Jin-hong Park, San Min Yoon, and Chang Sik Yu equally contributed to this study.
Randomized phase 3 trial comparing preoperative and postoperative chemoradiotherapy with capecitabine for locally advanced rectal cancer
Article first published online: 15 FEB 2011
Copyright © 2011 American Cancer Society
Volume 117, Issue 16, pages 3703–3712, 15 August 2011
How to Cite
Park, J.-h., Yoon, S. M., Yu, C. S., Kim, J. H., Kim, T. W. and Kim, J. C. (2011), Randomized phase 3 trial comparing preoperative and postoperative chemoradiotherapy with capecitabine for locally advanced rectal cancer. Cancer, 117: 3703–3712. doi: 10.1002/cncr.25943
- Issue published online: 3 AUG 2011
- Article first published online: 15 FEB 2011
- Manuscript Accepted: 22 DEC 2010
- Manuscript Revised: 14 DEC 2010
- Manuscript Received: 22 OCT 2010
- rectal cancer;
- preoperative chemoradiotherapy;
- postoperative chemoradiotherapy;
- randomized controlled trial
Although many trials have shown the efficacy of preoperative chemoradiotherapy (CRT) or postoperative CRT compared with surgery alone, the optimal sequence of radiotherapy and surgery is unclear. The authors reported the final results of this single institution prospective randomized phase 3 trial comparing preoperative CRT with postoperative CRT using capecitabine in survival, local control, sphincter preservation, and toxicity for the treatment of locally advanced rectal cancer.
Patients with locally advanced rectal cancer (cT3, potentially resectable cT4 or N+) were randomly assigned to receive preoperative or postoperative CRT. CRT consisted of 50 Gy/25 fractions and concurrent capecitabine (1,650 mg/m2/day). Total mesorectal excision was performed.
From March 2004 to April 2006, 240 patients were enrolled. Clinical characteristics were well balanced between both arms, except for more low-lying (<5 cm from anal verge) tumors in the preoperative CRT arm (60% vs 46%, P = .041). After a median follow-up time of 52 months, the 3- and 5-year disease-free survival, overall survival, and cumulative incidence of local recurrence were similar between both arms. However, for the patients with low-lying tumors, the preoperative CRT arm had a higher rate of sphincter preservation (68% vs 42%, P = .008). Acute and late complication rates were similar between both arms.
Although significant benefit of preoperative CRT in local control and survival was not demonstrated, the data showed that increased rate of sphincter preservation was possible in low-lying tumors without jeopardizing local control and surgical complication by preoperative CRT. Cancer 2011;. © 2011 American Cancer Society.
In locally advanced rectal cancer, adjuvant chemoradiotherapy (CRT) improves local control and survival.1 Although radical surgery followed by adjuvant CRT had been the standard treatment, there was a recent change from postoperative CRT to preoperative CRT based on several studies.2, 3 These trials showed the superiority of preoperative CRT over postoperative CRT, but there were several differences in the results between the trials.2, 3 The German trial demonstrated significant improvement in local control, sphincter preservation rate, and treatment-related toxicities in the preoperative CRT arm but not in disease-free survival (DFS) or overall survival (OS).2 However, the NSABP R-03 trial showed improved DFS and a trend toward improved OS in the preoperative CRT arm but not in local control, sphincter preservation, and treatment-related toxicities.3 In previous randomized trials that compared preoperative CRT with radiotherapy (RT) alone for the treatment of locally advanced rectal cancer, preoperative CRT improved local control.4, 5
In those randomized controlled trials for preoperative CRT, bolus injection or continuous infusion of 5-fluorouracil (5-FU) was used.2-5 Although CRT with protracted 5-FU infusion has proven to be superior to conventional bolus injection,6 protracted infusion is an inconvenient and invasive procedure, requires an ambulatory infusion pump, and is associated with increased risk of infection and thrombosis. Several studies have examined the use of new oral chemotherapeutic agents in the CRT protocol for treatment of rectal cancer with many biologic and clinical advantages including mimicking protracted infusion of 5-FU.7-10 In a prospective phase 2 trial of preoperative CRT with capecitabine, we reported encouraging rates of tumor down-staging, sphincter preservation, and low toxicity.11
With such encouraging outcomes, we designed a prospective, single institution phase 3 trial that compared preoperative CRT and postoperative CRT in survival, local control, sphincter preservation, and treatment-related toxicity for treatment of locally advanced rectal cancer. We hypothesized that the superiority of preoperative CRT over postoperative CRT could be effectively evaluated by a single institution study of a high-volume center for better quality assurance in surgery and RT with the potential advantages of the modern approach of CRT using capecitabine. We present the final results of our trial with a median follow-up time of 52 months.
