Ann Surg Oncol (2014) 21:1345–1351 DOI 10.1245/s10434-014-3484-1
ORIGINAL ARTICLE – COLORECTAL CANCER
Prognostic Impact of Circumferential Resection Margin in Rectal Cancer Treated with Preoperative Chemoradiotherapy Mi Ri Hwang, MD1, Ji Won Park, MD1,4,5,6, Sohee Park, PhD2, Hyekyoung Yoon, MS3, Dae Yong Kim, MD1, Hee Jin Chang, MD1, Sun Young Kim, MD1, Sung Chan Park, MD1, Hyo Seong Choi, MD1, Jae Hwan Oh, MD1, and Seung-Yong Jeong, MD4,5,6 1
Center for Colorectal Cancer, Research Institute & Hospital, National Cancer Center, Goyang, Gyeonggi-do, Korea; Department of Epidemiology and Heath Promotion, Graduate School of Public Health, Yonsei University, Seoul, Korea; 3 Cancer Biostatistics Branch, Research Institute, National Cancer Center, Goyang, Korea; 4Department of Surgery, Seoul National University College of Medicine, Seoul, Korea; 5Cancer Research Institute, Seoul National University, Seoul, Korea; 6Colorectal Cancer Center, Seoul National University Cancer Hospital, Seoul, Korea 2
ABSTRACT Background. The circumferential resection margin (CRM) is a strong prognostic factor in rectal cancer. The purpose of this study was to investigate the relationship between CRM distance and recurrence in patients with locally advanced rectal cancer who received preoperative chemoradiotherapy (CRT). Methods. We analyzed data for 561 patients who underwent preoperative CRT and curative surgery for locally advanced rectal cancer between August 2001 and December 2008. CRM was divided into four groups: group 1, CRM [ 2 mm; group 2, 1.1–2.0 mm; group 3, 0.1–1.0 mm; and group 4, 0 mm. We assessed the associations of CRM with local recurrence and disease-free survival. Results. Groups 1, 2, 3, and 4 comprised 487, 36, 20, and 18 patients, respectively. The local recurrence rate was highest and the disease-free survival rate was lowest in group 4, followed by groups 3, 2, and 1. Survival was similar between groups 2 and 1. Local recurrence rates were lower in groups 3, 2, and 1 than in group 4 [hazard ratio (HR) 0.28, 95 % confidence interval (CI) 0.09–0.91,
Electronic supplementary material The online version of this article (doi:10.1245/s10434-014-3484-1) contains supplementary material, which is available to authorized users. Ó Society of Surgical Oncology 2014 First Received: 2 August 2012; Published Online: 28 January 2014 J. W. Park, MD e-mail: [email protected]
P = 0.035; HR 0.11, 95 % CI 0.03–0.46, P = 0.002; HR 0.18, 95 % CI 0.08–0.42, P \ 0.0001, respectively]. Disease-free survival rates were higher in groups 3, 2, and 1 than in group 4 (HR 0.32, 95 % CI 0.13–0.75, P = 0.009; HR 0.24, 95 % CI 0.10–0.54, P = 0.001; HR 0.26, 95 % CI 0.14–0.48, P \ 0.0001, respectively). Conclusions. After preoperative CRT, CRM distance provides useful information for risk stratification in the recurrence of rectal cancer. Since the initial report of the prognostic importance of the circumferential resection margin (CRM) in rectal cancer, several studies have shown that CRM is a strong predictor of local recurrence, distant metastasis, and survival.1–3 Although preoperative chemoradiotherapy (CRT) is widely used for the treatment of locally advanced rectal cancer, its effects on the CRM remain unclear. In a downstaged patient group of one study, preoperative CRT was associated with an increase in the mean CRM because of tumor shrinkage, enabling curative surgery and sphincter preservation in patients with lower rectal cancer.4 However, another study demonstrated that preoperative radiotherapy (RT) did not compensate for CRM involvement.5 It was also reported that CRM was a strong predictor of local recurrence in patients who received preoperative CRT.6,7 The cutoff value for CRM involvement was 1 mm in many studies, although other studies used a cutoff value of 2 mm.2,3,8–13 Prior studies showed that a CRM of B1 mm was associated with high local recurrence and low survival rates, similar to a CRM of 0 mm, and the two have been grouped together as positive CRM.3,8 Therefore, a recent
1346
study recommended a CRM of B1 mm as a uniform cutoff value for CRM involvement.14 However, a pooled analysis suggested that a CRM of 0 mm is a more reasonable cutoff value for CRM involvement.15 The definition of CRM involvement for rectal cancer patients has yet to be firmly established. In particular, there are scant data regarding the prognostic ability of CRM distance in patients who have received preoperative CRT. Therefore, in the present study, we analyzed the prognostic ability of CRM distance in patients with locally advanced rectal cancer who had received preoperative CRT. METHODS Patients Between August 2001 and December 2008, a total of 655 patients with locally advanced rectal cancer underwent preoperative CRT at the National Cancer Center, Republic of Korea. Of the 655 patients, 35 refused surgery, and 17 with comorbidities or who rejected anal ablation were treated by local excision. Because CRM is not applicable to intraperitoneal cancer, 31 patients with intraperitoneal rectal cancer were also excluded. Intraperitoneal cancer was defined as cancers that were covered by the peritoneum in pathologic examination, as described in our previous report.16 Eleven patients were lost to follow-up within 6 months. After excluding these patients, we analyzed 561 patients in this study. Patients were divided into 4 groups according to CRM distance stated in the pathologic report: group 1, CRM [ 2 mm; group 2, 1.1–2.0 mm; group 3, 0.1–1.0 mm; and group 4, 0 mm. The study was performed in accordance with the guidelines of the Institutional Review Board of the National Cancer Center. This retrospective analysis was performed using data that were prospectively recorded in a database. Preoperative CRT RT was delivered to the whole pelvis at a dose of 45 Gy in 25 fractions, followed by a boost to the primary tumor of 5.4 Gy in three fractions over 5.5 weeks. The RT protocol was reported in detail elsewhere.17 After the introduction of preoperative CRT at our institution, we changed our policy as we moved from providing postoperative CRT toward providing preoperative CRT for clinical stage II or stage III rectal cancer. After a transition period, preoperative CRT was performed as the main treatment for locally advanced rectal cancer. The inclusion criterion for preoperative CRT was locally advanced rectal cancer located within 9 cm from the anal verge. Locally advanced rectal cancer was defined as
M. R. Hwang et al.
clinical T3/T4 or N? lesions, as determined by abdominal/ pelvic computed tomography and pelvic magnetic resonance imaging. Four regimens were used for preoperative CRT: (1) 5-fluorouracil and leucovorin [n = 341; two cycles of bolus infusion of 5-fluorouracil (400 mg/m2/day) and leucovorin (20 mg/m2/day) for 3 days during weeks 1 and 5 of RT]; (2) capecitabine [n = 134; oral administration of capecitabine (825 mg/m2) twice daily during RT without weekend breaks]; (3) capecitabine and irinotecan [n = 63; oral administration of capecitabine (825 mg/m2) twice daily during RT with weekend breaks and intravenous irinotecan (40 mg/m2/day) during each week of RT]; and (4) cetuximab, irinotecan, and capecitabine [n = 23; cetuximab (400 mg/m2/day) for 6 days before starting RT and 250 mg/m2/day on days 1, 8, 15, 22, and 29 during RT; irinotecan (40 mg/m2/day) on days 1, 8, 15, 22, and 29; and capecitabine (1,650 mg/m2/day) every Monday to Friday during RT]. At 4–8 weeks after completing preoperative CRT, all patients underwent radical proctectomy, including high ligation of the inferior mesenteric vessels and total mesorectal excision. The surgical procedures were low anterior resection (n = 455), Hartmann’s operation (n = 3), total proctocolectomy (n = 6), and abdominoperineal resection (n = 97). Regardless of the pathologic stage, four courses of postoperative maintenance chemotherapy were delivered starting 4 weeks after surgery. One of four adjuvant chemotherapeutic regimens was used: (1) 5-fluorouracil and leucovorin [four cycles of an intravenous bolus injection of 5-fluorouracil (400 mg/m2/day) and leucovorin (20 mg/m2/day) from days 1 to 5]; (2) capecitabine [six cycles of capecitabine (1,250 mg/m2) twice daily for 14 days, followed by 7 days of rest in each cycle]; (3) FOLFOX [eight cycles of oxaliplatin (85 mg/m2/day) on day 1, leucovorin (200 mg/m2/day) on day 1, an intravenous bolus of 5-fluorouracil (400 mg/m2/day) on day 1, and continuous infusion of 5-fluorouracil (2,400 mg/m2) for 46 h every 2 weeks]; or (4) FOLFIRI [eight cycles of irinotecan (180 mg/m2/day) on day 1, leucovorin (200 mg/ m2/day) on day 1, intravenous bolus of 5-fluorouracil (400 mg/m2/day) on day 1, and continuous infusion of 5-fluorouracil (2,400 mg/m2) for 46 h every 2 weeks]. Overall, 505 of 561 patients received postoperative chemotherapy. The other 56 (10.0 %) patients did not receive postoperative chemotherapy because the patient refused treatment (n = 35), because of postoperative complications (n = 2), or because of poor performance status (n = 19). For each patient, the chemotherapeutic regimens were selected according to the preferences of the medical oncologists or the patients or according to protocol-based clinical trials.
