- Email: [email protected]
A meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for resectable esophageal cancer
The American Journal of Surgery 185 (2003) 538 –543
Review
A meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for resectable esophageal cancer John D. Urschel, M.D.a,b,*, Hari Vasan, B.Sc.a b
a Department of Surgery, McMaster University, Hamilton, Ontario, Canada Section of Thoracic Surgery, Beth Israel Deaconess Medical Center, 110 Francis St., Suite 2A, Boston, MA 02215, USA
Manuscript received November 19, 2002; revised manuscript December 20, 2002
Abstract Background: Esophagectomy is a standard treatment for resectable esophageal cancer but relatively few patients are cured. Combining neoadjuvant chemoradiation with surgery may improve survival but treatment morbidity is a concern. We performed a meta-analysis of randomized controlled trials (RCTs) that compared the use of neoadjuvant chemoradiation and surgery with the use of surgery alone for esophageal cancer. Methods: Medline and manual searches were done to identify all published RCTs that compared neoadjuvant chemoradiation and surgery with surgery alone for esophageal cancer. A random-effects model was used and the odds ratio (OR) was the principal measure of effect. Systematic quantitative review was done for outcomes unique to the neoadjuvant chemoradiation treatment group, such as pathological complete response. Results: Nine RCTs that included 1,116 patients were selected with quality scores ranging from 1 to 3 (5-point Jadad scale). Odds ratio (95% confidence interval [CI]; P value), expressed as chemoradiation and surgery versus surgery alone (treatment versus control; values ⬍1 favor chemoradiation-surgery arm), was 0.79 (0.59, 1.06; P ⫽ 0.12) for 1-year survival, 0.77 (0.56, 1.05; P ⫽ 0.10) for 2-year survival, 0.66 (0.47, 0.92; P ⫽ 0.016) for 3-year survival, 2.50 (1.05, 5.96; P ⫽ 0.038) for rate of resection, 0.53 (0.33, 0.84; P ⫽ 0.007) for rate of complete resection, 1.72 (0.96, 3.07; P ⫽ 0.07) for operative mortality, 1.63 (0.99, 2.68; P ⫽ 0.053) for all treatment mortality, 0.38 (0.23, 0.63; P ⫽ 0.0002) for local-regional cancer recurrence, 0.88 (0.55, 1.41; P ⫽ 0.60) for distant cancer recurrence, and 0.47 (0.16, 1.45; P ⫽ 0.19) for all cancer recurrence. A complete pathological response to chemoradiation occurred in 21% of patients. The 3-year survival benefit was most pronounced when chemotherapy and radiotherapy were given concurrently (OR 0.45, 95% CI 0.26 to 0.79, P ⫽ 0.005) instead of sequentially (OR 0.82, 95% CI 0.54 to 1.25, P ⫽ 0.36). Conclusions: Compared with surgery alone, neoadjuvant chemoradiation and surgery improved 3-year survival and reduced local-regional cancer recurrence. It was associated with a lower rate of esophageal resection, but a higher rate of complete (R0) resection. There was a nonsignificant trend toward increased treatment mortality with neoadjuvant chemoradiation. Concurrent administration of neoadjuvant chemotherapy and radiotherapy was superior to sequential chemoradiation treatment scheduling. © 2003 Excerpta Medica, Inc. All rights reserved. Keywords: Esophageal neoplasms/surgery; Esophageal neoplasms/radiotherapy; Antineoplastic agents; Postoperative complications; Prospective studies; Randomized controlled trial; Meta-analysis
Esophagectomy remains a standard treatment for patients with resectable esophageal cancer, and long-term survival exceeds 20% in some surgical series [1–3]. Surgical results have improved in recent decades but most of this improvement can be attributed to advances in preoperative staging, * Corresponding author. Tel.: ⫹1-617-632-7373; fax: ⫹1-617-6327562. E-mail address: [email protected]
patient selection, and postoperative care, as opposed to the effectiveness of surgical therapy itself [3–5]. Most patients with seemingly resectable esophageal cancer have little prospect for cure. The proximity of the esophagus to vital mediastinal structures often compromises the completeness of cancer resection, and micrometastatic systemic disease is often present at the time of initial cancer diagnosis. These two limitations of surgical therapy set the stage for cancer recurrence, both local-regional and systemic. Radiotherapy
0002-9610/03/$ – see front matter © 2003 Excerpta Medica, Inc. All rights reserved. doi:10.1016/S0002-9610(03)00066-7
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
can control local-regional esophageal cancer and chemotherapy, usually cisplatin and 5-fluorouracil, has both local and systemic antineoplastic activity. Combining chemotherapy, radiotherapy, and surgery could improve survival in patients with resectable esophageal cancer [6 – 8]. In theory, neoadjuvant (induction) chemoradiation offers early treatment of micrometastatic disease, and it can facilitate surgical resection by downstaging cancers [6 – 8]. In addition, esophageal cancer patients seem to tolerate preoperative (neoadjuvant) chemoradiation better than postoperative (adjuvant) chemoradiation. Based on these premises many phase II trials of neoadjuvant chemoradiation followed by surgery have been done. Although many trials have generated promising results, there is a lingering concern, especially among surgeons, that neoadjuvant chemoradiation may cause an unacceptable increase in perioperative morbidity and mortality. In addition, it could be harmful by delaying definitive, albeit modestly effective, treatment with surgery. Randomized controlled trials (RCTs) have been performed to address these issues, but the results are not consistent. Many of the RCTs enrolled small numbers of patients; this limits their ability to detect a treatment benefit, even if a benefit exists (inadequate power). Meta-analysis can be useful in this situation. We performed a meta-analysis of randomized controlled trials that compared chemoradiation and surgery to surgery alone in patients with resectable esophageal cancer.
