. 1%). In addition, there was a significantly higher rate of computed tomography (CT) scanning of the chest at academic and nonacademic institutions (92% vs. 80%). Table 4 lists the major treatments given. Threequarters of the patients received chemotherapy and radiation therapy alone (54%), preoperatively (13.25%), or postoperatively (7.75%). One-quarter of patients 2502 CANCER June 15, 1999 / Volume 85 / Number 12 TABLE 2 Patient and Tumor Characteristics of 400 Patients with Esophageal Carcinoma Characteristic No. of patients Total no. Age (yrs) Range Median Gender Male Female KPS #70 80 90 100 Unknown Histology Squamous Adenocarcinoma Unknown Clinical stage I II III Unknown Race White Hispanic African American Asian Native American Unknown 400 TABLE 3 Percentage of Patients Undergoing Specified Evaluative Procedures Total A Non-A 69% 94% 3.5% 86% 75% 69% 94% 6% 92% 80% 68% 93% 1% 80% 70% (%) Esophagogram Endoscopy Endoscopic USa CT scan, chestb CT scan, abdomen 26–89 66.7 306 94 (76.5%) (23.5%) 38 165 176 12 9 (9.5%) (41.3%) (44%) (3%) (2.3%) 246 147 7 (61.5%) (36.8%) (1.8%) 60 158 118 64 (15%) (39.5%) (29.5%) (16%) 290 8 81 14 1 1 (73.8%) (2%) (20.3%) (3.5%) (0.3%) (0.3%) KPS: Karnofsky performance status. received radiation alone (20.25%), preoperatively (1.25%), or postoperatively (3.5%). Chemotherapy and radiation alone was the most frequently used regimen for both squamous cell and adenocarcinoma and for all stages. Trimodality treatment was more frequently used for patients with adenocarcinoma than for those with squamous cell carcinoma (34% vs. 14%). Only 29 patients (7%) were treated on an investigational protocol, including 15 on Institutional Review Board–approved institutional trials and 5 on NCI-sponsored group trials. Further information on chemotherapy administration is shown in Table 5. A significant difference in the rate of administration of chemotherapy between academic and nonacademic institutions was not apparent. Sixty-three percent of patients received 5-fluorouracil, 48% received cisplatin, and 6.8% received mitomycin. Academic institutions used cisplatin more frequently than nonacademic institutions (57% vs. 40%, P 5 0.001); however, nonacademic institutions A: academic; non-A: nonacademic; US: ultrasound; CT: computed tomography. a P 5 0.02 (A vs. Non-A). b P 5 0.002 (A vs. Non-A). used mitomycin more frequently (11% vs. 3%, P 5 0.001). CRT was given to 63% of patients, as detailed in Table 6. Academic institutions administered CRT significantly more often than nonacademic (67% vs. 58%, P 5 0.046). However, the rate at which CRT alone (without surgery) was given did not differ significantly between academic and nonacademic institutions (46% vs. 45%). Overall, preoperative CRT was given to 12% of patients, whereas postoperative CRT was given to 5%. Academic institutions used preoperative CRT more frequently than nonacademic institutions (18% vs. 7%), whereas postoperative CRT was used infrequently at both academic and nonacademic centers (3% vs. 6%). The most frequent dose range used was $50 gray (Gy) for those who received CRT alone as well as for the postoperative CRT group, whereas the most frequent dose range for the preoperative CRT group was 40 Gy to ,50 Gy. The parameters of radiation therapy are listed in Table 7. Eighty-three percent of patients were treated in a supine position. The median dose per fraction was 1.8 Gy and the median total dose was 50.4 Gy. The most frequent dose-per-fraction range was 1.9 –2.0 Gy, whereas the most frequent total dose range was 50 Gy to ,60 Gy. The median overall treatment duration was 40.5 days. Brachytherapy was given to 34 patients (8.5%); nearly all patients received brachytherapy in combination with external beam. High-dose-rate brachytherapy (HDR) was given more frequently than low-dose-rate brachytherapy (7.3% vs. 1.2%). Furthermore, nonacademic institutions used HDR brachytherapy more frequently than academic institutions (5.5% vs. 1.75%, P 5 0.002). The median HDR dose was 