Circumferential Margin in Rectal Cancer
Histopathologic Assessment The surgical specimens were examined grossly and microscopically. CRM was determined as described by Quirke et al.1,18 The nonperitonealized surfaces of the specimen were painted with black ink, and the specimen was fixed in 10 % formaldehyde overnight. The whole tumor, including surrounding nonneoplastic tissue and the suspected original lesion, was sectioned (4 mm thick) and embedded. To determine the CRM, the shortest distance was measured from the primary tumor to the adjacent mesorectal fascia. If a lymph node or tumor deposit was located nearer to the mesorectal fascia than the primary tumor, it was used to measure CRM. The CRM was measured using a ruler or a microscope graticule. For the assessment of perirectal lymph nodes, mesorectal fat was removed after tumor sampling, and all fat was thoroughly dissected to identify the lymph nodes. In cases with fewer than 12 harvested lymph nodes, macroscopic and microscopic examination was repeated to search for additional lymph nodes. The tumors were classified according to the World Health Organization classification system and were staged according to the tumor-nodemetastasis classification system (6th version).19,20 The grade of tumor regression was histologically classified by using the regression system proposed by Dworak et al.21 Pathologic examinations and the measurement of CRM were performed by a pathologist (H.J.C.). Follow-up Patients were followed up every 3 months for the first 2 years, every 6 months for the next 3 years, and every 6 months or yearly thereafter. For this analysis we included data up to the last follow-up, which ended in February 2012. Digital rectal examination, determination of carcinoembryonic antigen (CEA), chest x-ray, and abdominal/ pelvic computed tomography were performed at each follow-up. Surveillance colonoscopy was performed at 1- to 3-year intervals unless contraindicated because of advanced age or performance status. Recurrence was pathologically diagnosed by surgical resection, biopsy or cytology, and/or the detection of radiologically apparent lesions that grew over time. Local recurrence was defined as recurrence in areas contiguous with the bed of the primary rectal resection or recurrence at the site of anastomosis. Disease-free survival time was defined as the time between surgery and any type of recurrence. Statistical Analysis The local recurrence and disease-free survival rates were calculated by using the Kaplan–Meier product–limit
1347
method, and differences were tested with the log-rank test. Univariate and multivariate Cox proportional hazards models were used to identify possible prognostic factors. Risk factors that showed at least borderline significance (P \ 0.06) in the univariate model were included in the multivariate model to assess the independent effects of CRM involvement on local recurrence and disease-free survival after adjustment for potential confounding factors. All tests were two-sided, and P-values of \0.05 were considered statistically significant. RESULTS Patients, Treatments, and Pathologic Characteristics The median age was 58 years (range 22–83 years), and 367 patients (65.4 %) were male. The pretreatment clinical stage was III for most patients (80.0 %). The posttreatment clinical stage was I in 243 patients (43.3 %), II in 156 patients (27.8 %), and III in 162 patients (28.9 %). The CRM was [2 mm in 487 patients (86.8 %; group 1), 1.1–2 mm in 36 patients (6.4 %; group 2), 0.1–1 mm in 20 patients (3.6 %; group 3), and 0 mm (cancer cells in the CRM) in 18 patients (3.2 %; group 4). The CRM was not significantly different among the cT and cN grades (P = 0.180 and P = 0.153, respectively; Supplementary Table 1). However, the CRM decreased significantly as the ypT and ypN grades increased (P \ 0.0001, \0.0001, respectively; Supplementary Table 1). The CRM was significantly shorter in patients who underwent abdominoperineal resection than in patients who underwent sphincter preservation surgery (5.9 ± 8.2 vs. 9.4 ± 8.4 mm; P = 0.001). The mean follow-up time was 57.1 months (range 7–114 months). Local recurrence was found in 57 patients (10.2 %). The median time to the detection of local recurrence was 20 months (range 1–68 months). The overall recurrence rate was 25.3 %, and the median observation time was 16 months (range 1–110 months). Local Recurrence and Disease-Free Survival Rates According to the CRM Figure 1 shows the cumulative local recurrence and disease-free survival rates for each CRM group. The local recurrence rate decreased significantly with increasing CRM (P \ 0.0001). The disease-free survival rate increased significantly with increasing CRM distance (P \ 0.0001). The local recurrence rate was higher in group 4 than in the other groups and was followed in order by group 3 and groups 2 and 1. The 5-year local recurrence rate was 10.1 % in group 1, 10.3 % in group 2, 24.4 % in group 3, and 63.0 % in group 4. The local recurrence rate was
1348
M. R. Hwang et al.
free survival between group 4 and the other groups were statistically significant. The disease-free survival rate was similar between groups 1 and 2 (P = 0.296).
(a) Cumulative local recurrence rate 1.00
Group 1: >2.0 Group 2: 1.1-2.0 Group 3: 0.1-1.0 Group 4: 0
0.9 0.8 0.7
Univariate Analysis and Multivariate Analysis for Prognosis
0.6 0.5 0.4 0.3 0.2 0.1 00
10
20
30
40
50
60
70
80
90
100
194 12 7 3
142 7 3 2
94 2 3 1
48 0 1 1
16 0 0 1
Month Number at risk 487 457 36 32 20 17 18 11
419 26 12 5
379 24 10 4
324 21 9 4
246 15 8 3
(b) Disease free survival rate 1.00
Group 1: >2.0 Group 2: 1.1-2.0 Group 3: 0.1-1.0 Group 4: 0
0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 00
10
20
30
40
50
60
70
80
90
100
193 12 6 3
142 7 3 2
94 2 3 1
48 0 1 1
16 0 0 1
Univariate analysis using the log-rank test revealed that age, pretreatment CEA, type of surgery, stage, tumor regression grade, histologic grade, and CRM were significant risk factors for local recurrence. Pretreatment CEA, tumor distance from the anal verge, type of surgery, stage, tumor regression grade, and CRM were significant risk factors for disease-free survival (Table 1). Multivariate analysis revealed that the CRM was an independent risk factor for local recurrence (Table 2). Increases in CRM decreased the risk of local recurrence [vs. CRM = 0 mm; CRM = 0.1–1.0 mm: hazard ratio (HR) 0.28, 95 % confidence interval (CI) 0.08–0.91, P = 0.035; CRM = 1.1–2.0 mm: HR 0.11, 95 % CI 0.03–0.46, P = 0.002; CRM [ 2 mm: HR 0.18, 95 % CI 0.08–0.42, P \ 0.0001]. Disease stage and type of surgery were also risk factors for local recurrence. Both disease stage and CRM were significant prognostic factors for disease-free survival. Disease-free survival increased with increases in CRM (vs. CRM = 0 mm; CRM = 0.1–1.0 mm: HR 0.32, 95 % CI 0.13–0.75, P = 0.009; CRM = 1.1–2.0 mm: HR 0.24, 95 % CI 0.10–0.54, P = 0.001; CRM [ 2.0 mm: HR 0.26, 95 % CI 0.14–0.48, P \ 0.0001).