Methods Medline and manual searches were done (completed independently and in duplicate) to identify all published (manuscripts and abstracts) randomized controlled trials (RCTs) that compared neoadjuvant chemoradiation and surgery with surgery alone for resectable esophageal cancer. Trials were not excluded because of cancer histology (squamous or adenocarcinoma) or language of publication. The Medline search was done on PubMed (available at: http:// www.ncbi.nlm.nih.gov/entrez/query.fcgi). A set was created using the terms “esophageal neoplasms/surgery OR esophagectomy OR oesophagectomy OR esophageal cancer OR oesophageal cancer.” This yielded 21,523 citations (November 15, 2002). Another set was created using the terms “antineoplastic agents OR chemotherapy OR radiotherapy.” This yielded 1,455,501 citations. The two sets were combined using the Boolean operator “AND” to give 4,890 documents on chemotherapy, radiotherapy, and surgery for esophageal cancer. This set was limited to “randomized controlled trial” to yield 193 documents. These documents were reviewed to identify RCTs that compared neoadjuvant chemoradiation and surgery with surgery alone. Six citations were identified and retrieved [9 –14]. Manual searches were done by reviewing articles and abstracts cited in the reference lists of identified RCTs and also by reviewing the first author’s article and abstract file. This identified three
539
other trials that were published, to this point, in abstract form only [15–17]. We did not attempt to identify unpublished RCTs. In total, nine RCTs were found and these trials form the basis of the meta-analysis. Given the limited number of RCTs we designed the article selection process to be inclusive as opposed to exclusive. In other words, trials were not excluded because of trial quality (design), language, or insufficient number of patients. Nevertheless, trial validity assessment was done independently and in duplicate, and a trial quality score was assigned (scale of 1 to 5) according to the method of Jadad [18]. If reviewers disagreed on the quality scores, discrepancies were identified and a consensus was reached. Trial data abstraction was also done independently and in duplicate, but abstractors were not blinded to the trials’ authors or institution. The reviewers were familiar with the trials, and there were only nine trials, so blinding the abstracts was not practical. Any discrepancies in data abstraction were examined further and resolved by consensus. Outcomes assessed by meta-analysis included 1-year survival, 2-year survival, 3-year survival, rate of resection, rate of complete (R0) resection, operative mortality, anastomotic leaks, postoperative pulmonary complications, all treatment mortality, local-regional cancer recurrence, distant cancer recurrence, and all cancer recurrence. The intention to treat principle was used when calculating frequency of events, other than postoperative events (operative mortality, anastomotic leaks, postoperative pulmonary complications). For postoperative events, we used the actual number of patients undergoing surgery as the denominator because intention to treat analysis artificially lowered the incidence of adverse postoperative outcomes. Survival data was obtained from individual trials using the most reliable data available. Raw data were considered the most reliable data, followed by survival percentages, and derivation of survival from graphically presented survival curves. Resection was defined as any resection, curative or palliative; esophageal bypass and exploratory surgery were not included. Complete resection was defined as a microscopically complete (R0) resection performed with curative intent. Most of the trials expressed operative mortality as 30-day mortality, as opposed to in-hospital mortality, so 30-day mortality was used for data analysis. Anastomotic leaks were defined as clinical or radiological leaks. Postoperative pulmonary complications included pneumonia and respiratory failure. All treatment mortality was obtained by adding preoperative deaths (usually secondary to chemoradiation) and postoperative deaths. The most complete summation of these deaths was used from each individual trial. Local-regional cancer recurrence was defined as any localregional recurrence, as opposed to isolated local-regional recurrence. Similarly, distant cancer recurrence was defined as any distant recurrence. All cancer recurrence was defined as any type (local, regional, distant), or combination of types, of cancer recurrence. Sensitivity analyses were performed on the 3-year survival data to identify the impor-
540
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
tance of cancer histology (squamous or adenocarcinoma) and scheduling of chemoradiation (concurrent or sequential) on survival. Given the obvious heterogeneity of the trials (general patient characteristics, cancer stage, cancer histology, chemotherapeutic agents, radiation dose and scheduling, and type of surgical approach) we did not consider a fixed effects meta-analysis model to be methodologically sound, even if a test of heterogeneity was acceptable for several outcomes. Instead, a random effects model was used. This gives conservative confidence intervals and minimizes the risk of erroneously assigning benefit to the treatment group, if no benefit really exists. Odds ratio was the principal measure of effect. Odds ratios (OR) are presented as a point estimate with 95% confidence intervals (CI) and P values in parentheses. Odds ratios are calculated as treatment (chemoradiation and surgery) versus control (surgery alone), so a number less than one favors the treatment group (lower frequency of undesirable events). For resection, and complete resection, we used the reciprocal of these events (lack of resection and lack of a complete resection) to maintain consistency in presentation and direction of effect. Funnel plot analysis did not suggest publication bias against negative trials [19]. Biostat (Englewood, New Jersey) statistical software was used. Systematic quantitative review was used for outcomes that could not be accurately synthesized with traditional meta-analysis methods. These outcomes included pathological complete response to chemoradiation, and death secondary to chemoradiation (events only possible in the chemoradiation-surgery group). Crude pooling of data was done for these chemotherapy outcomes. Median survival data could not be combined using meta-analysis methods because there was insufficient documentation of original patient data in many trials.
Results The two trial assessors agreed on the selection of nine RCTs [9 –17]. Combining these trials yielded data on 1,116 patients. The RCT quality scores ranged from 1 to 3 (5-point scale), with a mean of 2.1. The quality scores were artificially low because of the importance placed on blinding in the scoring system, and the inherent difficulty in blinding a treatment such as chemoradiation [18,20]. Survival of the two patient groups was similar at one and two years, but 3-year survival in the neoadjuvant chemoradiation and surgery group was superior to that seen with surgery alone (Fig. 1). Odds ratio (95% CI; P value), expressed as chemoradiation and surgery versus surgery alone (treatment versus control; values ⬍1 favor chemoradiationsurgery arm), was 0.79 (0.59, 1.06; P ⫽ 0.12) for 1-year survival, 0.77 (0.56, 1.05; P ⫽ 0.10) for 2-year survival, and 0.66 (0.47, 0.92; P ⫽ 0.016) for 3-year survival. Patients treated with surgery alone were more likely to undergo an
Fig. 1. Three-year survival (odds ratio 0.66, 95% confidence interval 0.47 to 0.92; P ⫽ 0.016). (AC ⫽ adenocarcinoma; SCC ⫽ squamous cell carcinoma; CTRTS ⫽ chemotherapy, radiation therapy, surgery; S ⫽ surgery.)
esophageal resection than those treated with chemoradiation and surgery (OR 2.50, 95% CI 1.05 to 5.96; P ⫽ 0.038; Fig. 2). However, patients treated with chemoradiation and surgery had a higher rate of complete (R0) resection than those treated with surgery alone (OR 0.53, 95% CI 0.33 to 0.84; P ⫽ 0.007; Fig. 3). Three-year survival meta-analysis was repeated with RCTs separated according to histology (squamous or adenocarcinoma) and chemoradiation scheduling (concurrent or sequential). This admittedly created small sets of trials (3 to 5) for analysis. If only RCTs addressing squamous cancer were considered, the 3-year survival advantage of neoadjuvant chemoradiation and surgery was less apparent (OR 0.75, 95% CI 0.52 to 1.09; P ⫽ 0.13). Restricting the analysis to RCTs of adenocarcinoma was not feasible since there was only one trial of this type [12]. If meta-analysis was restricted to RCTs using concurrent chemoradiation, 3-year survival strongly favored the combination of neoadjuvant chemoradiation and surgery (OR 0.45, 95% CI 0.26 to 0.79; P ⫽ 0.005). Conversely, RCTs using sequential chemoradiation did not demonstrate a survival benefit at 3 years (OR 0.82, 95% CI 0.54 to 1.25; P ⫽ 0.36).
Fig. 2. Rate of resection (odds ratio 2.50, 95% confidence interval 1.05 to 5.96; P ⫽ 0.038). (AC ⫽ adenocarcinoma; CTRTS ⫽ chemotherapy, radiation therapy, surgery; S ⫽ surgery.)