MATERIALS AND METHODS
Eligibility and Randomization
The Institutional Review Board of Asan Medical Center approved this protocol. Patients were enrolled from March 2004 to April 2006. Patients with locally advanced rectal cancer (cT3 or potentially resectable cT4 or positive regional lymph node) on endorectal ultrasonography (EUS) and abdominopelvic computed tomography (CT) scan qualified for this study. Further eligibility criteria were similar to our previous phase 2 trial (Table 1).11 After written informed consent was obtained, all patients were randomly assigned to the preoperative CRT arm or the postoperative CRT arm (Fig. 1A). The treatment allocation was performed in a randomized permuted block method using random number tables and included stratification by gender.
|1. Histologically confirmed rectal adenocarcinoma|
|2. Tumor located below 10 cm from the anal verge|
|3. Clinical T3 or potentially resectable T4 stage or positive regional node on endorectal ultrasonography (EUS) and computed tomography (CT)|
|4. Patients must be >18 years and <76 years|
|5. Eastern Cooperative Oncology Group (ECOG) performance status 0-2|
|6. Adequate bone marrow reserve (white blood cell count ≥4,000/mm3, absolute neutrophil count ≥1,500/mm3, platelet count ≥100,000/mm3, hemoglobin ≥10 g/dL)|
|7. Adequate renal function (serum creatinine level ≤1.5 mg/dL, calculated creatinine clearance ≥50 mg/min)|
|8. Adequate liver function (liver transaminase levels ≤3 times the upper normal limits, serum bilirubin ≤1.5 mg/dL).|
|9. Signed informed consent prior to randomization|
|1. Evidence of distant metastasis|
|2. Previous history of chemotherapy or radiotherapy|
|3. History of malignancy during recent 5 years other than skin cancer|
|4. Pregnant or lactating woman|
|5. Family history of colorectal cancer|
The radiation dose for preoperative treatment was 46 Gy in 23 fractions to the whole pelvis followed by a boost dose of 4 Gy in 2 fractions. The radiation dose for postoperative RT was 50 Gy in 25 fractions to the whole pelvis. Detailed information on RT has been described previously.11
Capecitabine (825 mg/m2 twice per day, without weekend breaks) was initiated on the first day of RT and was delivered concurrently with RT. Adjuvant chemotherapy was initiated 4 weeks after surgery in the preoperative CRT arm and at 4 weeks after completion of CRT in the postoperative CRT arm. Adjuvant chemotherapy consisted of 4 cycles of capecitabine (2,500 mg/m2/day for 14 days, followed by a 1-week break) or 4 cycles of bolus 5-FU/leucovorin (375 mg 5-FU/m2/day and 20 mg leucovorin/m2/day for 5 days every 4 weeks), which was chosen based on patient economic status, because capecitabine is not covered by the medical insurance system of Korea. Patients were instructed to take capecitabine twice per day at 12-hour intervals, and to take one of the doses 1 hour before RT to maximize the radiosensitization effect based on our previous study.12
Total mesorectal excision (TME) was performed as the standard procedure, and the particular type of surgery was determined at the time of resection. All operations were carried out by specialist colorectal surgeons who had performed more than 200 TMEs each year for the past 5 years. Surgery was performed 4 to 6 weeks after completion of CRT.