Month Number at risk 487 453 32 36 20 17 10 18
413 26 11 5
376 24 10 4
322 21 8 4
245 15 7 3
FIG. 1 Kaplan–Meier curves of local recurrence and disease-free survival according to circumferential resection margin (group 1, CRM distance [2 mm; group 2, 1.1–2.0 mm; group 3, 0.1–1.0 mm; group 4, 0 mm). a Local recurrence. Group 1 versus 2, P = 0.838; group 1 versus 3, P = 0.002; group 1 versus 4, P \ 0.0001; group 2 versus 3, P = 0.075; group 2 versus 4, P \ 0.0001; group 3 versus 4, P = 0.071. b Disease-free survival. Group 1 versus 2, P = 0.296; group 1 versus 3, P = 0.004; group 1 versus 4, P \ 0.0001; group 2 versus 3, P = 0.175; group 2 versus 4, P \ 0.0001; group 3 versus 4, P = 0.027
significantly different between group 4 and groups 1 and 2. The difference between groups 3 and 4 was of borderline significance (P = 0.071). The local recurrence rate was similar between groups 1 and 2 (P = 0.838). The disease-free survival rate was much lower in group 4 than in the other groups. The 5-year disease-free survival rate was 77.7 % in group 1, 70.1 % in group 2, 51.8 % in group 3, and 22.2 % in group 4. The differences in disease-
DISCUSSION For irradiated rectal cancer, some studies defined CRM involvement as CRM B 1 or 2 mm.7,13,22,23 In one study, survival analysis was performed according to CRM in a cohort of heterogeneous patients with or without preoperative RT.2,14 To the best of our knowledge, survival analysis has yet to be performed according to CRM distance in patients who received preoperative CRT. Although the local recurrence rate was higher and the disease-free survival rate was lower in group 4 than in group 3 (0.1–1.0 mm), both groups exhibited a worse prognosis than patients with a CRM [ 1 mm [i.e., group 1 (CRM [ 2 mm) and group 2 (CRM 1–2 mm) combined] in terms of local recurrence (group 4 vs. groups 1 and 2, P \ 0.0001; group 3 vs. groups 1 and 2, P = 0.002) and disease-free survival (group 4 vs. groups 1 and 2, P \ 0.0001; group 3 vs. groups 1 and 2, P = 0.006). However, there were no significant differences in the local recurrence or disease-free survival rates between groups 1 and 2. Even in rectal cancer treated with
Circumferential Margin in Rectal Cancer
1349
TABLE 1 Univariate analysis of risk factors for local recurrence and disease-free survival in preoperative chemoradiotherapy Variable
No. of patients
Local recurrence 5-year
Age
P-value
Disease-free survival a
HR
95 % CI
5-year
0.039 311
14.6
1.00
[60 years
250
8.4
0.56
Male
367
13.7
1.00
Female
194
8.3
0.68
0.32–0.98
73.0
1.00
76.6
0.85
0.200
Pretreatment CEA (ng/ml) B5 [5
9.4
179
17.4
Tumor distance from anal verge
0.38–1.23
0.89
0.63–1.27
\0.0001 1.00 2.30
78.6 1.37–3.87
1.00
65.8
1.86
1.33–2.59
0.034
B5 cm
293
14.5
1.00
[5 cm
268
9.1
0.63
70.4 0.37–1.08
1.00
79.4
0.001
0.67
0.50–0.98
0.003
SSS
464
9.8
1.00
APR
97
21.9
2.42
Preoperative chemotherapy
0.61–1.19
1.00
75.6
0.092
Type of surgery
95 % CI
0.514 74.0
0.001 382
HR
0.348
B60 years Sex
P-valuea
76.9 1.38–4.23
1.00
63.4
0.343
1.77
1.22–2.58
0.227
5-Fluorouracil and leucovorin
341
14.8
1.00
70.8
1.00
Capecitabine
134
7.4
0.55
0.27–1.10
80.7
0.70
0.46–1.06
Capecitabine ? irinotecan
63
11.0
0.70
0.30–1.66
76.1
0.85
0.50–1.44
Capecitabine ? irinotecan ? cetuximab
23
8.9
0.72
0.17–2.97
87.0
0.47
0.15–1.47
Postoperative chemotherapy None Pyrimidine
0.691
0.107
56 480
14.3 11.4
1.00 0.71
0.32–1.57
80.8 74.6
1.00 1.17
0.63–2.17
Pyrimidine ? oxaliplatin
24
15.8
1.16
0.30–4.50
62.2
2.45
1.04–5.77
Pyrimidine ? irinotecan
1
0
–
100
\0.0001
yP stage
– \0.0001
0?1
243
5.6
1.00
86.4
1.00
2
156
10.1
2.32
1.05–5.12
75.5
1.98
1.23–3.16
3
162
24.9
5.91
2.97–11.73
55.6
4.38
2.89–6.63
1
69
23.9
1.00
2
252
13.6
0.40
0.23–0.71
72.7
0.48
0.33–0.70
3
71
3.5
0.07
0.02–0.32
92.8
0.15
0.07–0.32
64
1.8
0.13
0.04–0.42
77.2
0.27
0.14–0.50
WD/MD
530
11.4
1.00
PD/SRC
26
24.2
2.59
Tumor regression grade (Dworak’s)
4 Histologic grade
b
CRM 0 mm 0.1–1.0 mm 1.1–2.0 mm [2.0 mm
\0.0001
0.001 48.2
0.022
1.00
0.163 75.7 1.11–6.03
1.00
60.4
\0.0001
1.58
0.83–3.01
\0.0001
18
63.0
1.00
22.2
1.00
20
24.4
0.33
0.11–1.01
51.8
0.36
0.16–0.84
36
10.3
0.10
0.03–0.36
70.1
0.20
0.09–0.45
487
10.1
0.09
0.04–0.18
77.7
0.14
0.08–0.25
CEA carcinoembryonic antigen, SSS sphincter-saving surgery, APR abdominoperineal resection, WD well differentiated, MD moderately differentiated, PD poorly differentiated, SRC signet ring cell, CRM circumferential resection margin, HR hazard ratio, CI confidence interval a
Log-rank test
b
Missing value (n = 5)
1350
M. R. Hwang et al.
TABLE 2 Multivariate survival analysis (Cox proportional hazards model) of risk factors for local recurrence and disease-free survival in preoperative chemoradiotherapy Variable
Local recurrence HR
95 % CI
Disease-free survival P value HR
95 % CI
P-value
Age B60 years
1.00
[60 years
0.67 0.38–1.20
0.180
Pretreatment CEA B5 ng/ml
1.00
[5 ng/ml
1.49 0.85–2.63
1.00 0.167
1.37 0.97–1.94
0.073
Tumor distance from anal verge B5 cm
1.00
[5 cm
0.88 0.61–1.27
0.480
Type of surgery SSS
1.00
APR
1.91 1.03–3.54
yP stage
1.00 0.041
1.43 0.94–2.18
0.097 \0.0001
0.010
0?1 2
1.00 1.15 0.46–2.90
0.767
1.00 1.46 0.81–2.62
3
2.67 1.14–6.27
0.024
3.21 1.85–5.58 \0.0001
Tumor regression grade (Dworak’s) 1 1.00
0.168
0.207 0.177
1.00
2
0.62 0.33–1.18
0.145
0.78 0.52–1.18
0.239
3
0.21 0.04–0.97
0.045
0.42 0.18–0.97
0.041
4
0.35 0.08–1.53
0.164
0.88 0.39–2.00
0.758
Histologic grade
a
WD/MD
1.00
PD/SRC
1.19 0.43–3.31
CRM 0 mm
0.736 \0.0001
0.001 1.00
1.00
0.1–1.0 mm 0.28 0.08–0.91
0.035
0.32 0.13–0.75
0.009
1.1–2.0 mm 0.11 0.03–0.46
0.002
0.24 0.10–0.54
0.001
[2 mm
0.18 0.08–0.42 \0.0001 0.26 0.14–0.48 \0.0001
CEA carcinoembryonic antigen, SSS sphincter-saving surgery, APR abdominoperineal resection, WD well differentiated, MD moderately differentiated, PD poorly differentiated, SRC signet ring cell, CRM circumferential resection margin, HR hazard ratio, CI confidence interval a
Missing value (n = 5)
preoperative CRT, 1 mm may be considered as a uniform cutoff value for CRM involvement. In the present study, the local recurrence and diseasefree survival curves for group 2 (1.1–2.0 mm) were closer to those for group 1 ([2.0 mm) than those for group 3 (0.1–1.0 mm). In multivariate survival analysis, the HRs for group 2 were comparable with those for group 1 in terms of local recurrence and disease-free survival
(Table 2). A CRM of 1–2 mm after preoperative CRT did not increase the risk of a poor outcome relative to a CRM [ 2 mm. Consistent with previous studies, CRM was an independent prognostic factor for both local recurrence and disease-free survival in the present study. The results showed that the recurrence rate differed according to CRM and was greater in group 4 (0 mm) than in group 3 (0.1–1.0 mm). These results may reflect the effects of preoperative CRT on tumor margins. Preoperative CRT may sterilize tumor cells along the tumor border and provide a safer margin than in the absence of preoperative CRT. However, the precise biological mechanism by which CRT affects the tumor margin remains unclear. Earlier studies showed that short-course RT is not related to the rates of downstaging or CRM involvement.22,24,25 However, long-course CRT was reported to lower the rate of CRM involvement by enhancing tumor regression.26 It was reported that the CRM involvement rate was lower after long-course CRT than after shortcourse RT.9,22,26,27 In short-course RT, the cutoff values for CRM involvement that were associated with prognosis may differ from those of long-course CRT. Further studies are needed to determine the prognostic value of the CRM after short-course RT. In earlier studies of preoperative therapy, the CRM involvement rate ranged from 4 to 18 %.26–29 Consistent with earlier studies, the CRM involvement rate after preoperative CRT was 6.8 % in the present study. The skill of the surgeon is critical for avoiding CRM involvement, primarily by performing meticulous rectal dissection, because incomplete mesorectal excision can lead to CRM involvement. Even after complete mesorectal excision, CRM involvement may occur after invasion of an advanced tumor into the mesorectal fascia. Radial invasion of the tumor and the quality of mesorectal excision can affect CRM status. Preoperative CRT may achieve safer circumferential margins than the absence of preoperative treatment in patients with rectal cancer. There are some limitations to our study. The principal limitation of our study is its retrospective design. We did not calculate the sample size needed to ensure adequate power to detect a prognostic effect of CRM. A second limitation is the small number of patients with a short CRM. Very few studies have investigated the prognosis of rectal cancer according to the CRM. We believe that our data provide an important reference for risk stratification after preoperative CRT. Therefore, further studies with a prospective design and including a larger number of patients are needed to verify our results. In conclusion, our results indicate that CRM distance is an independent risk factor for local recurrence and diseasefree survival in patients with rectal cancer treated with
Circumferential Margin in Rectal Cancer
preoperative CRT. In this setting, CRM distance provides useful information for risk stratification in the recurrence of rectal cancer. ACKNOWLEDGMENT This study was supported by a grant from the National Cancer Center of Korea (NCC-0910160) and by the Converging Research Center Program funded by the Ministry of Education, Science and Technology (project 1131150). CONFLICT OF INTEREST conflicts of interest.