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
Fig. 3. Rate of complete resection (odds ratio 0.53, 95% confidence interval 0.33 to 0.84; P ⫽ 0.007). (AC ⫽ adenocarcinoma; SCC ⫽ squamous cell carcinoma; CTRTS ⫽ chemotherapy, radiation therapy, surgery; S ⫽ surgery.)
Data on patterns of cancer recurrence were provided in only three of nine RCTs. Patients treated with neoadjuvant chemoradiation and surgery had fewer local-regional cancer recurrences (OR 0.38, 95% CI 0.23 to 0.63; P ⫽ 0.0002). Distant recurrence (OR 0.88, 95% CI 0.55 to 1.41; P ⫽ 0.60) and all cancer recurrence (OR 0.47, 95% CI 0.16 to 1.45; P ⫽ 0.19) were similar in the two groups of patients. The rate of adverse treatment events was not significantly different in the two patient groups, but there was a trend in favor of surgery alone for both operative mortality and all treatment mortality. Odds ratio (95% CI; P value), expressed as chemoradiation and surgery versus surgery alone (values ⬍1 favor chemoradiation-surgery arm), was 1.72 (0.96, 3.07; P ⫽ 0.07) for operative mortality, 1.29 (0.66, 2.55; P ⫽ 0.46) for anastomotic leaks, 1.21 (0.65, 2.29; P ⫽ 0.55) for postoperative pulmonary complications, and 1.63 (0.99, 2.68; P ⫽ 0.053) for all treatment mortality. Pooled data analysis for the chemoradiation treated patients showed a complete pathological response in 21% of patients. Chemoradiation mortality (before surgery) was 1.2%. Data on the extent and quality of surgical resection, and lymphadenectomy, was difficult to discern from the published articles. One trial featured transhiatal esophagectomy exclusively [14]. The majority of patients treated in the other RCTs underwent transthoracic esophagectomies.
Comments Surgeons and oncologists remain disappointed with surgery as a solitary treatment modality for esophageal cancer. Not surprisingly, various combinations of chemotherapy, radiotherapy, and surgery (multimodality treatment) have been investigated [6 – 8]. Both neoadjuvant radiotherapy and surgery, and surgery and adjuvant radiotherapy, have been studied. Randomized controlled trials and meta-analyses of trials have not shown a significant survival advantage for these combinations of surgery and radiotherapy [21–24]. Surgery followed by adjuvant chemotherapy is a common treatment approach for many solid tumors, but
541
postesophagectomy chemotherapy has not been extensively studied; one randomized trial showed no benefit for surgery followed by chemotherapy [25]. Several randomized controlled trials have compared neoadjuvant chemotherapy and surgery to surgery alone [26 –30]. One large trial showed improved survival with neoadjuvant chemotherapy [26], but another large trial and a meta-analysis of all trials showed little real benefit [27,30]. The rate of pathological complete response was low. Current neoadjuvant chemotherapy regimes are probably not effective enough to improve survival [27,30]. The neoadjuvant combination of chemotherapy and radiation therapy, as opposed to neoadjuvant chemotherapy, has brought about dramatic clinical and pathological responses in nonrandomized esophageal cancer trials [6 – 8,31]. It is an intuitively appealing treatment strategy. Micrometastatic disease is treated early, treatment is delivered while tumor blood supply is intact, and neoadjuvant treatment can facilitate resection by downstaging tumors [6 – 8]. Chemotherapy and radiation therapy act synergistically at the local-regional cancer level. However, chemoradiation therapy can increase the morbidity and mortality of esophagectomy. Surgeons are particularly concerned about the technical difficulties of operating in a pretreated field, wound (anastomotic) healing problems, and postoperative pulmonary complications (especially acute respiratory distress syndrome) [32–35]. Finally, neoadjuvant chemoradiation may cause death independent of surgery (before surgery) [33]. Surgery followed by adjuvant chemotherapy and radiotherapy is used for some gastrointestinal malignancies, but esophageal cancer patients often lack the general strength to complete this postoperative treatment. Adjuvant chemoradiation after esophagectomy has not been investigated with randomized trials. Nevertheless, some centers have reported encouraging phase II data [36], and the therapeutic strategy is effective in gastric cancer [37]. Further consideration of this treatment approach for esophageal cancer is probably warranted. Our meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for esophageal cancer showed a survival benefit (3-year survival) for the combination of chemoradiation and surgery. The survival benefit is related, in part, to improved local-regional cancer control brought about by the neoadjuvant treatment. Neoadjuvant chemoradiation did not significantly reduce the incidence of cancer recurrence at distant sites. The sensitivity analysis of chemoradiation scheduling showed the importance of concurrent chemoradiation. Concurrent, as opposed to sequential, chemoradiation allows maximal antineoplastic synergy between chemotherapeutic agents and radiation treatment. The meta-analysis showed a marked difference in resection and complete resection rates for the two treatment groups. Patients treated with surgery alone were more likely to undergo esophageal resection, but those treated with
542
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
chemoradiation and surgery were more likely to have a complete (R0) resection performed. This suggests that neoadjuvant chemoradiation therapy downstages tumors and facilitates complete resection. The lower rate of local-regional cancer recurrence in the neoadjuvant group supports this contention. However, neoadjuvant therapy probably also acts as a selection tool for surgery [8,30]. Conceivably, patients with biologically aggressive tumors, who might undergo palliative resections if treated with surgery alone, may not respond to neoadjuvant chemoradiation. These patients may deteriorate during their neoadjuvant therapy and never have resection attempted. Our meta-analysis showed a nonsignificant trend toward increased operative mortality (OR 1.72, 95% CI 0.96 to 3.07, P ⫽ 0.07) and increased all treatment mortality (OR 1.63, 95% CI 0.99 to 2.68, P ⫽ 0.053) in the neoadjuvant group. There is little doubt that esophagectomy can be a technically challenging operation when performed after neoadjuvant chemoradiation treatment. Surgeons intuitively understand the general relationship between operative difficulty and postoperative complications. More specifically, radiation treatment can potentially set the stage for failures of wound healing (anastomotic leaks) and postoperative acute lung injury [8,32–35]. The risk of postoperative acute lung injury is related, in part, to the specifics of radiation therapy. Large radiation fractions and treatment of large volumes of lung tissue are undesirable. Both surgeons and oncologists are concerned that the survival benefits of neoadjuvant chemoradiation can be negated by an increase in postoperative deaths [33]. This was observed in the trial reported by Bosset [13]. In summary, this meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for esophageal cancer showed a survival benefit for neoadjuvant chemoradiation treatment. Compared with surgery alone, neoadjuvant chemoradiation and surgery improved 3-year survival and reduced localregional cancer recurrence. Neoadjuvant chemotherapy and radiotherapy was most effective when given concurrently. Neoadjuvant chemoradiation was associated with a lower rate of esophageal resection, but a higher rate of complete (R0) resection. There was a nonsignificant, but very concerning, trend toward increased treatment mortality with neoadjuvant chemoradiation.
