Int J Colorectal Dis DOI 10.1007/s00384-015-2182-1
ORIGINAL ARTICLE
Pattern of rectal cancer recurrence after curative surgery Minna Räsänen & Monika Carpelan-Holmström & Harri Mustonen & Laura Renkonen-Sinisalo & Anna Lepistö
Accepted: 3 March 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose After curative rectal cancer surgery, local recurrences manifest in 2.4–10 % and distant metastases in 20– 50 % of patients. The effectiveness of different surveillance regimens is not well established. We evaluated the pattern of recurrence and the utility of different surveillance instruments. Risk factors for recurrence were also recorded. Methods This retrospective study comprises 580 consecutive rectal cancer patients operated on at Helsinki University Central Hospital, Finland, during 2005–2011. Data were collected from patient records. After exclusions, 481 patients treated
M. Räsänen : M. Carpelan-Holmström : H. Mustonen : L. Renkonen-Sinisalo : A. Lepistö Department of Surgery, Helsinki University Central Hospital, Helsinki, Finland M. Carpelan-Holmström e-mail: [email protected] H. Mustonen e-mail: [email protected] L. Renkonen-Sinisalo e-mail: [email protected] A. Lepistö e-mail: [email protected] L. Renkonen-Sinisalo Research Programs Unit, Genome-Scale Biology, University of Helsinki, Helsinki, Finland M. Carpelan-Holmström : L. Renkonen-Sinisalo : A. Lepistö Kasarmikatu 11-13, Pl 263, FI-00029 HUS, Finland H. Mustonen Biomedicum, Haartmaninkatu 8, FI-00029 HUS, Finland M. Räsänen (*) Pl 340, FI-00029 HUS, Finland e-mail: [email protected]
with curative intent remained. Patients were followed up according to an intensive surveillance program. Results Rectal cancer recurrence was observed in 124 patients (25.8 %). Local recurrence manifested in 40 patients (8.3%) and distant metastases in 112 patients (23.3 %). Recurrences were observed a median of 1.3 years after surgery. Twelve patients had to be followed up to find one local recurrence and four patients to find one distant metastasis. Recurrences detected during regular follow-up visits were discovered on average earlier than those detected in additional visits arranged because of patient symptoms (p=0.023 for local recurrence, p=0.001 for distant metastases). All surveillance instruments were similarly useful in finding recurrence. Curative treatment was possible in 51 (41.1 %) of 124 patients with disease recurrence. Follow-up led to a 10.0 % chance of detecting recurrence that could be treated with curative intent. Conclusions Rectal cancer recurrences are detected earlier within a surveillance program than by symptoms alone. The most intensive follow-up should be focused on patients with known risk factors for recurrence. Keywords Rectal cancer . Recurrence . Follow-up . Surveillance
Introduction Rectal cancer is the third most common cancer in the world, and its incidence is on the rise. Both the total mesorectal excision (TME) procedure [1] and chemoradiation have improved survival. Five-year survival figures have increased up to 60 %, reaching the level of colon cancer survival [2, 3]. Local recurrence manifests after rectal cancer operation in 2.4–10 % of patients, and one third of these are resectable
Int J Colorectal Dis
[4–7]. Distant metastases, especially in the liver and lungs, occur in 20–50 % of patients, and an increasing proportion of these can be treated with curative intent [6, 8, 9]. Distant metastases reduce survival; thus, efforts should be directed to preventing systemic disease [10]. In many countries, postoperative follow-up is arranged to detect local recurrences and distant metastases. Systematic follow-up is both time-consuming and expensive for the patients as well as the healthcare system. Thus far, its effectiveness has not been well documented. The American Society of Clinical Oncology (ASCO) compared low-intensity and highintensity programs of colorectal cancer surveillance, and based on this, published practice guidelines [11] according to which intensive follow-up are recommended in rectal cancer. Local recurrence is far more frequent in patients treated for rectal cancer than for colon cancer [12]. This study aimed to evaluate the pattern of rectal cancer recurrence in a consecutive patient series and the utility of different surveillance instruments.
Materials and methods Patient characteristics. Data on 580 consecutive rectal cancer patients treated at Helsinki University Central Hospital during 2005–2011 were collected from patient records. In total, 93 patients were excluded for having palliative treatment only or having non-curable distant metastases at the time of primary operation. Five patients died within 30 days of surgery, and one patient was lost to follow-up; these six patients were also excluded from the analysis. The median follow-up time was 3.8 years (range 0.02–8.42). The follow-up period was defined as time from day of surgery until the last contact with the healthcare system or death. Cause of death was verified from official death certificates. Altogether, 481 patients treated with a radical intention were included in the analysis. Of these, 29 had synchronous metastatic disease, which was curatively operated on. Of these synchronous metastases, 20 were located in the liver, five in the lungs, and four in both. Patient and tumor characteristics, surgical procedures, preoperative chemoradiation therapies, pTNM stage, and complication rate of colorectal/anal anastomosis, verified by computer tomography (CT) or fibersigmoidoscopy (fsg), are shown in Tables 1 and 2. Postoperative chemotherapy was offered as adjuvant treatment for lymph node-positive patients, after a long course of preoperative (chemo)radiation, for those with positive resection margins, and selectively for patients who had an emergency operation or preoperative tumor perforation. Adjuvant chemotherapy was also given to patients with distant metastases. Patients in poor general condition did not receive postoperative adjuvant therapy, and some patients refused adjuvant treatment.