The authors indicated no potential
REFERENCES 1. Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection. Histopathological study of lateral tumour spread and surgical excision. Lancet. 1986;2:996–99. 2. Birbeck KF, Macklin CP, Tiffin NJ, et al. Rates of circumferential resection margin involvement vary between surgeons and predict outcomes in rectal cancer surgery. Ann Surg. 2002;235: 449–57. 3. Wibe A, Rendedal PR, Svensson E, et al. Prognostic significance of the circumferential resection margin following total mesorectal excision for rectal cancer. Br J Surg. 2002;89:327–34. 4. Rullier E, Goffre B, Bonnel C, Zerbib F, Caudry M, Saric J. Preoperative radiochemotherapy and sphincter-saving resection for T3 carcinomas of the lower third of the rectum. Ann Surg. 2001;234:633–40. 5. den Dulk M, Collette L, van de Velde CJ, et al. Quality of surgery in T3-4 rectal cancer: involvement of circumferential resection margin not influenced by preoperative treatment. Results from EORTC trial 22921. Eur J Cancer. 2007;43:1821–8. 6. Nagtegaal ID, Quirke P. What is the role for the circumferential margin in the modern treatment of rectal cancer? J Clin Oncol. 2008;26:303–12. 7. Gosens MJ, Klaassen RA, Tan-Go I, et al. Circumferential margin involvement is the crucial prognostic factor after multimodality treatment in patients with locally advanced rectal carcinoma. Clin Cancer Res. 2007;13:6617–23. 8. Adam IJ, Mohamdee MO, Martin IG, et al. Role of circumferential margin involvement in the local recurrence of rectal cancer. Lancet. 1994;344:707–11. 9. Nagtegaal ID, Marijnen CA, Kranenbarg EK, van de Velde CJ, van Krieken JH. Circumferential margin involvement is still an important predictor of local recurrence in rectal carcinoma: not one millimeter but two millimeters is the limit. Am J Surg Pathol. 2002;26:350–7. 10. Glynne-Jones R, Mawdsley S, Novell JR. The clinical significance of the circumferential resection margin following preoperative pelvic chemo-radiotherapy in rectal cancer: why we need a common language. Colorectal Dis. 2006;8:800–7. 11. de Haas-Kock DF, Baeten CG, Jager JJ, et al. Prognostic significance of radial margins of clearance in rectal cancer. Br J Surg. 1996;83:781–5. 12. Tilney HS, Rasheed S, Northover JM, Tekkis PP. The influence of circumferential resection margins on long-term outcomes following rectal cancer surgery. Dis Colon Rectum. 2009;52: 1723–9. 13. Luna-Perez P, Bustos-Cholico E, Alvarado I, et al. Prognostic significance of circumferential margin involvement in rectal adenocarcinoma treated with preoperative chemoradiotherapy and low anterior resection. J Surg Oncol. 2005;90:20–5.
1351 14. Wittekind C, Compton C, Quirke P, et al. A uniform residual tumor (R) classification: integration of the R classification and the circumferential margin status. Cancer. 2009;115:3483–8. 15. Dent OF, Haboubi N, Chapuis PH, et al. Assessing the evidence for an association between circumferential tumour clearance and local recurrence after resection of rectal cancer. Colorectal Dis. 2007;9:112–21. 16. Hwang MR, Park JW, Kim DY, et al. Prognostic impact of peritonealisation in rectal cancer treated with preoperative chemoradiotherapy: extraperitoneal versus intraperitoneal rectal cancer. Radiother Oncol. 2010;94:353–8. 17. Kim DY, Jung KH, Kim TH, et al. Comparison of 5-fluorouracil/ leucovorin and capecitabine in preoperative chemoradiotherapy for locally advanced rectal cancer. Int J Radiat Oncol Biol Phys. 2007;67:378–84. 18. Quirke P, Dixon MF. The prediction of local recurrence in rectal adenocarcinoma by histopathological examination. Int J Colorectal Dis. 1988;3:127–31. 19. Greene FL, Page DL, Fleming ID, et al. American Joint Committee on Cancer staging manual. 6th ed. New York: Springer; 2002. 20. Hamilton SR, Vogelstein B, Kudo S, et al. Carcinoma of the colon and rectum. In: Hamilton SR, Aaltonen LA editors. World Health Organization classification of tumours. Pathology and genetics of tumours of the digestive system. Lyon: IARC Press, Lyon; 2000. p. 104–19. 21. Dworak O, Keilholz L, Hoffmann A. Pathological features of rectal cancer after preoperative radiochemotherapy. Int J Colorectal Dis. 1997;12:19–23. 22. Marijnen CA, Nagtegaal ID, Kapiteijn E, et al. Radiotherapy does not compensate for positive resection margins in rectal cancer patients: report of a multicenter randomized trial. Int J Radiat Oncol Biol Phys. 2003;55:1311–20. 23. Mawdsley S, Glynne-Jones R, Grainger J, et al. Can histopathologic assessment of circumferential margin after preoperative pelvic chemoradiotherapy for T3-T4 rectal cancer predict for 3-year disease-free survival? Int J Radiat Oncol Biol Phys. 2005; 63:745–52. 24. Marijnen CA, Nagtegaal ID, Klein Kranenbarg E, et al. No downstaging after short-term preoperative radiotherapy in rectal cancer patients. J Clin Oncol. 2001;19:1976–84. 25. Sebag-Montefiore D, Stephens RJ, Steele R, et al. Preoperative radiotherapy versus selective postoperative chemoradiotherapy in patients with rectal cancer (MRC CR07 and NCIC-CTG C016): a multicentre, randomised trial. Lancet. 2009;373:811–20. 26. Bujko K, Nowacki MP, Nasierowska-Guttmejer A, et al. Sphincter preservation following preoperative radiotherapy for rectal cancer: report of a randomised trial comparing short-term radiotherapy vs. conventionally fractionated radiochemotherapy. Radiother Oncol. 2004;72:15–24. 27. Bosset JF, Calais G, Mineur L, et al. Enhanced tumorocidal effect of chemotherapy with preoperative radiotherapy for rectal cancer: preliminary results—EORTC 22921. J Clin Oncol. 2005;23: 5620–7. 28. Wheeler JM, Dodds E, Warren BF, et al. Preoperative chemoradiotherapy and total mesorectal excision surgery for locally advanced rectal cancer: correlation with rectal cancer regression grade. Dis Colon Rectum. 2004;47:2025–31. 29. Rullier A, Laurent C, Vendrely V, Le Bail B, Bioulac-Sage P, Rullier E. Impact of colloid response on survival after preoperative radiotherapy in locally advanced rectal carcinoma. Am J Surg Pathol. 2005;29:602–6.