Pretreatment evaluation included complete blood counts (CBC) and biochemical profiles, serum carcinoembryonic antigen, colonoscopy with biopsy, abdominopelvic CT scan, EUS, and chest CT scan. Clinical staging was based on abdominopelvic CT and EUS findings and the American Joint Committee on Cancer (AJCC) staging system.13
Treatment-related toxicity was evaluated according to National Cancer Institute Common Toxicity Criteria version 2.0.14 During CRT, patients were examined weekly. After completion of treatment, patients were scheduled for follow-up every 3 months for first 2 years, then every 6 months thereafter. Complete history and physical examination, complete blood count (CBC), biochemical profile, serum carcinoembryonic antigen (CEA), and chest radiography were performed at each follow-up. Abdominopelvic CT scan was performed every 6 months for first 2 years, then annually thereafter. Colonoscopic evaluation was performed annually.
After surgery and tumor fixation, the largest diameter of the tumor was recorded. Then tumor was examined by slicing it into 3- to 5-mm sections and stained with hematoxylin and eosin. If tumor cells were not found in initial examination, thinner slices were taken for further inspection. Complete lymph node dissection from perirectal fat tissue and microscopic examination were also performed. Pathologic stage was determined using the TNM classification system, as recommended by the AJCC.
The primary end point was 3-year DFS. This study was designed to have 85% statistical power to detect an increase in DFS of 15% at 3 years in the preoperative CRT group, with a 2-sided significance level of 0.05. This required a sample size of 216 patients per treatment arm. The secondary end points were OS, local or distant relapse, sphincter preservation, and treatment-related toxicities.
All statistical analysis was performed using SPSS (version 14.0, SPSS Inc, Chicago, IL). Chi-square tests or Fisher exact tests were used to compare patient characteristics between both arms for categorical variables, and Student t tests were used to compare quantitative variables. All eligible patients were included in the evaluation of OS, DFS, and the cumulative incidence of local and distant recurrences based on intent-to-treat analysis. All events were determined from the first day of RT in the preoperative CRT arm and from the date of surgery in the postoperative CRT arm. Pathologic confirmation of recurrent disease was encouraged. If histologic evidence was not available, a clear demonstration of recurrent lesions or serial enlargement of the lesions based on radiology was accepted as the evidence of treatment failure. Elevated CEA alone was not considered as evidence of recurrence. Local recurrence was defined as tumor recurrence within radiation field in pelvic cavity, and distant metastasis was defined as any recurrence outside of pelvic cavity. All survival curves were estimated using the Kaplan-Meier method and were compared using the log-rank test. Two-tailed probability (P) values less than .05 were considered to indicate a significant difference.
Accrual of the Patients and Patient Characteristics
Between March 2004 and April 2006, 240 patients were randomly assigned to receive preoperative or postoperative CRT. We could not enroll the planned number of patients because patient accrual decreased after publication of the study by Sauer et al.2 In addition, randomization could have been affected by investigator preference for preoperative treatment for low-lying tumors; hence, we closed this protocol earlier than initially planned.
Among the 240 patients initially enrolled, 20 were determined to be ineligible for the following reasons: 10 patients withdrew consent, 5 had distant metastases that were detected after enrollment, 3 had early stage disease (cT2N0) on additional staging work-ups, 1 had a history of pelvic irradiation for uterine cervical cancer, and 1 had recurrent rectal cancer. Most of the characteristics of the 220 eligible patients were relatively well balanced between both arms (Table 2). However, there were significantly more patients with low-lying tumors (<5 cm from the anal verge) in the preoperative treatment arm (Table 2).
