Journal of Surgical Oncology 2015;111:439–444
Adjuvant Chemotherapy After Neoadjuvant Chemoradiation and Curative Resection for Rectal Cancer: Is it Necessary for all Patients? KYUNG UK JUNG, MD,1 HEE CHEOL KIM, MD, PhD,2* JOON OH PARK, MD, PhD,3 YOUNG SUK PARK, MD, PhD,3 HEE CHUL PARK, MD, PhD,4 DOO HO CHOI, MD, PhD,4 YONG BEOM CHO, MD, PhD,2 SEONG HYEON YUN, MD, PhD,2 WOO YONG LEE, MD, PhD,2 1 AND HO-KYUNG CHUN, MD, PhD 1
Department of Surgery, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea 2 Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea 3 Department of Medicine, Division of Hematology-Oncology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea 4 Department of Radiation Oncology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Background: The benefit of adjuvant chemotherapy for patients with locally advanced rectal cancer who have received neoadjuvant concurrent chemoradiation therapy (CCRT) and undergone curative resection remains unclear. Methods: This study was a retrospective review of prospectively collected data. Patients with locally advanced rectal cancer who underwent curative surgery after neoadjuvant CCRT between January 2006 and March 2011 were identified. Four hundred forty-one patients who completed adjuvant chemotherapy (chemo group) were compared with 35 patients who did not receive any adjuvant treatment (nonchemo group). Results: The 5-year disease-free survival (DFS) was significantly higher in the chemo group (78.5% vs. 63.1%, P ¼ 0.016). After stratification of the patients according to nodal status, these differences were no longer significant, but there were trends toward inferior DFS in the nonchemo group in all survival curves. In multivariate Cox regression analysis, no adjuvant chemotherapy (HR, 2.306; 95% CI, 1.101–4.829; P ¼ 0.027) emerged as an independent prognostic factor associated with decreased DFS. Conclusions: Adjuvant chemotherapy was significantly associated with increased DFS among patients who had undergone neoadjuvant CCRT and radical resection for locally advanced rectal cancer. Adjuvant chemotherapy should be considered in every patient after neoadjuvant CCRT irrespective of the final pathology stage.
J. Surg. Oncol. 2015;111:439–444. ß 2014 Wiley Periodicals, Inc.
KEY WORDS: neoadjuvant CCRT; rectal cancer; postoperative chemotherapy
INTRODUCTION
PATIENTS AND METHODS
The current standard treatment for locally advanced rectal cancer is the combination of concurrent chemoradiation therapy (CCRT) and surgical resection. With better local control and less toxicity proven by a large-scale randomized controlled trial [1], the international consensus for cancer treatment generally recommends neoadjuvant CCRT rather than postoperative one for patients with clinically T3–4 or N þ rectal cancer [2–4]. In principle, patients undergo planned postoperative chemotherapy after neoadjuvant CCRT and radical resection. This strategy has been applied to all patients regardless of their surgical pathology results [2,3]. Even if the surgical specimen obtained from radical resection reveals a complete response without any viable tumor cells, the patient is expected to complete the adjuvant chemotherapy schedule over 4–6 months according to their clinical stage, which was estimated before neoadjuvant treatment. However, several reports have shown that the clinical outcomes are more strongly associated with the pathology staging of the surgical specimen after neoadjuvant CCRT rather than the pretreatment clinical staging [5,6], and it is unclear whether patients with a good response benefit from adjuvant chemotherapy. Several studies have been published on this issue, but with somewhat conflicting results [7–10]. The aim of this study was to assess the influence of postoperative chemotherapy on the oncological outcome according to the pathology stage in patients with rectal cancer undergoing neoadjuvant CCRT and curative resection.
Patients and Data Collection
ß 2014 Wiley Periodicals, Inc.
This study was a retrospective review of a prospectively collected database maintained since 1996. This database includes data for all patients with colorectal cancer undergoing surgery in Samsung Medical Center, Seoul, Korea.