ORIGINAL ARTICLE – COLORECTAL CANCER
Prognostic Impact of Circumferential Resection Margin in Rectal Cancer Treated with Preoperative Chemoradiotherapy Mi Ri Hwang, MD1, Ji Won Park, MD1,4,5,6, Sohee Park, PhD2, Hyekyoung Yoon, MS3, Dae Yong Kim, MD1, Hee Jin Chang, MD1, Sun Young Kim, MD1, Sung Chan Park, MD1, Hyo Seong Choi, MD1, Jae Hwan Oh, MD1, and Seung-Yong Jeong, MD4,5,6 1
Center for Colorectal Cancer, Research Institute & Hospital, National Cancer Center, Goyang, Gyeonggi-do, Korea; Department of Epidemiology and Heath Promotion, Graduate School of Public Health, Yonsei University, Seoul, Korea; 3 Cancer Biostatistics Branch, Research Institute, National Cancer Center, Goyang, Korea; 4Department of Surgery, Seoul National University College of Medicine, Seoul, Korea; 5Cancer Research Institute, Seoul National University, Seoul, Korea; 6Colorectal Cancer Center, Seoul National University Cancer Hospital, Seoul, Korea 2
ABSTRACT Background. The circumferential resection margin (CRM) is a strong prognostic factor in rectal cancer. The purpose of this study was to investigate the relationship between CRM distance and recurrence in patients with locally advanced rectal cancer who received preoperative chemoradiotherapy (CRT). Methods. We analyzed data for 561 patients who underwent preoperative CRT and curative surgery for locally advanced rectal cancer between August 2001 and December 2008. CRM was divided into four groups: group 1, CRM [ 2 mm; group 2, 1.1–2.0 mm; group 3, 0.1–1.0 mm; and group 4, 0 mm. We assessed the associations of CRM with local recurrence and disease-free survival. Results. Groups 1, 2, 3, and 4 comprised 487, 36, 20, and 18 patients, respectively. The local recurrence rate was highest and the disease-free survival rate was lowest in group 4, followed by groups 3, 2, and 1. Survival was similar between groups 2 and 1. Local recurrence rates were lower in groups 3, 2, and 1 than in group 4 [hazard ratio (HR) 0.28, 95 % confidence interval (CI) 0.09–0.91,
Electronic supplementary material The online version of this article (doi:10.1245/s10434-014-3484-1) contains supplementary material, which is available to authorized users. Ó Society of Surgical Oncology 2014 First Received: 2 August 2012; Published Online: 28 January 2014 J. W. Park, MD e-mail: [email protected]
P = 0.035; HR 0.11, 95 % CI 0.03–0.46, P = 0.002; HR 0.18, 95 % CI 0.08–0.42, P \ 0.0001, respectively]. Disease-free survival rates were higher in groups 3, 2, and 1 than in group 4 (HR 0.32, 95 % CI 0.13–0.75, P = 0.009; HR 0.24, 95 % CI 0.10–0.54, P = 0.001; HR 0.26, 95 % CI 0.14–0.48, P \ 0.0001, respectively). Conclusions. After preoperative CRT, CRM distance provides useful information for risk stratification in the recurrence of rectal cancer. Since the initial report of the prognostic importance of the circumferential resection margin (CRM) in rectal cancer, several studies have shown that CRM is a strong predictor of local recurrence, distant metastasis, and survival.1–3 Although preoperative chemoradiotherapy (CRT) is widely used for the treatment of locally advanced rectal cancer, its effects on the CRM remain unclear. In a downstaged patient group of one study, preoperative CRT was associated with an increase in the mean CRM because of tumor shrinkage, enabling curative surgery and sphincter preservation in patients with lower rectal cancer.4 However, another study demonstrated that preoperative radiotherapy (RT) did not compensate for CRM involvement.5 It was also reported that CRM was a strong predictor of local recurrence in patients who received preoperative CRT.6,7 The cutoff value for CRM involvement was 1 mm in many studies, although other studies used a cutoff value of 2 mm.2,3,8–13 Prior studies showed that a CRM of B1 mm was associated with high local recurrence and low survival rates, similar to a CRM of 0 mm, and the two have been grouped together as positive CRM.3,8 Therefore, a recent
1346
study recommended a CRM of B1 mm as a uniform cutoff value for CRM involvement.14 However, a pooled analysis suggested that a CRM of 0 mm is a more reasonable cutoff value for CRM involvement.15 The definition of CRM involvement for rectal cancer patients has yet to be firmly established. In particular, there are scant data regarding the prognostic ability of CRM distance in patients who have received preoperative CRT. Therefore, in the present study, we analyzed the prognostic ability of CRM distance in patients with locally advanced rectal cancer who had received preoperative CRT. METHODS Patients Between August 2001 and December 2008, a total of 655 patients with locally advanced rectal cancer underwent preoperative CRT at the National Cancer Center, Republic of Korea. Of the 655 patients, 35 refused surgery, and 17 with comorbidities or who rejected anal ablation were treated by local excision. Because CRM is not applicable to intraperitoneal cancer, 31 patients with intraperitoneal rectal cancer were also excluded. Intraperitoneal cancer was defined as cancers that were covered by the peritoneum in pathologic examination, as described in our previous report.16 Eleven patients were lost to follow-up within 6 months. After excluding these patients, we analyzed 561 patients in this study. Patients were divided into 4 groups according to CRM distance stated in the pathologic report: group 1, CRM [ 2 mm; group 2, 1.1–2.0 mm; group 3, 0.1–1.0 mm; and group 4, 0 mm. The study was performed in accordance with the guidelines of the Institutional Review Board of the National Cancer Center. This retrospective analysis was performed using data that were prospectively recorded in a database. Preoperative CRT RT was delivered to the whole pelvis at a dose of 45 Gy in 25 fractions, followed by a boost to the primary tumor of 5.4 Gy in three fractions over 5.5 weeks. The RT protocol was reported in detail elsewhere.17 After the introduction of preoperative CRT at our institution, we changed our policy as we moved from providing postoperative CRT toward providing preoperative CRT for clinical stage II or stage III rectal cancer. After a transition period, preoperative CRT was performed as the main treatment for locally advanced rectal cancer. The inclusion criterion for preoperative CRT was locally advanced rectal cancer located within 9 cm from the anal verge. Locally advanced rectal cancer was defined as
M. R. Hwang et al.
clinical T3/T4 or N? lesions, as determined by abdominal/ pelvic computed tomography and pelvic magnetic resonance imaging. Four regimens were used for preoperative CRT: (1) 5-fluorouracil and leucovorin [n = 341; two cycles of bolus infusion of 5-fluorouracil (400 mg/m2/day) and leucovorin (20 mg/m2/day) for 3 days during weeks 1 and 5 of RT]; (2) capecitabine [n = 134; oral administration of capecitabine (825 mg/m2) twice daily during RT without weekend breaks]; (3) capecitabine and irinotecan [n = 63; oral administration of capecitabine (825 mg/m2) twice daily during RT with weekend breaks and intravenous irinotecan (40 mg/m2/day) during each week of RT]; and (4) cetuximab, irinotecan, and capecitabine [n = 23; cetuximab (400 mg/m2/day) for 6 days before starting RT and 250 mg/m2/day on days 1, 8, 15, 22, and 29 during RT; irinotecan (40 mg/m2/day) on days 1, 8, 15, 22, and 29; and capecitabine (1,650 mg/m2/day) every Monday to Friday during RT]. At 4–8 weeks after completing preoperative CRT, all patients underwent radical proctectomy, including high ligation of the inferior mesenteric vessels and total mesorectal excision. The surgical procedures were low anterior resection (n = 455), Hartmann’s operation (n = 3), total proctocolectomy (n = 6), and abdominoperineal resection (n = 97). Regardless of the pathologic stage, four courses of postoperative maintenance chemotherapy were delivered starting 4 weeks after surgery. One of four adjuvant chemotherapeutic regimens was used: (1) 5-fluorouracil and leucovorin [four cycles of an intravenous bolus injection of 5-fluorouracil (400 mg/m2/day) and leucovorin (20 mg/m2/day) from days 1 to 5]; (2) capecitabine [six cycles of capecitabine (1,250 mg/m2) twice daily for 14 days, followed by 7 days of rest in each cycle]; (3) FOLFOX [eight cycles of oxaliplatin (85 mg/m2/day) on day 1, leucovorin (200 mg/m2/day) on day 1, an intravenous bolus of 5-fluorouracil (400 mg/m2/day) on day 1, and continuous infusion of 5-fluorouracil (2,400 mg/m2) for 46 h every 2 weeks]; or (4) FOLFIRI [eight cycles of irinotecan (180 mg/m2/day) on day 1, leucovorin (200 mg/ m2/day) on day 1, intravenous bolus of 5-fluorouracil (400 mg/m2/day) on day 1, and continuous infusion of 5-fluorouracil (2,400 mg/m2) for 46 h every 2 weeks]. Overall, 505 of 561 patients received postoperative chemotherapy. The other 56 (10.0 %) patients did not receive postoperative chemotherapy because the patient refused treatment (n = 35), because of postoperative complications (n = 2), or because of poor performance status (n = 19). For each patient, the chemotherapeutic regimens were selected according to the preferences of the medical oncologists or the patients or according to protocol-based clinical trials.