References [1] Nigro JJ, DeMeester SR, Hagen JA, et al. Node status in transmural esophageal adenocarcinoma and outcome after en bloc esophagectomy. J Thorac Cardiovasc Surg 1999;117:960 – 8. [2] Lerut T, Coosemans W, De Leyn P, et al. Is there a role for radical esophagectomy? Eur J Cardiothorac Surg 1999;16(suppl 1):S44 –7. [3] Ando N, Ozawa S, Kitagawa Y, et al. Improvement in the results of surgical treatment of advanced squamous esophageal carcinoma during 15 consecutive years. Ann Surg 2000;232:225–32.
[4] Whooley BP, Law S, Murthy SC, et al. Analysis of reduced death and complication rates after esophageal resection. Ann Surg 2001;233: 338 – 44. [5] Orringer MB, Marshall B, Iannettoni MD. Transhiatal esophagectomy: clinical experience and refinements. Ann Surg 1999;230:392– 403. [6] Sutton P, Clark P. Neo-adjuvant treatment for oesophageal cancer. GI Cancer 2000;3:231– 8. [7] Lehnert T. Multimodal therapy for squamous carcinoma of the esophagus. Br J Surg 1999;86:727–39. [8] Geh JI, Crellin AM, Glynne-Jones R. Preoperative (neoadjuvant) chemoradiotherapy in oesophageal cancer. Br J Surg 2001;88:338 – 56. [9] Nygaard K, Hagen S, Hansen HS, et al. Pre-operative radiotherapy prolongs survival in operable esophageal carcinoma: a randomized, multicenter study of pre-operative radiotherapy and chemotherapy. The second Scandinavian trial in esophageal cancer. World J Surg 1992;16:1104 –10. [10] Apinop C, Puttisak P, Preecha N. A prospective study of combined therapy in esophageal cancer. Hepatogastroenterology 1994;41:391–3. [11] Le Prise E, Etienne PL, Meunier B, et al. A randomized study of chemotherapy, radiation therapy, and surgery versus surgery for localized squamous cell carcinoma of the esophagus. Cancer 1994;73: 1779 – 84. [12] Walsh TN, Noonan N, Hollywood D, et al. A comparison of multimodality therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996;335:462–7. [13] 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;337:161–7. [14] 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;19:305–13. [15] Law S, Kwong DLW, Tung HM, et al. Preoperative chemoradiation for squamous cell esophageal cancer: a prospective randomized trial [abstract]. Can J Gastroenterol 1998;12(suppl B):161. [16] Walsh TN, McDonnell CO, Mulligan ED, et al. Multimodal therapy versus surgery alone for squamous cell carcinoma of the esophagus: a prospective randomized trial [abstract]. Gastroenterology 2000; 118(suppl 2):1008. [17] Burmeister BH, Smithers BM, Fitzgerald L, et al. A randomized phase III trial of preoperative chemoradiation followed by surgery (CR-S) versus surgery alone (S) for localized resectable cancer of the esophagus [abstract]. Proc Am Soc Clin Oncol 2002;21:518. [18] Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1–12. [19] Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. QUOROM Group. Br J Surg 2000;87:1448 – 54. [20] Urschel JD, Goldsmith CH, Tandan VR, Miller JD. Users’ guide to evidence-based surgery: how to use an article evaluating surgical interventions. Can J Surg 2001;44:95–100. [21] Arnott SJ, Duncan W, Gignoux M, et al. Preoperative radiotherapy in esophageal carcinoma: a meta-analysis using individual patient data (Oesophageal Cancer Collaborative Group). Int J Radiat Oncol Biol Phys 1998;41:579 – 83. [22] Teniere P, Hay JM, Fingerhut A, Fagniez PL. Postoperative radiation therapy does not increase survival after curative resection for squamous cell carcinoma of the middle and lower esophagus as shown by a multicenter controlled trial. French University Association for Surgical Research. Surg Gynecol Obstet 1991;173:123–30. [23] Fok M, Sham JS, Choy D, et al. Postoperative radiotherapy for carcinoma of the esophagus: a prospective, randomized controlled study. Surgery 1993;113:138 – 47.
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543 [24] Zieren HU, Muller JM, Jacobi CA, et al. Adjuvant postoperative radiation therapy after curative resection of squamous cell carcinoma of the thoracic esophagus: a prospective randomized study. World J Surg 1995;19:444 –9. [25] Ando N, Iizuka T, Kakegawa T, et al. A randomized trial of surgery with and without chemotherapy for localized squamous carcinoma of the thoracic esophagus: the Japan Clinical Oncology Group study. J Thorac Cardiovasc Surg 1997;114:205–9. [26] Medical Research Council Oesophageal Cancer Working Group. Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. Lancet 2002;359:1727– 33. [27] Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 1998;339:1979 – 84. [28] Law S, Fok M, Chow S, et al. Preoperative chemotherapy versus surgical therapy alone for squamous cell carcinoma of the esophagus: a prospective randomized trial. J Thorac Cardiovasc Surg 1997;114: 210 –7. [29] Ancona E, Ruol A, Santi S, et al. Only pathologic complete response to neoadjuvant chemotherapy improves significantly the long term survival of patients with resectable esophageal squamous cell carcinoma: final report of a randomized, controlled trial of preoperative chemotherapy versus surgery alone. Cancer 2001;91:2165–74.
543
[30] Urschel JD, Vasan H, Blewett CJ. A meta-analysis of randomized controlled trials that compared neoadjuvant chemotherapy and surgery to surgery alone for resectable esophageal cancer. Am J Surg 2002;183:274 –9. [31] Leichman L, Steiger Z, Seydel HG, et al. Preoperative chemotherapy and radiation therapy for patients with cancer of the esophagus: a potentially curative approach. J Clin Oncol 1984;2:75–9. [32] Urschel JD. Esophagogastrostomy anastomotic leaks complicating esophagectomy—a review. Am J Surg 1995;169:634 – 40. [33] Keller SM, Ryan LM, Coia LR, et al. High dose chemoradiotherapy followed by esophagectomy for adenocarcinoma of the esophagus and gastroesophageal junction: results of a phase II study of the Eastern Cooperative Oncology Group. Cancer 1998;83:1908 –16. [34] Avendano CE, Flume PA, et al. Pulmonary complications after esophagectomy. Ann Thorac Surg 2002;73:922– 6. [35] Schilling MK, Eichenberger M, Maurer CA, et al. Ketoconazole and pulmonary failure after esophagectomy: a prospective clinical trial. Dis Esophagus 2001;14:37– 40. [36] Bedard EL, Inculet RI, Malthaner RA, et al. The role of surgery and postoperative chemoradiation therapy in patients with lymph node positive esophageal carcinoma. Cancer 2001;91:2423–30. [37] Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 2001;345:725– 30.