Abbreviations: CCRT, concurrent chemoradiation therapy; CT, computed tomography; MRI, magnetic resonance imaging; AJCC, American Joint Committee on Cancer; 5-FU, 5-fluorouracil; RT, radiotherapy; BMI, body mass index; DFS, disease-free survival; HR, hazard ratio; CI, confidence interval; EORTC, European Organization for Research and Treatment of Cancer. Conflicts and source of funding: None * Correspondence to: Hee Cheol Kim, MD, PhD, Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, #50, Irwon-dong, Gangnam-gu, Seoul, 135-710, Republic of Korea. Fax: 82-2-3410-6980. E-mail: [email protected] Received 7 June 2014; Accepted 9 October 2014 DOI 10.1002/jso.23835 Published online 9 December 2014 in Wiley Online Library (wileyonlinelibrary.com).
440
Jung et al.
Between January 2006 and March 2011, 661 patients with locally advanced rectal cancer (radiological T3–4 or N þ) underwent some type of surgical procedure after neoadjuvant CCRT in Samsung Medical Center. Only patients with newly diagnosed primary rectal cancer were included, and patients with recurrence or a history of any kind of treatment with curative intent (radical resection, local excision with CCRT, only CCRT in early rectal cancer, wait and see after CCRT, etc.) were excluded. Among these 661, we excluded sequentially seven patients with anal cancer, 30 with local excision, 25 with definitely palliative resection (remnant disease after surgery although the initial purpose of the resection was curative), and 22 with stage IV disease at the time of the operation after neoadjuvant CCRT (Fig. 1). After excluding these patients, 26 patients who had a history of cancer or who had developed cancer in another body site during the follow-up period were excluded. Among the remaining 551 patients, 441 underwent postoperative chemotherapy and completed the entire schedule as initially planned (chemo group). Thirty-five patients did not receive any adjuvant chemotherapy (nonchemo group) after surgery. Fifty-one patients received more than one cycle of chemotherapy, but did not complete the initially planned schedule. We could not verify whether 24 patients had completed the adjuvant chemotherapy schedule. The clinical and pathological features and oncological outcomes were compared between the two groups. Confirmation of the pathology obtained by colonoscopic biopsy was performed for all patients before treatment. Pretreatment workup for deciding the clinical stage included a digital rectal examination, transrectal ultrasonography, abdominopelvic computed tomography (CT), rectal magnetic resonance imaging (MRI), and chest CT or X-ray. Clinical and pathology classification and staging were determined
according to the American Joint Committee on Cancer (AJCC) TNM staging system. Preoperative chemotherapy included 5-fluorouracil (5FU), usually as a bolus administration of 5-FU at 500 mg/m2/day during weeks 1 and 5 of radiotherapy (RT), or 850 mg/m2 of oral Xeloda twice a day during the entire period of RT. Preoperative RT was delivered using the 3- or 4-field technique, and the median dose was 4400 cGy in 22 fractions. The patients underwent radical surgery after 6–8 weeks after completion of neoadjuvant CCRT. All operations were performed by an experienced colorectal surgical team, according to the principles of radical surgery including the total mesorectal excision technique. All patients were considered candidates for postoperative chemotherapy. They underwent 5-FU based chemotherapy within 3–6 weeks after surgery except when the patient’s general health was too poor to withstand the toxicity of chemotherapeutic agents or the patient refused chemotherapy. Our follow-up policy was every three months in the first year after surgery, every six months for the next two years, and every year thereafter. The examinations performed included physical examination, carcinoembryonic antigen, colonoscopy, abdominopelvic CT, chest CT or X-ray, and positron emission tomography scanning. Rectal and liver MRI was used selectively when there was a suspicion of recurrence. Determination of recurrence was made by a multidisciplinary committee comprising colorectal surgeons, hepatobiliary surgeons, thoracic surgeons, medical oncologists, radiologists, and pathologists. Disease-free survival (DFS) in any patient was defined as the interval between the date of surgery and the date of the first detection of recurrence. All time-to-event data were censored at the last known follow-up without evidence of recurrence. Because we didn’t calculated overall survival but disease free survival, the death of unknown or other cause were not counted as event and censored at the time of death.