|Variables||Preoperative CRT Arm (n=107)||Postoperative CRT Arm (n=113)||P|
|Male||67 (63)||71 (63)|
|Female||40 (37)||42 (37)|
|Normal||88 (82)||98 (87)|
|Increased||19 (18)||15 (13)|
|Low (<5 cm)||64 (60)||52 (46)|
|Middle (5-10 cm)||43 (40)||61 (54)|
|Well differentiated||17 (18)||9 (8)|
|Moderately differentiated||71 (76)||93 (85)|
|Poorly differentiated||5 (6)||8 (7)|
|T3N0||35 (32)||36 (32)|
|T4N0||0 (0)||1 (1)|
|T2N+||1 (1)||3 (2)|
|T3N+||70 (66)||72 (64)|
|T4N+||1 (1)||1 (1)|
A total of 107 patients were randomly assigned to the preoperative CRT arm and 113 patients to the postoperative CRT arm. CRT was not administered to 16 patients assigned to the postoperative CRT arm (Fig. 1B). Two patients in the preoperative CRT arm did not have surgical resection after preoperative CRT, and there were protocol violations in 11 patients, including 10 patients in the postoperative CRT arm who were transferred to regional hospitals and received nonprotocol-based treatment (Fig. 1B).
RT was completed according to protocol in 106 patients from the preoperative CRT arm and 86 patients from the postoperative CRT arm (99% vs 76%, P < .001) (Table 3). Concurrent chemotherapy was completed according to protocol in 100 patients from the preoperative CRT arm and 85 patients from the postoperative CRT arm (93% vs 75%, P < .001). Thus, 99 patients (93%) in the preoperative CRT arm and 84 patients (74%) in the postoperative CRT arm completed CRT as protocol.
|Variables||Preoperative CRT Arm (n=107)||Postoperative CRT Arm (n=113)||P|
|50 Gy||106 (99)||86 (76)|
|48 Gy||1 (1)||1 (1)|
|No CRT||0 (0)||16 (14)|
|Full dose||100 (93)||85 (75)|
|Dose reduction||3 (3)||1 (1)|
|Incompletion of the full cycle||4 (4)||1 (1)|
|No CRT||0 (0)||16 (14)|
|Capecitabine||82 (77)||66 (58)|
|Incompletion of the full cycle||4||2|
|5-FU/Leucovorin||13 (12)||17 (15)|
|Incompletion of the full cycle||0||2|
|No chemotherapy||12 (11)||20 (18)|
|Protocol violation||0 (0)||10 (9)||.002|
Among all patients, except the patients who did not received the postoperative CRT (16 patients), 16 (7%) never started adjuvant chemotherapy due to disease progression (5 patients), postoperative complication (3 patients), refusal (3 patients), toxic effects of concurrent CRT (1 patient), and other causes (4 patients). Treatment compliance with the adjuvant chemotherapy was better in the preoperative CRT arm. Adjuvant chemotherapy was administered to 95 patients in the preoperative CRT arm and 83 patients in the postoperative CRT arm (89% vs 73%, P = .002) as full dose or modified dose/cycle. The full protocol dose was administered to 68 patients in the preoperative-CRT arm and 56 patients in the postoperative CRT arm (64% vs 50%, P = .036).
Surgery and Pathologic Findings
Complete resection confirmed by pathologic examination was performed in 217 of 218 patients who had resection. In 1 patient, the tumor could not be completely resected due to extensive vaginal invasion. The sphincter-sparing rate was similar in the 2 groups; however, among patients with low-lying tumors, the preoperative CRT arm had a higher rate of sphincter-sparing surgery (Table 4).
|All Patients||Patients With Low-Lying Tumorsa|
|Preoperative CRT Arm||Postoperative CRT Arm||Preoperative CRT Arm||Postoperative CRT Arm|
|APR||19 (18)||31 (27)||18 (29)||29 (56)|
|LAR||84 (80)||81 (72)||42 (68)||22 (42)|
|Others||2 (2)||1 (1)||2 (3)||1 (2)|
Pathologic stage was lower in the preoperative CRT arm than in the postoperative CRT arm (Table 5). Complete disappearance of the primary tumor in pathology specimens (ypCR-T) occurred in 22 patients (21%), and 83 patients (79%) had no tumor cells in their lymph node specimen (ypCR-N). Four of the ypCR-T patients had residual tumor cells in lymph nodes; thus, complete tumor disappearance occurred in 18 patients (17%). The mean tumor size was significantly smaller in the preoperative CRT arm and microscopic tumor invasions such as perineural and/or lymphovascular invasion were less frequent in preoperative CRT arm (Table 5).