Review
A meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for resectable esophageal cancer John D. Urschel, M.D.a,b,*, Hari Vasan, B.Sc.a b
a Department of Surgery, McMaster University, Hamilton, Ontario, Canada Section of Thoracic Surgery, Beth Israel Deaconess Medical Center, 110 Francis St., Suite 2A, Boston, MA 02215, USA
Manuscript received November 19, 2002; revised manuscript December 20, 2002
Abstract Background: Esophagectomy is a standard treatment for resectable esophageal cancer but relatively few patients are cured. Combining neoadjuvant chemoradiation with surgery may improve survival but treatment morbidity is a concern. We performed a meta-analysis of randomized controlled trials (RCTs) that compared the use of neoadjuvant chemoradiation and surgery with the use of surgery alone for esophageal cancer. Methods: Medline and manual searches were done to identify all published RCTs that compared neoadjuvant chemoradiation and surgery with surgery alone for esophageal cancer. A random-effects model was used and the odds ratio (OR) was the principal measure of effect. Systematic quantitative review was done for outcomes unique to the neoadjuvant chemoradiation treatment group, such as pathological complete response. Results: Nine RCTs that included 1,116 patients were selected with quality scores ranging from 1 to 3 (5-point Jadad scale). Odds ratio (95% confidence interval [CI]; P value), expressed as chemoradiation and surgery versus surgery alone (treatment versus control; values ⬍1 favor chemoradiation-surgery arm), was 0.79 (0.59, 1.06; P ⫽ 0.12) for 1-year survival, 0.77 (0.56, 1.05; P ⫽ 0.10) for 2-year survival, 0.66 (0.47, 0.92; P ⫽ 0.016) for 3-year survival, 2.50 (1.05, 5.96; P ⫽ 0.038) for rate of resection, 0.53 (0.33, 0.84; P ⫽ 0.007) for rate of complete resection, 1.72 (0.96, 3.07; P ⫽ 0.07) for operative mortality, 1.63 (0.99, 2.68; P ⫽ 0.053) for all treatment mortality, 0.38 (0.23, 0.63; P ⫽ 0.0002) for local-regional cancer recurrence, 0.88 (0.55, 1.41; P ⫽ 0.60) for distant cancer recurrence, and 0.47 (0.16, 1.45; P ⫽ 0.19) for all cancer recurrence. A complete pathological response to chemoradiation occurred in 21% of patients. The 3-year survival benefit was most pronounced when chemotherapy and radiotherapy were given concurrently (OR 0.45, 95% CI 0.26 to 0.79, P ⫽ 0.005) instead of sequentially (OR 0.82, 95% CI 0.54 to 1.25, P ⫽ 0.36). Conclusions: Compared with surgery alone, neoadjuvant chemoradiation and surgery improved 3-year survival and reduced local-regional cancer recurrence. It was associated with a lower rate of esophageal resection, but a higher rate of complete (R0) resection. There was a nonsignificant trend toward increased treatment mortality with neoadjuvant chemoradiation. Concurrent administration of neoadjuvant chemotherapy and radiotherapy was superior to sequential chemoradiation treatment scheduling. © 2003 Excerpta Medica, Inc. All rights reserved. Keywords: Esophageal neoplasms/surgery; Esophageal neoplasms/radiotherapy; Antineoplastic agents; Postoperative complications; Prospective studies; Randomized controlled trial; Meta-analysis
Esophagectomy remains a standard treatment for patients with resectable esophageal cancer, and long-term survival exceeds 20% in some surgical series [1–3]. Surgical results have improved in recent decades but most of this improvement can be attributed to advances in preoperative staging, * Corresponding author. Tel.: ⫹1-617-632-7373; fax: ⫹1-617-6327562. E-mail address: [email protected]
patient selection, and postoperative care, as opposed to the effectiveness of surgical therapy itself [3–5]. Most patients with seemingly resectable esophageal cancer have little prospect for cure. The proximity of the esophagus to vital mediastinal structures often compromises the completeness of cancer resection, and micrometastatic systemic disease is often present at the time of initial cancer diagnosis. These two limitations of surgical therapy set the stage for cancer recurrence, both local-regional and systemic. Radiotherapy
0002-9610/03/$ – see front matter © 2003 Excerpta Medica, Inc. All rights reserved. doi:10.1016/S0002-9610(03)00066-7
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
can control local-regional esophageal cancer and chemotherapy, usually cisplatin and 5-fluorouracil, has both local and systemic antineoplastic activity. Combining chemotherapy, radiotherapy, and surgery could improve survival in patients with resectable esophageal cancer [6 – 8]. In theory, neoadjuvant (induction) chemoradiation offers early treatment of micrometastatic disease, and it can facilitate surgical resection by downstaging cancers [6 – 8]. In addition, esophageal cancer patients seem to tolerate preoperative (neoadjuvant) chemoradiation better than postoperative (adjuvant) chemoradiation. Based on these premises many phase II trials of neoadjuvant chemoradiation followed by surgery have been done. Although many trials have generated promising results, there is a lingering concern, especially among surgeons, that neoadjuvant chemoradiation may cause an unacceptable increase in perioperative morbidity and mortality. In addition, it could be harmful by delaying definitive, albeit modestly effective, treatment with surgery. Randomized controlled trials (RCTs) have been performed to address these issues, but the results are not consistent. Many of the RCTs enrolled small numbers of patients; this limits their ability to detect a treatment benefit, even if a benefit exists (inadequate power). Meta-analysis can be useful in this situation. We performed a meta-analysis of randomized controlled trials that compared chemoradiation and surgery to surgery alone in patients with resectable esophageal cancer.