Table 1
Patient characteristics
Characteristic
N
Median age in years (range) Gender Male Female Synchronic metastasis (curatively treated) Liver Lungs Liver and lungs Synchronic colon cancer Neoadjuvant therapy No Short radiotherapy (5×5 Gy) Long chemoradiation (45–54 Gy) Modified chemoradiationa Type of surgery Anterior resection (AR) AR with coloanal anastomosis Hartmann’s procedure Abdominoperineal excision (APE) Local excision Endoscopic extirpation Tumor distance from anal verge ≤6 cm >6 cm Data missingb Stage 1 2 3a 3b 3c 4 Data missingc Macroscopically radical surgery Yes No Anastomotic complication 30 days from operation (late) Leakage Abscess Fistula Postoperative chemotherapy No Yes
66 (22–90) 257 224 29 20 5 4 12
53.4 46.6 6.0 4.2 1.0 0.8 2.5
222 195 58 6
46.2 40.5 12.1 1.2
292 34 25 94 31 5
60.7 7.1 5.2 19.5 6.5 1.0
198 282 1
41.2 58.6 0.2
116 129 19 82 66 29 40
24.1 26.8 4.0 17.1 13.7 6.0 8.3
466 15
96.9 3.1
40/326 26 14 34/326 17 7 10
12.3 8.0 4.3 10.4 5.2 2.1 3.1
276 205
57.4 42.6
a
Chemo- or radiotherapy, non-standard regimen
b
Crohn’s disease, not thought to have cancer before operation
c
Percent
Local excision (31), endoscopic extirpation (5), AR (1), T0 downstage (2), no N status (1)
Int J Colorectal Dis Table 2
Tumor characteristics
Characteristic Histology Adenocarcinoma Mucinous carcinoma Grade 1 2 3 4 Missinga pT 0b 1 2 3 4 pN 0 1 2 xc Missingd Circumferential lateral margin (CRM) ≥1 mm 80 years) was decided case by case. Statistical analysis. Cumulative incidences of local recurrences and distant metastases were calculated by KaplanMeier analysis. Potential risk factors for disease recurrence were analyzed with Cox proportional hazard regression analysis. Risk factors with p
ORIGINAL ARTICLE
Pattern of rectal cancer recurrence after curative surgery Minna Räsänen & Monika Carpelan-Holmström & Harri Mustonen & Laura Renkonen-Sinisalo & Anna Lepistö
Accepted: 3 March 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose After curative rectal cancer surgery, local recurrences manifest in 2.4–10 % and distant metastases in 20– 50 % of patients. The effectiveness of different surveillance regimens is not well established. We evaluated the pattern of recurrence and the utility of different surveillance instruments. Risk factors for recurrence were also recorded. Methods This retrospective study comprises 580 consecutive rectal cancer patients operated on at Helsinki University Central Hospital, Finland, during 2005–2011. Data were collected from patient records. After exclusions, 481 patients treated
M. Räsänen : M. Carpelan-Holmström : H. Mustonen : L. Renkonen-Sinisalo : A. Lepistö Department of Surgery, Helsinki University Central Hospital, Helsinki, Finland M. Carpelan-Holmström e-mail: [email protected] H. Mustonen e-mail: [email protected] L. Renkonen-Sinisalo e-mail: [email protected] A. Lepistö e-mail: [email protected] L. Renkonen-Sinisalo Research Programs Unit, Genome-Scale Biology, University of Helsinki, Helsinki, Finland M. Carpelan-Holmström : L. Renkonen-Sinisalo : A. Lepistö Kasarmikatu 11-13, Pl 263, FI-00029 HUS, Finland H. Mustonen Biomedicum, Haartmaninkatu 8, FI-00029 HUS, Finland M. Räsänen (*) Pl 340, FI-00029 HUS, Finland e-mail: [email protected]
with curative intent remained. Patients were followed up according to an intensive surveillance program. Results Rectal cancer recurrence was observed in 124 patients (25.8 %). Local recurrence manifested in 40 patients (8.3%) and distant metastases in 112 patients (23.3 %). Recurrences were observed a median of 1.3 years after surgery. Twelve patients had to be followed up to find one local recurrence and four patients to find one distant metastasis. Recurrences detected during regular follow-up visits were discovered on average earlier than those detected in additional visits arranged because of patient symptoms (p=0.023 for local recurrence, p=0.001 for distant metastases). All surveillance instruments were similarly useful in finding recurrence. Curative treatment was possible in 51 (41.1 %) of 124 patients with disease recurrence. Follow-up led to a 10.0 % chance of detecting recurrence that could be treated with curative intent. Conclusions Rectal cancer recurrences are detected earlier within a surveillance program than by symptoms alone. The most intensive follow-up should be focused on patients with known risk factors for recurrence. Keywords Rectal cancer . Recurrence . Follow-up . Surveillance
Introduction Rectal cancer is the third most common cancer in the world, and its incidence is on the rise. Both the total mesorectal excision (TME) procedure [1] and chemoradiation have improved survival. Five-year survival figures have increased up to 60 %, reaching the level of colon cancer survival [2, 3]. Local recurrence manifests after rectal cancer operation in 2.4–10 % of patients, and one third of these are resectable
Int J Colorectal Dis
[4–7]. Distant metastases, especially in the liver and lungs, occur in 20–50 % of patients, and an increasing proportion of these can be treated with curative intent [6, 8, 9]. Distant metastases reduce survival; thus, efforts should be directed to preventing systemic disease [10]. In many countries, postoperative follow-up is arranged to detect local recurrences and distant metastases. Systematic follow-up is both time-consuming and expensive for the patients as well as the healthcare system. Thus far, its effectiveness has not been well documented. The American Society of Clinical Oncology (ASCO) compared low-intensity and highintensity programs of colorectal cancer surveillance, and based on this, published practice guidelines [11] according to which intensive follow-up are recommended in rectal cancer. Local recurrence is far more frequent in patients treated for rectal cancer than for colon cancer [12]. This study aimed to evaluate the pattern of rectal cancer recurrence in a consecutive patient series and the utility of different surveillance instruments.