|Variables||Preoperative CRT Arm (n=105)||Postoperative CRT Arm (n=113)||P|
|T0||22 (21)||0 (0)|
|T1||3 (3)||1 (1)|
|T2||31 (29)||16 (14)|
|T3||49 (47)||92 (81)|
|T4||0 (0)||4 (4)|
|N0||83 (79)||36 (32)|
|N1||16 (15)||44 (39)|
|N2||6 (6)||33 (29)|
|0||18 (17)||0 (0)|
|I||26 (25)||9 (8)|
|II||34 (32)||24 (21)|
|III||26 (25)||77 (68)|
|IV||1 (1)||3 (3)|
|No||94 (89)||90 (80)|
|Yes||11 (11)||23 (20)|
|No||96 (91)||70 (62)|
|Yes||9 (9)||43 (38)|
|Complete||105 (100)||112 (99)|
|Incomplete||0 (0)||1 (1)|
Acute Toxicities and Postoperative Complications
Preoperative and postoperative CRT was relatively well tolerated in most patients, and there was no treatment-related mortality. Most acute adverse events were mild to moderate, and all of them resolved spontaneously with supportive care (Table 6). The overall incidence of acute toxicities was comparable between both arms (15% vs 16%, P = .827).
|Preoperative CRT Arm (n=107)||Postoperative CRT Arm (n=87)|
|Dermatitis||3 (3)||5 (6)|
|Proctitis||3 (3)||4 (5)|
|Hand-foot syndrome||3 (3)||2 (2)|
|Leukopenia||3 (3)||4 (5)|
|Anemia||6 (6)||1 (1)|
|Thrombocytopenia||1 (1)||0 (0)|
|Any grade 3 or higher toxicities||16 (15)||14 (16)|
|Ileus/obstruction||5 (5)||5 (6)|
|Anastomotic leakage||2 (2)||5 (6)|
|Wound healing||1 (1)||1 (1)|
|Fistula||1 (1)||0 (0)|
|Any grade 3 or higher toxicities||8 (9)||10 (12)|
|Pelvic abscess||2 (2)||0 (0)|
|Anastomotic leakage||2 (2)||0 (0)|
|Ureteral stricture||2 (2)||0 (0)|
|Ileus/obstruction||1 (1)||2 (2)|
|Fistula||1 (1)||0 (0)|
|Rectal stricture||0 (0)||1 (1)|
|Any grade 3 or higher toxicities||8 (8)||3 (3)|
Perioperative (within 2 months of surgery) and late complications (more than 2 months after surgery) were also similar between both arms (Table 6). All patients recovered after supportive care or surgical treatment. Among the patients with perioperative complications, 8 patients in the preoperative CRT arm and 10 patients in the postoperative CRT arm had grade 3 or 4 toxicities (9% vs 12%, P = .359). Among the patients with late complications, 8 patients in the preoperative CRT arm and 3 patients in the postoperative CRT arm had grade 3 or 4 toxicities (8% vs 3%, P = .350).
Patterns of Failure and Survival
After a median follow-up time of 52 months, 24% of patients in the preoperative CRT arm and 25% of patients in the postoperative CRT arm experienced treatment failure. Follow-up data were missing for 1 patient in the postoperative CRT arm. In the preoperative CRT arm, we observed local failure alone in 3 patients (3%), distant failure alone in 23 patients (21%), and synchronous local and distant failure in 1 patient (1%). In the postoperative CRT arm, we observed local failure alone in 2 patients (2%), distant failure alone in 22 patients (19%), and synchronous local and distant failure in 5 patients (4%).