10 Gy and the median number of applications was 2. CONCLUSIONS This study establishes the national benchmarks for the evaluation and treatment of esophageal carcinoma at radiation facilities in the U.S. As it is based on infor- Radiation for Esophageal Carcinoma/Coia et al. 2503 TABLE 4 Treatments Used in the Study Overall and by Stage and Histology No. of patients Chemo and RT (total) Alone With surgery Pre-op Post-op RT (total) Alone With surgery Pre-op Post-op % Overall Squamous Adeno Stage I Stage II Stage III 75.0% 54.0% 21.0% 13.25% 7.75% 25.0% 20.25% 4.75% 1.25% 3.5% 300 216 84 53 31 100 81 19 5 14 176 142 34 25 9 70 59 11 3 8 120 70 50 28 22 27 19 8 2 6 44 33 11 10 1 16 13 3 0 3 126 88 38 28 10 32 30 2 1 1 93 67 26 11 15 25 18 7 2 5 RT: radiotherapy; Adeno: adenocarcinoma. TABLE 5 Frequency and Type of Chemotherapy Usage TABLE 6 Timing of Concurrent Chemotherapy and Radiation Therapy and Dose Ranges of Radiation Used % of patients Total Chemotherapy given Agents used 5-FU Cisplatina Mitomycina A Total A Non-A 63% 46% 4% 9% 33% 12% 3% 5.5% 3.8% 5% 0.3% 2% 3% 67% 46% 58% 45% 18% 7% 3% 6% Non-A 75% 76% 74% 63% 48% 6.8% 65% 57% 3% 62% 40% 11% A: academic; Non-A: nonacademic; 5-FU: 5-fluorouracil. a P 5 0.001. mation available from radiation therapy charts and hospital charts, it may occasionally underestimate procedure performance. Nonetheless, the study indicates that such evaluation and treatment are similar at academic and nonacademic facilities. It also suggests early penetrance of clinical trial results into national practice, as CRT was used more frequently than radiation alone. Regarding pretherapy evaluation, this study indicates that endoscopic ultrasound was infrequently used. The current AJCC T classification is based on the depth of esophageal wall penetration by the tumor. Only an initial surgical resection (i.e., esophagectomy) or endoscopic ultrasound can determine this depth. Patients not undergoing surgery cannot be completely staged according to the 1997 AJCC system without the use of endoscopic ultrasound. Because endoscopic ultrasound is infrequently used, and because the depth of wall penetration has not been shown to be of prognostic importance in the management of patients with initial chemoradiation, a modification of the current AJCC staging system to take into account readily Concurrent chemo and XRTa No surgery ,40 Gy 40–,50 Gy $50 Gy Pre-op ,40 Gy 40–,50 Gy $50 Gy Post-op ,40 Gy 40–,50 Gy $50 Gy A: academic; Non-A: nonacademic; XRT: radiation therapy; Gy: gray. a P 5 0.046. available and prognostically important tumor measures, such as length and degree of circumferential involvement, seems justified. Esophagography, which remains an important indicator of tumor length and location for the radiation oncologist, was used to evaluate over two-thirds of the patients. The less frequent use of esophagography than of CT of the thorax may reflect the increased reliance of the radiation oncologist on the CT scan for tumor localization. It should be also noted that the CT scan has been demonstrated to be less accurate in determining tumor length than esophagography or esophagoscopy; however, simulation is successful for most patients with the aid of a barium swallow.12 While not a diagnostic substitute for esophagography, the use of barium at simulation 2504 CANCER June 15, 1999 / Volume 85 / Number 12 TABLE 7 Radiation Parameters Parameters RT given External beam only Brachytherapy only Both Machine energy 6 MV 10 MV 18 MV Other Unknown Patient position Supine Prone Both Unknown Total dose Range #30 Gy .30–,40 Gy 40–,50 Gy 50–,60 Gy $60 Gy Total dose Median Dose/FX Range #1.8 Gy 1.9–2.0 Gy .2.0 Gy–,3.0 Gy $3.0 Gy Median Treatment duration Median Brachytherapy Overalla HDRa Median dose (Range) Median no. of applications (Range) LDR TABLE 8 Comparison of Treatments Delivered in the PCS and ACS Studies No. of patients (%) 366 3 31 (91.5%) (0.8%) (7.8%) 139 87 79 88 4 (35.0%) (22.0%) (20.0%) (22.0%) (1.0%) 330 38 8 21 (83.0%) (10.0%) (2.0%) (5.0%) Treatment ACS (%) ACS(M) (%) PCS (%) Surgery alone RT alone CT alone SX 1 RT SX 1 CT RT 1 CT All three No Rx Unknown 18 13 7 4 3 30 9 14 3 — 23 — 7.5 — 55 16 — — — 20 — 5 — 54 21 — — ACS: American Cancer Society; ACS(M): ACS data limited to those receiving radiation; PCS: Patterns of Care Study; RT: radiation therapy; CT: chemotherapy; SX: surgery; RX: treatment. 