Materials and methods Patient characteristics. Data on 580 consecutive rectal cancer patients treated at Helsinki University Central Hospital during 2005–2011 were collected from patient records. In total, 93 patients were excluded for having palliative treatment only or having non-curable distant metastases at the time of primary operation. Five patients died within 30 days of surgery, and one patient was lost to follow-up; these six patients were also excluded from the analysis. The median follow-up time was 3.8 years (range 0.02–8.42). The follow-up period was defined as time from day of surgery until the last contact with the healthcare system or death. Cause of death was verified from official death certificates. Altogether, 481 patients treated with a radical intention were included in the analysis. Of these, 29 had synchronous metastatic disease, which was curatively operated on. Of these synchronous metastases, 20 were located in the liver, five in the lungs, and four in both. Patient and tumor characteristics, surgical procedures, preoperative chemoradiation therapies, pTNM stage, and complication rate of colorectal/anal anastomosis, verified by computer tomography (CT) or fibersigmoidoscopy (fsg), are shown in Tables 1 and 2. Postoperative chemotherapy was offered as adjuvant treatment for lymph node-positive patients, after a long course of preoperative (chemo)radiation, for those with positive resection margins, and selectively for patients who had an emergency operation or preoperative tumor perforation. Adjuvant chemotherapy was also given to patients with distant metastases. Patients in poor general condition did not receive postoperative adjuvant therapy, and some patients refused adjuvant treatment.
Table 1
Patient characteristics
Characteristic
N
Median age in years (range) Gender Male Female Synchronic metastasis (curatively treated) Liver Lungs Liver and lungs Synchronic colon cancer Neoadjuvant therapy No Short radiotherapy (5×5 Gy) Long chemoradiation (45–54 Gy) Modified chemoradiationa Type of surgery Anterior resection (AR) AR with coloanal anastomosis Hartmann’s procedure Abdominoperineal excision (APE) Local excision Endoscopic extirpation Tumor distance from anal verge ≤6 cm >6 cm Data missingb Stage 1 2 3a 3b 3c 4 Data missingc Macroscopically radical surgery Yes No Anastomotic complication 30 days from operation (late) Leakage Abscess Fistula Postoperative chemotherapy No Yes
66 (22–90) 257 224 29 20 5 4 12
53.4 46.6 6.0 4.2 1.0 0.8 2.5
222 195 58 6
46.2 40.5 12.1 1.2
292 34 25 94 31 5
60.7 7.1 5.2 19.5 6.5 1.0
198 282 1
41.2 58.6 0.2
116 129 19 82 66 29 40
24.1 26.8 4.0 17.1 13.7 6.0 8.3
466 15
96.9 3.1
40/326 26 14 34/326 17 7 10
12.3 8.0 4.3 10.4 5.2 2.1 3.1
276 205
57.4 42.6
a
Chemo- or radiotherapy, non-standard regimen
b
Crohn’s disease, not thought to have cancer before operation
c
Percent
Local excision (31), endoscopic extirpation (5), AR (1), T0 downstage (2), no N status (1)
Int J Colorectal Dis Table 2
Tumor characteristics
Characteristic Histology Adenocarcinoma Mucinous carcinoma Grade 1 2 3 4 Missinga pT 0b 1 2 3 4 pN 0 1 2 xc Missingd Circumferential lateral margin (CRM) ≥1 mm 80 years) was decided case by case. Statistical analysis. Cumulative incidences of local recurrences and distant metastases were calculated by KaplanMeier analysis. Potential risk factors for disease recurrence were analyzed with Cox proportional hazard regression analysis. Risk factors with p