Circumferential Margin in Rectal Cancer
Histopathologic Assessment The surgical specimens were examined grossly and microscopically. CRM was determined as described by Quirke et al.1,18 The nonperitonealized surfaces of the specimen were painted with black ink, and the specimen was fixed in 10 % formaldehyde overnight. The whole tumor, including surrounding nonneoplastic tissue and the suspected original lesion, was sectioned (4 mm thick) and embedded. To determine the CRM, the shortest distance was measured from the primary tumor to the adjacent mesorectal fascia. If a lymph node or tumor deposit was located nearer to the mesorectal fascia than the primary tumor, it was used to measure CRM. The CRM was measured using a ruler or a microscope graticule. For the assessment of perirectal lymph nodes, mesorectal fat was removed after tumor sampling, and all fat was thoroughly dissected to identify the lymph nodes. In cases with fewer than 12 harvested lymph nodes, macroscopic and microscopic examination was repeated to search for additional lymph nodes. The tumors were classified according to the World Health Organization classification system and were staged according to the tumor-nodemetastasis classification system (6th version).19,20 The grade of tumor regression was histologically classified by using the regression system proposed by Dworak et al.21 Pathologic examinations and the measurement of CRM were performed by a pathologist (H.J.C.). Follow-up Patients were followed up every 3 months for the first 2 years, every 6 months for the next 3 years, and every 6 months or yearly thereafter. For this analysis we included data up to the last follow-up, which ended in February 2012. Digital rectal examination, determination of carcinoembryonic antigen (CEA), chest x-ray, and abdominal/ pelvic computed tomography were performed at each follow-up. Surveillance colonoscopy was performed at 1- to 3-year intervals unless contraindicated because of advanced age or performance status. Recurrence was pathologically diagnosed by surgical resection, biopsy or cytology, and/or the detection of radiologically apparent lesions that grew over time. Local recurrence was defined as recurrence in areas contiguous with the bed of the primary rectal resection or recurrence at the site of anastomosis. Disease-free survival time was defined as the time between surgery and any type of recurrence. Statistical Analysis The local recurrence and disease-free survival rates were calculated by using the Kaplan–Meier product–limit
1347
method, and differences were tested with the log-rank test. Univariate and multivariate Cox proportional hazards models were used to identify possible prognostic factors. Risk factors that showed at least borderline significance (P \ 0.06) in the univariate model were included in the multivariate model to assess the independent effects of CRM involvement on local recurrence and disease-free survival after adjustment for potential confounding factors. All tests were two-sided, and P-values of \0.05 were considered statistically significant. RESULTS Patients, Treatments, and Pathologic Characteristics The median age was 58 years (range 22–83 years), and 367 patients (65.4 %) were male. The pretreatment clinical stage was III for most patients (80.0 %). The posttreatment clinical stage was I in 243 patients (43.3 %), II in 156 patients (27.8 %), and III in 162 patients (28.9 %). The CRM was [2 mm in 487 patients (86.8 %; group 1), 1.1–2 mm in 36 patients (6.4 %; group 2), 0.1–1 mm in 20 patients (3.6 %; group 3), and 0 mm (cancer cells in the CRM) in 18 patients (3.2 %; group 4). The CRM was not significantly different among the cT and cN grades (P = 0.180 and P = 0.153, respectively; Supplementary Table 1). However, the CRM decreased significantly as the ypT and ypN grades increased (P \ 0.0001, \0.0001, respectively; Supplementary Table 1). The CRM was significantly shorter in patients who underwent abdominoperineal resection than in patients who underwent sphincter preservation surgery (5.9 ± 8.2 vs. 9.4 ± 8.4 mm; P = 0.001). The mean follow-up time was 57.1 months (range 7–114 months). Local recurrence was found in 57 patients (10.2 %). The median time to the detection of local recurrence was 20 months (range 1–68 months). The overall recurrence rate was 25.3 %, and the median observation time was 16 months (range 1–110 months). Local Recurrence and Disease-Free Survival Rates According to the CRM Figure 1 shows the cumulative local recurrence and disease-free survival rates for each CRM group. The local recurrence rate decreased significantly with increasing CRM (P \ 0.0001). The disease-free survival rate increased significantly with increasing CRM distance (P \ 0.0001). The local recurrence rate was higher in group 4 than in the other groups and was followed in order by group 3 and groups 2 and 1. The 5-year local recurrence rate was 10.1 % in group 1, 10.3 % in group 2, 24.4 % in group 3, and 63.0 % in group 4. The local recurrence rate was
1348
M. R. Hwang et al.
free survival between group 4 and the other groups were statistically significant. The disease-free survival rate was similar between groups 1 and 2 (P = 0.296).
(a) Cumulative local recurrence rate 1.00
Group 1: >2.0 Group 2: 1.1-2.0 Group 3: 0.1-1.0 Group 4: 0
0.9 0.8 0.7
Univariate Analysis and Multivariate Analysis for Prognosis
0.6 0.5 0.4 0.3 0.2 0.1 00
10
20
30
40
50
60
70
80
90
100
194 12 7 3
142 7 3 2
94 2 3 1
48 0 1 1
16 0 0 1
Month Number at risk 487 457 36 32 20 17 18 11
419 26 12 5
379 24 10 4
324 21 9 4
246 15 8 3
(b) Disease free survival rate 1.00
Group 1: >2.0 Group 2: 1.1-2.0 Group 3: 0.1-1.0 Group 4: 0
0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 00
10
20
30
40
50
60
70
80
90
100
193 12 6 3
142 7 3 2
94 2 3 1
48 0 1 1
16 0 0 1
Univariate analysis using the log-rank test revealed that age, pretreatment CEA, type of surgery, stage, tumor regression grade, histologic grade, and CRM were significant risk factors for local recurrence. Pretreatment CEA, tumor distance from the anal verge, type of surgery, stage, tumor regression grade, and CRM were significant risk factors for disease-free survival (Table 1). Multivariate analysis revealed that the CRM was an independent risk factor for local recurrence (Table 2). Increases in CRM decreased the risk of local recurrence [vs. CRM = 0 mm; CRM = 0.1–1.0 mm: hazard ratio (HR) 0.28, 95 % confidence interval (CI) 0.08–0.91, P = 0.035; CRM = 1.1–2.0 mm: HR 0.11, 95 % CI 0.03–0.46, P = 0.002; CRM [ 2 mm: HR 0.18, 95 % CI 0.08–0.42, P \ 0.0001]. Disease stage and type of surgery were also risk factors for local recurrence. Both disease stage and CRM were significant prognostic factors for disease-free survival. Disease-free survival increased with increases in CRM (vs. CRM = 0 mm; CRM = 0.1–1.0 mm: HR 0.32, 95 % CI 0.13–0.75, P = 0.009; CRM = 1.1–2.0 mm: HR 0.24, 95 % CI 0.10–0.54, P = 0.001; CRM [ 2.0 mm: HR 0.26, 95 % CI 0.14–0.48, P \ 0.0001).