The estimated 3- and 5-year DFS (77% vs 74% and 73% vs 74%, P = .8656) and OS (91% vs 90% and 83% vs 85%, P = .6204) did not differ significantly between both arms (Figs. 2A and 2B). There was also no significant difference between both arms in the 3- and 5-year cumulative incidence of local recurrence (3% vs 6% and 5% vs 6%, P = .3925) and distant metastasis (21% vs 24% and 23% vs 24%, P = .7384) (Figs. 2C and 2D). Although tumor location was explored to investigate any influence of the treatment effect between the 2 arms on survival or local control, there was no evidence of any interactions.
Compliance and Toxicity
In the current trial, preoperative and postoperative CRT with capecitabine was well tolerated by most patients, and there was no significant difference in acute and late toxicities in both arms. Compliance rate was higher in the preoperative CRT arm. However, most of the noncompliant patients in the postoperative CRT arm were those who were transferred from distant cancer centers and travel problems led to noncompliance. Although they were regarded as a noncompliance group in our protocol, they all received one of the standard treatments recommended by National Comprehensive Cancer Network (NCCN) guidelines in regional cancer centers.
Postoperative CRT arm of our trial had a more favorable toxicity profile and compliance than those of the German or NSABR-R03 trial,2, 3 although we cannot explain. It is possible that this difference might be due to the use of different chemotherapeutic agents. The design of the current trial was similar to other phase 3 trials,2, 3 but we used oral capecitabine based on encouraging results of our previous studies.11, 12 Clearly, more data are needed to determine whether continuous infusion or bolus injection of 5-FU is more effective for treatment of rectal cancer.15 However, previous retrospective studies comparing capecitabine and 5-FU indicated that capecitabine was at least as effective as 5-FU in preoperative CRT regimen.16-19 The NSABP R-04 trial is expected to provide important information about the efficacy of oral capecitabine versus 5-FU chemotherapy.20 This trial has completed accrual and the results are pending. Surgeon factor could be another possible reason for the superior result of the current study than the German trial. Because the current trial was a single institution study of a high-volume center and our retrospective study also showed relatively low incidence of late complications in the postoperative RT group, better quality in surgery and RT could be possible reasons.21
In our current study, we used a 7-day-per-week capecitabine schedule based on our previous phase 2 trial that showed encouraging rates of tumor down-staging and sphincter preservation with a low toxicity profile.11 However, a 7-day-per-week schedule is not the golden standard of care, and a 5-day-per-week schedule used in many centers could be superior to the 7-day regimen in the view of compliance and toxicity, with similar efficacy.22 We think that further study concerning this issue will be needed for the establishment of an ideal regimen of capecitabine with preoperative RT in rectal cancer.
One of the rationales for preoperative CRT is the enhancement of sphincter preservation in patients with low rectal cancer. The current study showed that low-lying rectal cancer patients in the preoperative CRT arm had significantly better sphincter preservation than did patients in the postoperative CRT arm, but sphincter preservation did not produce negative effects in terms of complications and disease control in the preoperative CRT arm.
However, functional outcome after sphincter preservation was not examined in the German and the current trial.2 The ultimate goal of sphincter preservation must be improvement in quality of life and not simply preservation of the sphincter complex. We suggest that future trials use a functional end point to assess sphincter preservation.