Methods Medline and manual searches were done (completed independently and in duplicate) to identify all published (manuscripts and abstracts) randomized controlled trials (RCTs) that compared neoadjuvant chemoradiation and surgery with surgery alone for resectable esophageal cancer. Trials were not excluded because of cancer histology (squamous or adenocarcinoma) or language of publication. The Medline search was done on PubMed (available at: http:// www.ncbi.nlm.nih.gov/entrez/query.fcgi). A set was created using the terms “esophageal neoplasms/surgery OR esophagectomy OR oesophagectomy OR esophageal cancer OR oesophageal cancer.” This yielded 21,523 citations (November 15, 2002). Another set was created using the terms “antineoplastic agents OR chemotherapy OR radiotherapy.” This yielded 1,455,501 citations. The two sets were combined using the Boolean operator “AND” to give 4,890 documents on chemotherapy, radiotherapy, and surgery for esophageal cancer. This set was limited to “randomized controlled trial” to yield 193 documents. These documents were reviewed to identify RCTs that compared neoadjuvant chemoradiation and surgery with surgery alone. Six citations were identified and retrieved [9 –14]. Manual searches were done by reviewing articles and abstracts cited in the reference lists of identified RCTs and also by reviewing the first author’s article and abstract file. This identified three
539
other trials that were published, to this point, in abstract form only [15–17]. We did not attempt to identify unpublished RCTs. In total, nine RCTs were found and these trials form the basis of the meta-analysis. Given the limited number of RCTs we designed the article selection process to be inclusive as opposed to exclusive. In other words, trials were not excluded because of trial quality (design), language, or insufficient number of patients. Nevertheless, trial validity assessment was done independently and in duplicate, and a trial quality score was assigned (scale of 1 to 5) according to the method of Jadad [18]. If reviewers disagreed on the quality scores, discrepancies were identified and a consensus was reached. Trial data abstraction was also done independently and in duplicate, but abstractors were not blinded to the trials’ authors or institution. The reviewers were familiar with the trials, and there were only nine trials, so blinding the abstracts was not practical. Any discrepancies in data abstraction were examined further and resolved by consensus. Outcomes assessed by meta-analysis included 1-year survival, 2-year survival, 3-year survival, rate of resection, rate of complete (R0) resection, operative mortality, anastomotic leaks, postoperative pulmonary complications, all treatment mortality, local-regional cancer recurrence, distant cancer recurrence, and all cancer recurrence. The intention to treat principle was used when calculating frequency of events, other than postoperative events (operative mortality, anastomotic leaks, postoperative pulmonary complications). For postoperative events, we used the actual number of patients undergoing surgery as the denominator because intention to treat analysis artificially lowered the incidence of adverse postoperative outcomes. Survival data was obtained from individual trials using the most reliable data available. Raw data were considered the most reliable data, followed by survival percentages, and derivation of survival from graphically presented survival curves. Resection was defined as any resection, curative or palliative; esophageal bypass and exploratory surgery were not included. Complete resection was defined as a microscopically complete (R0) resection performed with curative intent. Most of the trials expressed operative mortality as 30-day mortality, as opposed to in-hospital mortality, so 30-day mortality was used for data analysis. Anastomotic leaks were defined as clinical or radiological leaks. Postoperative pulmonary complications included pneumonia and respiratory failure. All treatment mortality was obtained by adding preoperative deaths (usually secondary to chemoradiation) and postoperative deaths. The most complete summation of these deaths was used from each individual trial. Local-regional cancer recurrence was defined as any localregional recurrence, as opposed to isolated local-regional recurrence. Similarly, distant cancer recurrence was defined as any distant recurrence. All cancer recurrence was defined as any type (local, regional, distant), or combination of types, of cancer recurrence. Sensitivity analyses were performed on the 3-year survival data to identify the impor-
540
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
tance of cancer histology (squamous or adenocarcinoma) and scheduling of chemoradiation (concurrent or sequential) on survival. Given the obvious heterogeneity of the trials (general patient characteristics, cancer stage, cancer histology, chemotherapeutic agents, radiation dose and scheduling, and type of surgical approach) we did not consider a fixed effects meta-analysis model to be methodologically sound, even if a test of heterogeneity was acceptable for several outcomes. Instead, a random effects model was used. This gives conservative confidence intervals and minimizes the risk of erroneously assigning benefit to the treatment group, if no benefit really exists. Odds ratio was the principal measure of effect. Odds ratios (OR) are presented as a point estimate with 95% confidence intervals (CI) and P values in parentheses. Odds ratios are calculated as treatment (chemoradiation and surgery) versus control (surgery alone), so a number less than one favors the treatment group (lower frequency of undesirable events). For resection, and complete resection, we used the reciprocal of these events (lack of resection and lack of a complete resection) to maintain consistency in presentation and direction of effect. Funnel plot analysis did not suggest publication bias against negative trials [19]. Biostat (Englewood, New Jersey) statistical software was used. Systematic quantitative review was used for outcomes that could not be accurately synthesized with traditional meta-analysis methods. These outcomes included pathological complete response to chemoradiation, and death secondary to chemoradiation (events only possible in the chemoradiation-surgery group). Crude pooling of data was done for these chemotherapy outcomes. Median survival data could not be combined using meta-analysis methods because there was insufficient documentation of original patient data in many trials.
Results The two trial assessors agreed on the selection of nine RCTs [9 –17]. Combining these trials yielded data on 1,116 patients. The RCT quality scores ranged from 1 to 3 (5-point scale), with a mean of 2.1. The quality scores were artificially low because of the importance placed on blinding in the scoring system, and the inherent difficulty in blinding a treatment such as chemoradiation [18,20]. Survival of the two patient groups was similar at one and two years, but 3-year survival in the neoadjuvant chemoradiation and surgery group was superior to that seen with surgery alone (Fig. 1). Odds ratio (95% CI; P value), expressed as chemoradiation and surgery versus surgery alone (treatment versus control; values ⬍1 favor chemoradiationsurgery arm), was 0.79 (0.59, 1.06; P ⫽ 0.12) for 1-year survival, 0.77 (0.56, 1.05; P ⫽ 0.10) for 2-year survival, and 0.66 (0.47, 0.92; P ⫽ 0.016) for 3-year survival. Patients treated with surgery alone were more likely to undergo an
Fig. 1. Three-year survival (odds ratio 0.66, 95% confidence interval 0.47 to 0.92; P ⫽ 0.016). (AC ⫽ adenocarcinoma; SCC ⫽ squamous cell carcinoma; CTRTS ⫽ chemotherapy, radiation therapy, surgery; S ⫽ surgery.)
esophageal resection than those treated with chemoradiation and surgery (OR 2.50, 95% CI 1.05 to 5.96; P ⫽ 0.038; Fig. 2). However, patients treated with chemoradiation and surgery had a higher rate of complete (R0) resection than those treated with surgery alone (OR 0.53, 95% CI 0.33 to 0.84; P ⫽ 0.007; Fig. 3). Three-year survival meta-analysis was repeated with RCTs separated according to histology (squamous or adenocarcinoma) and chemoradiation scheduling (concurrent or sequential). This admittedly created small sets of trials (3 to 5) for analysis. If only RCTs addressing squamous cancer were considered, the 3-year survival advantage of neoadjuvant chemoradiation and surgery was less apparent (OR 0.75, 95% CI 0.52 to 1.09; P ⫽ 0.13). Restricting the analysis to RCTs of adenocarcinoma was not feasible since there was only one trial of this type [12]. If meta-analysis was restricted to RCTs using concurrent chemoradiation, 3-year survival strongly favored the combination of neoadjuvant chemoradiation and surgery (OR 0.45, 95% CI 0.26 to 0.79; P ⫽ 0.005). Conversely, RCTs using sequential chemoradiation did not demonstrate a survival benefit at 3 years (OR 0.82, 95% CI 0.54 to 1.25; P ⫽ 0.36).
Fig. 2. Rate of resection (odds ratio 2.50, 95% confidence interval 1.05 to 5.96; P ⫽ 0.038). (AC ⫽ adenocarcinoma; CTRTS ⫽ chemotherapy, radiation therapy, surgery; S ⫽ surgery.)