Month Number at risk 487 453 32 36 20 17 10 18
413 26 11 5
376 24 10 4
322 21 8 4
245 15 7 3
FIG. 1 Kaplan–Meier curves of local recurrence and disease-free survival according to circumferential resection margin (group 1, CRM distance [2 mm; group 2, 1.1–2.0 mm; group 3, 0.1–1.0 mm; group 4, 0 mm). a Local recurrence. Group 1 versus 2, P = 0.838; group 1 versus 3, P = 0.002; group 1 versus 4, P \ 0.0001; group 2 versus 3, P = 0.075; group 2 versus 4, P \ 0.0001; group 3 versus 4, P = 0.071. b Disease-free survival. Group 1 versus 2, P = 0.296; group 1 versus 3, P = 0.004; group 1 versus 4, P \ 0.0001; group 2 versus 3, P = 0.175; group 2 versus 4, P \ 0.0001; group 3 versus 4, P = 0.027
significantly different between group 4 and groups 1 and 2. The difference between groups 3 and 4 was of borderline significance (P = 0.071). The local recurrence rate was similar between groups 1 and 2 (P = 0.838). The disease-free survival rate was much lower in group 4 than in the other groups. The 5-year disease-free survival rate was 77.7 % in group 1, 70.1 % in group 2, 51.8 % in group 3, and 22.2 % in group 4. The differences in disease-
DISCUSSION For irradiated rectal cancer, some studies defined CRM involvement as CRM B 1 or 2 mm.7,13,22,23 In one study, survival analysis was performed according to CRM in a cohort of heterogeneous patients with or without preoperative RT.2,14 To the best of our knowledge, survival analysis has yet to be performed according to CRM distance in patients who received preoperative CRT. Although the local recurrence rate was higher and the disease-free survival rate was lower in group 4 than in group 3 (0.1–1.0 mm), both groups exhibited a worse prognosis than patients with a CRM [ 1 mm [i.e., group 1 (CRM [ 2 mm) and group 2 (CRM 1–2 mm) combined] in terms of local recurrence (group 4 vs. groups 1 and 2, P \ 0.0001; group 3 vs. groups 1 and 2, P = 0.002) and disease-free survival (group 4 vs. groups 1 and 2, P \ 0.0001; group 3 vs. groups 1 and 2, P = 0.006). However, there were no significant differences in the local recurrence or disease-free survival rates between groups 1 and 2. Even in rectal cancer treated with
Circumferential Margin in Rectal Cancer
1349
TABLE 1 Univariate analysis of risk factors for local recurrence and disease-free survival in preoperative chemoradiotherapy Variable
No. of patients
Local recurrence 5-year
Age
P-value
Disease-free survival a
HR
95 % CI
5-year
0.039 311
14.6
1.00
[60 years
250
8.4
0.56
Male
367
13.7
1.00
Female
194
8.3
0.68
0.32–0.98
73.0
1.00
76.6
0.85
0.200
Pretreatment CEA (ng/ml) B5 [5
9.4
179
17.4
Tumor distance from anal verge
0.38–1.23
0.89
0.63–1.27
\0.0001 1.00 2.30
78.6 1.37–3.87
1.00
65.8
1.86
1.33–2.59
0.034
B5 cm
293
14.5
1.00
[5 cm
268
9.1
0.63
70.4 0.37–1.08
1.00
79.4
0.001
0.67
0.50–0.98
0.003
SSS
464
9.8
1.00
APR
97
21.9
2.42
Preoperative chemotherapy
0.61–1.19
1.00
75.6
0.092
Type of surgery
95 % CI
0.514 74.0
0.001 382
HR
0.348
B60 years Sex
P-valuea
76.9 1.38–4.23
1.00
63.4
0.343
1.77
1.22–2.58
0.227
5-Fluorouracil and leucovorin
341
14.8
1.00
70.8
1.00
Capecitabine
134
7.4
0.55
0.27–1.10
80.7
0.70
0.46–1.06
Capecitabine ? irinotecan
63
11.0
0.70
0.30–1.66
76.1
0.85
0.50–1.44
Capecitabine ? irinotecan ? cetuximab
23
8.9
0.72
0.17–2.97
87.0
0.47
0.15–1.47
Postoperative chemotherapy None Pyrimidine
0.691
0.107
56 480
14.3 11.4
1.00 0.71
0.32–1.57
80.8 74.6
1.00 1.17
0.63–2.17
Pyrimidine ? oxaliplatin
24
15.8
1.16
0.30–4.50
62.2
2.45
1.04–5.77
Pyrimidine ? irinotecan
1
0
–
100
\0.0001
yP stage
– \0.0001
0?1
243
5.6
1.00
86.4
1.00
2
156
10.1
2.32
1.05–5.12
75.5
1.98
1.23–3.16
3
162
24.9
5.91
2.97–11.73
55.6
4.38
2.89–6.63
1
69
23.9
1.00
2
252
13.6
0.40
0.23–0.71
72.7
0.48
0.33–0.70
3
71
3.5
0.07
0.02–0.32
92.8
0.15
0.07–0.32
64
1.8
0.13
0.04–0.42
77.2
0.27
0.14–0.50
WD/MD
530
11.4
1.00
PD/SRC
26
24.2
2.59
Tumor regression grade (Dworak’s)
4 Histologic grade
b
CRM 0 mm 0.1–1.0 mm 1.1–2.0 mm [2.0 mm
\0.0001
0.001 48.2
0.022
1.00
0.163 75.7 1.11–6.03
1.00
60.4
\0.0001
1.58
0.83–3.01
\0.0001
18
63.0
1.00
22.2
1.00
20
24.4
0.33
0.11–1.01
51.8
0.36
0.16–0.84
36
10.3
0.10
0.03–0.36
70.1
0.20
0.09–0.45
487
10.1
0.09
0.04–0.18
77.7
0.14
0.08–0.25
CEA carcinoembryonic antigen, SSS sphincter-saving surgery, APR abdominoperineal resection, WD well differentiated, MD moderately differentiated, PD poorly differentiated, SRC signet ring cell, CRM circumferential resection margin, HR hazard ratio, CI confidence interval a
Log-rank test
b
Missing value (n = 5)
1350
M. R. Hwang et al.
TABLE 2 Multivariate survival analysis (Cox proportional hazards model) of risk factors for local recurrence and disease-free survival in preoperative chemoradiotherapy Variable
Local recurrence HR
95 % CI
Disease-free survival P value HR
95 % CI
P-value
Age B60 years
1.00
[60 years
0.67 0.38–1.20
0.180
Pretreatment CEA B5 ng/ml
1.00
[5 ng/ml
1.49 0.85–2.63
1.00 0.167
1.37 0.97–1.94
0.073
Tumor distance from anal verge B5 cm
1.00
[5 cm
0.88 0.61–1.27
0.480
Type of surgery SSS
1.00
APR
1.91 1.03–3.54
yP stage
1.00 0.041
1.43 0.94–2.18
0.097 \0.0001
0.010
0?1 2
1.00 1.15 0.46–2.90
0.767
1.00 1.46 0.81–2.62
3
2.67 1.14–6.27
0.024
3.21 1.85–5.58 \0.0001
Tumor regression grade (Dworak’s) 1 1.00
0.168
0.207 0.177
1.00
2
0.62 0.33–1.18
0.145
0.78 0.52–1.18
0.239
3
0.21 0.04–0.97
0.045
0.42 0.18–0.97
0.041
4
0.35 0.08–1.53
0.164
0.88 0.39–2.00
0.758
Histologic grade
a
WD/MD
1.00
PD/SRC
1.19 0.43–3.31
CRM 0 mm
0.736 \0.0001
0.001 1.00
1.00
0.1–1.0 mm 0.28 0.08–0.91
0.035
0.32 0.13–0.75
0.009
1.1–2.0 mm 0.11 0.03–0.46
0.002
0.24 0.10–0.54
0.001
[2 mm
0.18 0.08–0.42 \0.0001 0.26 0.14–0.48 \0.0001
CEA carcinoembryonic antigen, SSS sphincter-saving surgery, APR abdominoperineal resection, WD well differentiated, MD moderately differentiated, PD poorly differentiated, SRC signet ring cell, CRM circumferential resection margin, HR hazard ratio, CI confidence interval a
Missing value (n = 5)
preoperative CRT, 1 mm may be considered as a uniform cutoff value for CRM involvement. In the present study, the local recurrence and diseasefree survival curves for group 2 (1.1–2.0 mm) were closer to those for group 1 ([2.0 mm) than those for group 3 (0.1–1.0 mm). In multivariate survival analysis, the HRs for group 2 were comparable with those for group 1 in terms of local recurrence and disease-free survival
(Table 2). A CRM of 1–2 mm after preoperative CRT did not increase the risk of a poor outcome relative to a CRM [ 2 mm. Consistent with previous studies, CRM was an independent prognostic factor for both local recurrence and disease-free survival in the present study. The results showed that the recurrence rate differed according to CRM and was greater in group 4 (0 mm) than in group 3 (0.1–1.0 mm). These results may reflect the effects of preoperative CRT on tumor margins. Preoperative CRT may sterilize tumor cells along the tumor border and provide a safer margin than in the absence of preoperative CRT. However, the precise biological mechanism by which CRT affects the tumor margin remains unclear. Earlier studies showed that short-course RT is not related to the rates of downstaging or CRM involvement.22,24,25 However, long-course CRT was reported to lower the rate of CRM involvement by enhancing tumor regression.26 It was reported that the CRM involvement rate was lower after long-course CRT than after shortcourse RT.9,22,26,27 In short-course RT, the cutoff values for CRM involvement that were associated with prognosis may differ from those of long-course CRT. Further studies are needed to determine the prognostic value of the CRM after short-course RT. In earlier studies of preoperative therapy, the CRM involvement rate ranged from 4 to 18 %.26–29 Consistent with earlier studies, the CRM involvement rate after preoperative CRT was 6.8 % in the present study. The skill of the surgeon is critical for avoiding CRM involvement, primarily by performing meticulous rectal dissection, because incomplete mesorectal excision can lead to CRM involvement. Even after complete mesorectal excision, CRM involvement may occur after invasion of an advanced tumor into the mesorectal fascia. Radial invasion of the tumor and the quality of mesorectal excision can affect CRM status. Preoperative CRT may achieve safer circumferential margins than the absence of preoperative treatment in patients with rectal cancer. There are some limitations to our study. The principal limitation of our study is its retrospective design. We did not calculate the sample size needed to ensure adequate power to detect a prognostic effect of CRM. A second limitation is the small number of patients with a short CRM. Very few studies have investigated the prognosis of rectal cancer according to the CRM. We believe that our data provide an important reference for risk stratification after preoperative CRT. Therefore, further studies with a prospective design and including a larger number of patients are needed to verify our results. In conclusion, our results indicate that CRM distance is an independent risk factor for local recurrence and diseasefree survival in patients with rectal cancer treated with
Circumferential Margin in Rectal Cancer
preoperative CRT. In this setting, CRM distance provides useful information for risk stratification in the recurrence of rectal cancer. ACKNOWLEDGMENT This study was supported by a grant from the National Cancer Center of Korea (NCC-0910160) and by the Converging Research Center Program funded by the Ministry of Education, Science and Technology (project 1131150). CONFLICT OF INTEREST conflicts of interest.