In the current trial, even though we performed meticulous pathologic examination described previously,11 our ypCR rate or other histologic parameters were comparable with other studies that used capecitabine,16-18, 23-26 and our ypCR rate was higher than that of the German trial (17% vs 8%) in spite of more adverse factors (Table 7).2
|Authors||Sauer et al2||Rho et al3||Current Trial|
|No. of eligible patients||799||254||220|
|Median follow-up (y)||3.8||8.4||4.3|
|Clinical node-positive disease||52a||Unknown||67|
|pCR in preoperative CRT arm||8||15||17|
|pT1-2N0 in postoperative CRT arm||18||Unknown||8|
|Sphincter preservation rated||39/19e||48/39f||68/42g|
|5-year local recurrence rated||6/13||11/11||5/6|
|5-year disease-free survival rated||68/65||65/53||73/74|
|5-year overall survival rated||76/74||75/66||83/85|
|Any grade 3 or higher acute toxicitiesd||27/40||52/49||15/16|
The exact reason for these differences is not clear, but we propose the following 2 explanations. First, use of capecitabine during the whole period of RT with an optimized time interval between capecitabine intake and radiation might have enhanced the radiosensitizing effect of capecitabine relative to bolus 5-FU injection.12 Second, racial and genetic differences among the patients of East and West may account for the differences. The first explanation seems to be reasonable but not conclusive because pathologic complete response rate has been reported to be slightly higher in the capecitabine-based regimen than 5-FU.2, 3, 16-18, 20, 27 The second explanations are difficult to prove but commonly used.
Local Control and Survival
Previous German and NSABP phase 3 trials demonstrated the superiority of preoperative CRT in terms of local control and DFS (Table 7),2, 3 but we could not observe such benefits in the current study. Because the statistical power of the current trial was limited owing to premature closure, the superiority of preoperative CRT over postoperative CRT might not be excluded. However, our postoperative CRT arm had better 5-year OS and DFS rates (85% and 74%) than those reported in the German trial (74% and 65%) and NSABP-R03 trial (66% and 53%).2, 3 The 5-year cumulative incidence of local recurrence in our postoperative-CRT arm was similar to that of the preoperative CRT arm of the German trial (6%).2
One possible reason for this discordance may be that patients in the 2 studies had different distribution of cancer stage. However, we believe that this is unlikely because the proportion of clinical T4 disease was only 3% in the German trial and pathologic stage 3 disease was more prevalent in our study (68% and 40%). Moreover, the rate of overstaging in our study was 8%, which was lower than in the German trial (18%).2 Relatively high compliance rate in our study may be another possible explanation for this discordance. Pre- and postoperative CRT were both well tolerated by most patients in our study. Other trials reported very high rates of acute toxicities of grade 3 or higher (40%-49%), much higher than for the patients of our postoperative CRT arm.2, 3 Moreover, almost half of the patients in the German trial did not receive planned postoperative CRT.2 Another possible reason for our better results may be associated with a single institution study performed in a high-volume hospital. Hospital volume factor seems to play an important role in achieving high local control.28 The current trial was conducted in a high-volume tertiary referral center (with >2,500 beds), and participating surgeons were those who had performed more than 200 TMEs each year for the past 5 years. Also, the parameters for RT in a single institution study would likely be more uniform than in the multicenter trials.29 RT was performed by a single radiation oncologist who majored in gastrointestinal oncology.
One of the major strengths of the current single institution study of a high-volume center is that it allowed for better quality assurance in surgery and RT than a multicenter trial and the multidisciplinary teamwork that goes with that, so the quality of overall care was probably better and more homogeneous than the German trial. Also, the current trial included only middle and distal tumors, which had a higher risk of local recurrence than proximal tumors and which made the current trial more appropriate for evaluation of the role of RT as a local modality. This trial was closed prematurely; although the German trial that used bolus 5-FU globally accepted at that time hampered our ability to complete patient accrual, our trial provided a more modern approach in preoperative CRT for locally advanced rectal cancer with the oral fluoropyrimidine, capecitabine.
The results of the current trial suggested that preoperative or postoperative CRT with capecitabine was well tolerated by most rectal cancer patients and that treatment compliance was excellent. Perioperative and long-term late complications were not increased by use of preoperative CRT. Although we could not demonstrate significant benefit of preoperative CRT in terms of local control and survival, the rate of sphincter-sparing surgery was higher in preoperative CRT patients with low-lying rectal tumors without jeopardizing local control and surgical safety by preoperative CRT.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 13American Joint Committee on Cancer staging manual. 6th ed. New York: Springer, Inc., 2002., , , .