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
Fig. 3. Rate of complete resection (odds ratio 0.53, 95% confidence interval 0.33 to 0.84; P ⫽ 0.007). (AC ⫽ adenocarcinoma; SCC ⫽ squamous cell carcinoma; CTRTS ⫽ chemotherapy, radiation therapy, surgery; S ⫽ surgery.)
Data on patterns of cancer recurrence were provided in only three of nine RCTs. Patients treated with neoadjuvant chemoradiation and surgery had fewer local-regional cancer recurrences (OR 0.38, 95% CI 0.23 to 0.63; P ⫽ 0.0002). Distant recurrence (OR 0.88, 95% CI 0.55 to 1.41; P ⫽ 0.60) and all cancer recurrence (OR 0.47, 95% CI 0.16 to 1.45; P ⫽ 0.19) were similar in the two groups of patients. The rate of adverse treatment events was not significantly different in the two patient groups, but there was a trend in favor of surgery alone for both operative mortality and all treatment mortality. Odds ratio (95% CI; P value), expressed as chemoradiation and surgery versus surgery alone (values ⬍1 favor chemoradiation-surgery arm), was 1.72 (0.96, 3.07; P ⫽ 0.07) for operative mortality, 1.29 (0.66, 2.55; P ⫽ 0.46) for anastomotic leaks, 1.21 (0.65, 2.29; P ⫽ 0.55) for postoperative pulmonary complications, and 1.63 (0.99, 2.68; P ⫽ 0.053) for all treatment mortality. Pooled data analysis for the chemoradiation treated patients showed a complete pathological response in 21% of patients. Chemoradiation mortality (before surgery) was 1.2%. Data on the extent and quality of surgical resection, and lymphadenectomy, was difficult to discern from the published articles. One trial featured transhiatal esophagectomy exclusively [14]. The majority of patients treated in the other RCTs underwent transthoracic esophagectomies.
Comments Surgeons and oncologists remain disappointed with surgery as a solitary treatment modality for esophageal cancer. Not surprisingly, various combinations of chemotherapy, radiotherapy, and surgery (multimodality treatment) have been investigated [6 – 8]. Both neoadjuvant radiotherapy and surgery, and surgery and adjuvant radiotherapy, have been studied. Randomized controlled trials and meta-analyses of trials have not shown a significant survival advantage for these combinations of surgery and radiotherapy [21–24]. Surgery followed by adjuvant chemotherapy is a common treatment approach for many solid tumors, but
541
postesophagectomy chemotherapy has not been extensively studied; one randomized trial showed no benefit for surgery followed by chemotherapy [25]. Several randomized controlled trials have compared neoadjuvant chemotherapy and surgery to surgery alone [26 –30]. One large trial showed improved survival with neoadjuvant chemotherapy [26], but another large trial and a meta-analysis of all trials showed little real benefit [27,30]. The rate of pathological complete response was low. Current neoadjuvant chemotherapy regimes are probably not effective enough to improve survival [27,30]. The neoadjuvant combination of chemotherapy and radiation therapy, as opposed to neoadjuvant chemotherapy, has brought about dramatic clinical and pathological responses in nonrandomized esophageal cancer trials [6 – 8,31]. It is an intuitively appealing treatment strategy. Micrometastatic disease is treated early, treatment is delivered while tumor blood supply is intact, and neoadjuvant treatment can facilitate resection by downstaging tumors [6 – 8]. Chemotherapy and radiation therapy act synergistically at the local-regional cancer level. However, chemoradiation therapy can increase the morbidity and mortality of esophagectomy. Surgeons are particularly concerned about the technical difficulties of operating in a pretreated field, wound (anastomotic) healing problems, and postoperative pulmonary complications (especially acute respiratory distress syndrome) [32–35]. Finally, neoadjuvant chemoradiation may cause death independent of surgery (before surgery) [33]. Surgery followed by adjuvant chemotherapy and radiotherapy is used for some gastrointestinal malignancies, but esophageal cancer patients often lack the general strength to complete this postoperative treatment. Adjuvant chemoradiation after esophagectomy has not been investigated with randomized trials. Nevertheless, some centers have reported encouraging phase II data [36], and the therapeutic strategy is effective in gastric cancer [37]. Further consideration of this treatment approach for esophageal cancer is probably warranted. Our meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for esophageal cancer showed a survival benefit (3-year survival) for the combination of chemoradiation and surgery. The survival benefit is related, in part, to improved local-regional cancer control brought about by the neoadjuvant treatment. Neoadjuvant chemoradiation did not significantly reduce the incidence of cancer recurrence at distant sites. The sensitivity analysis of chemoradiation scheduling showed the importance of concurrent chemoradiation. Concurrent, as opposed to sequential, chemoradiation allows maximal antineoplastic synergy between chemotherapeutic agents and radiation treatment. The meta-analysis showed a marked difference in resection and complete resection rates for the two treatment groups. Patients treated with surgery alone were more likely to undergo esophageal resection, but those treated with
542
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543
chemoradiation and surgery were more likely to have a complete (R0) resection performed. This suggests that neoadjuvant chemoradiation therapy downstages tumors and facilitates complete resection. The lower rate of local-regional cancer recurrence in the neoadjuvant group supports this contention. However, neoadjuvant therapy probably also acts as a selection tool for surgery [8,30]. Conceivably, patients with biologically aggressive tumors, who might undergo palliative resections if treated with surgery alone, may not respond to neoadjuvant chemoradiation. These patients may deteriorate during their neoadjuvant therapy and never have resection attempted. Our meta-analysis showed a nonsignificant trend toward increased operative mortality (OR 1.72, 95% CI 0.96 to 3.07, P ⫽ 0.07) and increased all treatment mortality (OR 1.63, 95% CI 0.99 to 2.68, P ⫽ 0.053) in the neoadjuvant group. There is little doubt that esophagectomy can be a technically challenging operation when performed after neoadjuvant chemoradiation treatment. Surgeons intuitively understand the general relationship between operative difficulty and postoperative complications. More specifically, radiation treatment can potentially set the stage for failures of wound healing (anastomotic leaks) and postoperative acute lung injury [8,32–35]. The risk of postoperative acute lung injury is related, in part, to the specifics of radiation therapy. Large radiation fractions and treatment of large volumes of lung tissue are undesirable. Both surgeons and oncologists are concerned that the survival benefits of neoadjuvant chemoradiation can be negated by an increase in postoperative deaths [33]. This was observed in the trial reported by Bosset [13]. In summary, this meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for esophageal cancer showed a survival benefit for neoadjuvant chemoradiation treatment. Compared with surgery alone, neoadjuvant chemoradiation and surgery improved 3-year survival and reduced localregional cancer recurrence. Neoadjuvant chemotherapy and radiotherapy was most effective when given concurrently. Neoadjuvant chemoradiation was associated with a lower rate of esophageal resection, but a higher rate of complete (R0) resection. There was a nonsignificant, but very concerning, trend toward increased treatment mortality with neoadjuvant chemoradiation.