The authors indicated no potential
REFERENCES 1. Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection. Histopathological study of lateral tumour spread and surgical excision. Lancet. 1986;2:996–99. 2. Birbeck KF, Macklin CP, Tiffin NJ, et al. Rates of circumferential resection margin involvement vary between surgeons and predict outcomes in rectal cancer surgery. Ann Surg. 2002;235: 449–57. 3. Wibe A, Rendedal PR, Svensson E, et al. Prognostic significance of the circumferential resection margin following total mesorectal excision for rectal cancer. Br J Surg. 2002;89:327–34. 4. Rullier E, Goffre B, Bonnel C, Zerbib F, Caudry M, Saric J. Preoperative radiochemotherapy and sphincter-saving resection for T3 carcinomas of the lower third of the rectum. Ann Surg. 2001;234:633–40. 5. den Dulk M, Collette L, van de Velde CJ, et al. Quality of surgery in T3-4 rectal cancer: involvement of circumferential resection margin not influenced by preoperative treatment. Results from EORTC trial 22921. Eur J Cancer. 2007;43:1821–8. 6. Nagtegaal ID, Quirke P. What is the role for the circumferential margin in the modern treatment of rectal cancer? J Clin Oncol. 2008;26:303–12. 7. Gosens MJ, Klaassen RA, Tan-Go I, et al. Circumferential margin involvement is the crucial prognostic factor after multimodality treatment in patients with locally advanced rectal carcinoma. Clin Cancer Res. 2007;13:6617–23. 8. Adam IJ, Mohamdee MO, Martin IG, et al. Role of circumferential margin involvement in the local recurrence of rectal cancer. Lancet. 1994;344:707–11. 9. Nagtegaal ID, Marijnen CA, Kranenbarg EK, van de Velde CJ, van Krieken JH. Circumferential margin involvement is still an important predictor of local recurrence in rectal carcinoma: not one millimeter but two millimeters is the limit. Am J Surg Pathol. 2002;26:350–7. 10. Glynne-Jones R, Mawdsley S, Novell JR. The clinical significance of the circumferential resection margin following preoperative pelvic chemo-radiotherapy in rectal cancer: why we need a common language. Colorectal Dis. 2006;8:800–7. 11. de Haas-Kock DF, Baeten CG, Jager JJ, et al. Prognostic significance of radial margins of clearance in rectal cancer. Br J Surg. 1996;83:781–5. 12. Tilney HS, Rasheed S, Northover JM, Tekkis PP. The influence of circumferential resection margins on long-term outcomes following rectal cancer surgery. Dis Colon Rectum. 2009;52: 1723–9. 13. Luna-Perez P, Bustos-Cholico E, Alvarado I, et al. Prognostic significance of circumferential margin involvement in rectal adenocarcinoma treated with preoperative chemoradiotherapy and low anterior resection. J Surg Oncol. 2005;90:20–5.
1351 14. Wittekind C, Compton C, Quirke P, et al. A uniform residual tumor (R) classification: integration of the R classification and the circumferential margin status. Cancer. 2009;115:3483–8. 15. Dent OF, Haboubi N, Chapuis PH, et al. Assessing the evidence for an association between circumferential tumour clearance and local recurrence after resection of rectal cancer. Colorectal Dis. 2007;9:112–21. 16. Hwang MR, Park JW, Kim DY, et al. Prognostic impact of peritonealisation in rectal cancer treated with preoperative chemoradiotherapy: extraperitoneal versus intraperitoneal rectal cancer. Radiother Oncol. 2010;94:353–8. 17. Kim DY, Jung KH, Kim TH, et al. Comparison of 5-fluorouracil/ leucovorin and capecitabine in preoperative chemoradiotherapy for locally advanced rectal cancer. Int J Radiat Oncol Biol Phys. 2007;67:378–84. 18. Quirke P, Dixon MF. The prediction of local recurrence in rectal adenocarcinoma by histopathological examination. Int J Colorectal Dis. 1988;3:127–31. 19. Greene FL, Page DL, Fleming ID, et al. American Joint Committee on Cancer staging manual. 6th ed. New York: Springer; 2002. 20. Hamilton SR, Vogelstein B, Kudo S, et al. Carcinoma of the colon and rectum. In: Hamilton SR, Aaltonen LA editors. World Health Organization classification of tumours. Pathology and genetics of tumours of the digestive system. Lyon: IARC Press, Lyon; 2000. p. 104–19. 21. Dworak O, Keilholz L, Hoffmann A. Pathological features of rectal cancer after preoperative radiochemotherapy. Int J Colorectal Dis. 1997;12:19–23. 22. Marijnen CA, Nagtegaal ID, Kapiteijn E, et al. Radiotherapy does not compensate for positive resection margins in rectal cancer patients: report of a multicenter randomized trial. Int J Radiat Oncol Biol Phys. 2003;55:1311–20. 23. Mawdsley S, Glynne-Jones R, Grainger J, et al. Can histopathologic assessment of circumferential margin after preoperative pelvic chemoradiotherapy for T3-T4 rectal cancer predict for 3-year disease-free survival? Int J Radiat Oncol Biol Phys. 2005; 63:745–52. 24. Marijnen CA, Nagtegaal ID, Klein Kranenbarg E, et al. No downstaging after short-term preoperative radiotherapy in rectal cancer patients. J Clin Oncol. 2001;19:1976–84. 25. Sebag-Montefiore D, Stephens RJ, Steele R, et al. Preoperative radiotherapy versus selective postoperative chemoradiotherapy in patients with rectal cancer (MRC CR07 and NCIC-CTG C016): a multicentre, randomised trial. Lancet. 2009;373:811–20. 26. Bujko K, Nowacki MP, Nasierowska-Guttmejer A, et al. Sphincter preservation following preoperative radiotherapy for rectal cancer: report of a randomised trial comparing short-term radiotherapy vs. conventionally fractionated radiochemotherapy. Radiother Oncol. 2004;72:15–24. 27. Bosset JF, Calais G, Mineur L, et al. Enhanced tumorocidal effect of chemotherapy with preoperative radiotherapy for rectal cancer: preliminary results—EORTC 22921. J Clin Oncol. 2005;23: 5620–7. 28. Wheeler JM, Dodds E, Warren BF, et al. Preoperative chemoradiotherapy and total mesorectal excision surgery for locally advanced rectal cancer: correlation with rectal cancer regression grade. Dis Colon Rectum. 2004;47:2025–31. 29. Rullier A, Laurent C, Vendrely V, Le Bail B, Bioulac-Sage P, Rullier E. Impact of colloid response on survival after preoperative radiotherapy in locally advanced rectal carcinoma. Am J Surg Pathol. 2005;29:602–6.