20 42 98 130 107 (5.0%) (10.5%) (24.5%) (32.5%) (26.8%) 50.4 Gy (Range 5 6.0–96.7 Gy) 237 121 9 26 1.8 Gy (59.3%) (30.3%) (2.3%) (6.5%) (Range 5 1.25–6.5 Gy) 40.5 days (Range 5 1–123 days) No. of patients (%) A Non-A 34 (8.5%) 29 (7.3%) 10 Gy (10–26 Gy 2 (2–4) 5 (1.2%) 8 (4%) 7 (3.5%) 15 Gy (10–26 Gy) 2 (2–3) 1 (0.5%) 26 (12.8%) 22 (11%) 10 Gy (5–25 Gy) 2 2–4) 4 (2%) A: academic; Non-A: nonacademic; RT: radiotherapy; MV: megavolts; Gy: gray; HDR: high-dose-rate brachytherapy; LDR: low-dose-rate brachytherapy. a P 5 0.002. (not accounted for in this study) is an effective tumor localization procedure. Regarding management, this study demonstrated that chemotherapy and radiation is the most frequently used regimen in the treatment of esophageal carcinoma. Fifty-four percent of all patients received chemotherapy and radiation; 21% received radiation alone; 20% had surgery, radiation, and chemotherapy; and 5% had surgery and radiation. Chemotherapy and radiation was the most frequently used treatment regardless of stage or histology. It is interesting to compare results of this PCS study with those of the American College of Surgeons Commission on Cancer and the American Cancer Society (ACS), as reported by Daly et al.13 (Table 8). They analyzed 5256 cases of esophageal carcinoma treated in 1993 and accrued to the National Cancer Data Base. These cases represented nearly half of the patients with esophageal carcinoma treated in the U.S. in 1993. The findings of the study of Daly et al. are remarkably similar to those of the PCS study presented herein. Of 8 possible treatment regimens listed in Table 8, ranging from surgery alone to trimodality treatment, they also found that the combination of radiation therapy and chemotherapy was the most frequently used treatment. A comparison of the percentage of patients treated with radiation alone, surgery and radiation, radiation and chemotherapy, or all three modalities in the two studies can be made by eliminating patients who were not treated with radiation in the study of Daly et al. In both studies, approximately 20% of patients received radiation alone, less than 10% received surgery plus radiation, more than half received radiation plus chemotherapy, and roughly 20% received trimodality treatment. The remarkable similarity in the results attest to the power of the methodology of this PCS study, which established the treatment usage frequency from a much more limited sample size (in this case, 400 patients vs. over 5000 patients.) CRT should be used more frequently than radiation alone, as the Herskovic et al. trial demonstrated superior results with CRT for the great majority of patients with esophageal carcinoma. That study involved patients with Stages I–III esophageal carci- Radiation for Esophageal Carcinoma/Coia et al. noma and included both squamous cell carcinoma and adenocarcinoma histologies (although only 15% of patients had adenocarcinoma). Evidence of the sustained benefit of CRT was published by Al-Sarraf et al. who reported long term results of the Intergroup trial.14 However, the appropriate relative frequency of the use of CRT to radiation alone is unknown, as some patients may not be candidates for chemotherapy due to medical condition, performance status, or other factors. This PCS study confirmed that CRT alone is used more frequently than radiation alone (45% vs. 20%). Furthermore, there was no difference in the frequency of use of CRT alone between academic and nonacademic facilities. PCS plans to restudy the issue to determine whether the relative use of CRT over radiation alone will change over time. Overall, CRT (alone, preoperative, or postoperative) was given somewhat more frequently at academic than nonacademic facilities (67% vs. 58%), predominantly due to its more frequent use at academic institutions in the preoperative setting. In particular, trimodality treatment was used more than twice as frequently for patients with adenocarcinoma than for those with squamous cell carcinoma. The use of preoperative CRT has been shown to be superior to surgery alone in two randomized studies; in one study only patients with adenocarcinoma were included,15 and in the other approximately three-fourths of patients had adenocarcinoma.16 Nonetheless, the value of preoperative CRT compared with surgery alone remains controversial and is being investigated further in a current Intergroup study. This study successfully determined the national practice patterns for patients with esophageal carcinoma. Changes in the evaluation and treatment of esophageal carcinoma patients continue, and PCS plans to examine these changes with a future process survey of patients with esophageal carcinoma treated in 1996 –1997. The current analysis will also serve as a useful benchmark to compare with evaluation and treatment given in other countries, particularly Japan, where the histologic presentation and management of disease differ substantially from that observed in the U.S. Such studies will give us further insight into how we may better evaluate and treat patients with esophageal carcinoma. 2505 REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Coia LR, Hanks GE. Quality Assessment in the USA: How the Patterns of Care Study has made a difference. Semin Radiat Oncol 1997;7:146 –56. Landis SH, Murray T, Bolden S, Wingo P. Cancer statistics, 1998. CA Cancer J Clin 1998;48:6 –29. Blot WJ, DeVesa SS, Fraumeni JF Jr. Continuing climbing rates of esophageal adenocarcinoma: an update. JAMA 1993; 270:1320. Blot WJ. Esophageal cancer trends and risk factors. Semin Oncol 1994;21:403–10. Coia LR, Engstrom P, Paul A. Nonsurgical management of esophageal cancer: report of a study of combined radiotherapy and chemotherapy. J Clin Oncol 1987;5:1783–90. Keane TJ, Harwood AR, Elhakim T, Rider WD, Cummings BJ, Ginsberg RJ. Radical radiation therapy with 5-fluorouracil infusion and mitomycin C for oesophageal squamous carcinoma. Radiother Oncol 1985;4:205–10. Chan A, Wong A, Arthur K. Concomitant 5-FU infusion, mitomycin-C and radical radiation therapy in esophageal squamous cell carcinoma. Int J Radiat Oncol Biol Phys 1989; 16:59 – 65. Herskovic A, Martz K, al-Sarraf M, Leichman L, Brindle J, Vaitkevicius V, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992;326:1593– 8. PCS Consensus Committee. 1996 decision trees and management guidelines. Semin Radiat Oncol 1997;7:163– 81. Sedransk N, Sedransk J. Distinguishing among distributions using data from complex sample design. J Am Stat Assoc 1979;74:754 – 60. Owen JB, Sedransk J, Pajak TF. National averages for process and outcome in radiation oncology: methodology of the Patterns of Care Study. Semin Radiat Oncol 1997;7:101–7. Lefor A, Merino M, Steinberg S. Computerized tomographic prediction of extraluminal spread and prognostic implications of lesion width in esophageal cancer. Cancer 1988;62: 1277–92. Daly JM, Karnell LH, Menck HR. National Cancer Data Base report on esophageal cancer. Cancer 1996;78:1820 – 8. al-Sarraf M, Martz K, Herskovic A, Leichman L, Brindle JS, Vaitkevicius VK, et al. Progress report of combined chemoradiotherapy versus radiotherapy alone in patients with esophageal cancer: an Intergroup study. J Clin Oncol 1997; 15:277– 84. Walsh TN, Noonan N, Hollywood D, Kelly A, Keeling N, Hennessy TP. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996; 335:462–7. Urba S, Orringer M, Turrisi A, Whyte R, Iannettoni M, Forastiere A. A randomized trial comparing transhiatal esophagectomy to preoperative concurrent chemoradiation followed by esophagectomy in local-regional esophageal cancer. Proc Am Soc Clin Oncol 1997;16:276a.