References [1] Nigro JJ, DeMeester SR, Hagen JA, et al. Node status in transmural esophageal adenocarcinoma and outcome after en bloc esophagectomy. J Thorac Cardiovasc Surg 1999;117:960 – 8. [2] Lerut T, Coosemans W, De Leyn P, et al. Is there a role for radical esophagectomy? Eur J Cardiothorac Surg 1999;16(suppl 1):S44 –7. [3] Ando N, Ozawa S, Kitagawa Y, et al. Improvement in the results of surgical treatment of advanced squamous esophageal carcinoma during 15 consecutive years. Ann Surg 2000;232:225–32.
[4] Whooley BP, Law S, Murthy SC, et al. Analysis of reduced death and complication rates after esophageal resection. Ann Surg 2001;233: 338 – 44. [5] Orringer MB, Marshall B, Iannettoni MD. Transhiatal esophagectomy: clinical experience and refinements. Ann Surg 1999;230:392– 403. [6] Sutton P, Clark P. Neo-adjuvant treatment for oesophageal cancer. GI Cancer 2000;3:231– 8. [7] Lehnert T. Multimodal therapy for squamous carcinoma of the esophagus. Br J Surg 1999;86:727–39. [8] Geh JI, Crellin AM, Glynne-Jones R. Preoperative (neoadjuvant) chemoradiotherapy in oesophageal cancer. Br J Surg 2001;88:338 – 56. [9] Nygaard K, Hagen S, Hansen HS, et al. Pre-operative radiotherapy prolongs survival in operable esophageal carcinoma: a randomized, multicenter study of pre-operative radiotherapy and chemotherapy. The second Scandinavian trial in esophageal cancer. World J Surg 1992;16:1104 –10. [10] Apinop C, Puttisak P, Preecha N. A prospective study of combined therapy in esophageal cancer. Hepatogastroenterology 1994;41:391–3. [11] Le Prise E, Etienne PL, Meunier B, et al. A randomized study of chemotherapy, radiation therapy, and surgery versus surgery for localized squamous cell carcinoma of the esophagus. Cancer 1994;73: 1779 – 84. [12] Walsh TN, Noonan N, Hollywood D, et al. A comparison of multimodality therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996;335:462–7. [13] 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;337:161–7. [14] 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;19:305–13. [15] Law S, Kwong DLW, Tung HM, et al. Preoperative chemoradiation for squamous cell esophageal cancer: a prospective randomized trial [abstract]. Can J Gastroenterol 1998;12(suppl B):161. [16] Walsh TN, McDonnell CO, Mulligan ED, et al. Multimodal therapy versus surgery alone for squamous cell carcinoma of the esophagus: a prospective randomized trial [abstract]. Gastroenterology 2000; 118(suppl 2):1008. [17] Burmeister BH, Smithers BM, Fitzgerald L, et al. A randomized phase III trial of preoperative chemoradiation followed by surgery (CR-S) versus surgery alone (S) for localized resectable cancer of the esophagus [abstract]. Proc Am Soc Clin Oncol 2002;21:518. [18] Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1–12. [19] Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. QUOROM Group. Br J Surg 2000;87:1448 – 54. [20] Urschel JD, Goldsmith CH, Tandan VR, Miller JD. Users’ guide to evidence-based surgery: how to use an article evaluating surgical interventions. Can J Surg 2001;44:95–100. [21] Arnott SJ, Duncan W, Gignoux M, et al. Preoperative radiotherapy in esophageal carcinoma: a meta-analysis using individual patient data (Oesophageal Cancer Collaborative Group). Int J Radiat Oncol Biol Phys 1998;41:579 – 83. [22] Teniere P, Hay JM, Fingerhut A, Fagniez PL. Postoperative radiation therapy does not increase survival after curative resection for squamous cell carcinoma of the middle and lower esophagus as shown by a multicenter controlled trial. French University Association for Surgical Research. Surg Gynecol Obstet 1991;173:123–30. [23] Fok M, Sham JS, Choy D, et al. Postoperative radiotherapy for carcinoma of the esophagus: a prospective, randomized controlled study. Surgery 1993;113:138 – 47.
J.D. Urschel and H. Vasan / The American Journal of Surgery 185 (2003) 538 –543 [24] Zieren HU, Muller JM, Jacobi CA, et al. Adjuvant postoperative radiation therapy after curative resection of squamous cell carcinoma of the thoracic esophagus: a prospective randomized study. World J Surg 1995;19:444 –9. [25] Ando N, Iizuka T, Kakegawa T, et al. A randomized trial of surgery with and without chemotherapy for localized squamous carcinoma of the thoracic esophagus: the Japan Clinical Oncology Group study. J Thorac Cardiovasc Surg 1997;114:205–9. [26] Medical Research Council Oesophageal Cancer Working Group. Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. Lancet 2002;359:1727– 33. [27] Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 1998;339:1979 – 84. [28] Law S, Fok M, Chow S, et al. Preoperative chemotherapy versus surgical therapy alone for squamous cell carcinoma of the esophagus: a prospective randomized trial. J Thorac Cardiovasc Surg 1997;114: 210 –7. [29] Ancona E, Ruol A, Santi S, et al. Only pathologic complete response to neoadjuvant chemotherapy improves significantly the long term survival of patients with resectable esophageal squamous cell carcinoma: final report of a randomized, controlled trial of preoperative chemotherapy versus surgery alone. Cancer 2001;91:2165–74.
543
[30] Urschel JD, Vasan H, Blewett CJ. A meta-analysis of randomized controlled trials that compared neoadjuvant chemotherapy and surgery to surgery alone for resectable esophageal cancer. Am J Surg 2002;183:274 –9. [31] Leichman L, Steiger Z, Seydel HG, et al. Preoperative chemotherapy and radiation therapy for patients with cancer of the esophagus: a potentially curative approach. J Clin Oncol 1984;2:75–9. [32] Urschel JD. Esophagogastrostomy anastomotic leaks complicating esophagectomy—a review. Am J Surg 1995;169:634 – 40. [33] Keller SM, Ryan LM, Coia LR, et al. High dose chemoradiotherapy followed by esophagectomy for adenocarcinoma of the esophagus and gastroesophageal junction: results of a phase II study of the Eastern Cooperative Oncology Group. Cancer 1998;83:1908 –16. [34] Avendano CE, Flume PA, et al. Pulmonary complications after esophagectomy. Ann Thorac Surg 2002;73:922– 6. [35] Schilling MK, Eichenberger M, Maurer CA, et al. Ketoconazole and pulmonary failure after esophagectomy: a prospective clinical trial. Dis Esophagus 2001;14:37– 40. [36] Bedard EL, Inculet RI, Malthaner RA, et al. The role of surgery and postoperative chemoradiation therapy in patients with lymph node positive esophageal carcinoma. Cancer 2001;91:2423–30. [37] Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 2001;345:725– 30.