Statistics The data were analyzed using PASW Statistics software (version 20.0; IBM SPSS Statistics, Armonk, NY, USA). Continuous variables are presented as the median (interquartile range) and were analyzed with the Mann–Whitney U test. Categorical variables are presented as the number (percentage) and were analyzed with the chi-squared and the Fisher’s exact tests. Kaplan–Meier curves were constructed to evaluate the DFS. Comparisons between groups were performed using the log–rank test. To separate out the effects of explanatory variables (basic demographics - age, sex and body mass index; pathologic parameters—histology, stage; oncologic result of operation—distal resection margin, circumferential resection margin; adjuvant chemotherapy), multivariable survival analysis for DFS was performed using a backward stepwise Cox proportional-hazards regression model. The results are presented as the hazard ratio (HR) and 95% confidence interval (CI). Statistical results were considered significant at P < 0.05.
RESULTS
Fig. 1. Schema of the patient selection flow. Journal of Surgical Oncology
Chemo (n ¼ 441) and nonchemo (n ¼ 35) patients showed similar demographics and clinicopathological characteristics except for age and body mass index (BMI) (Table I). The chemo group was significantly younger than the nonchemo group (54.0 vs. 64.0 years, P ¼ 0.002). BMI was significantly larger in the chemo group than in the nonchemo group (23.8 vs. 21.7 kg/m2,P ¼ 0.020). The gender of patients, location and size of the tumor, histology, and postoperative pathology stage did not differ between the two groups. The chemotherapy regimen during neoadjuvant CCRT, radiation dose and fraction, interval between CCRT and the operation, operation type, and margin status were also similar in both groups (Table II).
CTx After Neoadjuvant CCRT
441
TABLE I. Demographic and Clinicopathological Characteristics of Patients
Age (years) Gender Male Female BMI (kg/m2) Location of cancer (from AV, cm) Size of cancer (long diameter, cm) Histology of cancer* WD þ MD PD þ MUC þ SRC ypT stage (0 : 1 : 2 : 3 : 4) ypN stage (0 : 1 : 2) yp Stage (0 : I : II : III)
Chemo (n ¼ 441)
Nonchemo (n ¼ 35)
P-value
54.0 (46.5–62.5)
64.0 (47.0–72.0)
0.002 0.144
285 (64.6%) 156 (35.4%) 23.8 (21.9–25.5) 4.0 (3.0–6.0) 2.1 (1.0–3.0)
27 (77.1%) 8 (22.9%) 21.7 (20.7–25.0) 4.0 (3.0–7.0) 2.5 (0.7–3.5)
403 (91.6%) 37 (8.4%) 82 : 19 : 116 : 206 : 18 301 : 100 : 40 75 : 107 : 118 : 141
32 (91.4%) 3 (8.6%) 9 : 4 : 4 : 17 : 1 23 : 9 : 3 9 : 8 : 6 : 12
0.020 0.735 0.695 1.000 0.095 0.922 0.454
AV, anal verge; BMI, body mass index; WD, well differentiated; MD, moderatelyl differentiated; PD, poorly differentiated; MUC, mucinous; SRC, signet ring cell. * *Data from one patient was missing.
Table III shows the details of recurrence and mortality for the patients. The overall median follow-up period was 47.8 months. The median follow-up period was nonsignificantly shorter in the nonchemo group than in the chemo group (42.1 and 48.4 months, respectively). Ninety patients (20.4%) in the chemo group and 12 patients (34.3%) in the nonchemo group exhibited some kind of recurrence during the follow-up. The median duration to the detection of recurrence was significantly shorter in the nonchemo group (40.9 vs. 48.3 months, P ¼ 0.013). The Kaplan–Meier survival curve showed significantly different DFS between the groups (P ¼ 0.016, Fig. 2a). The 5-year DFS rate was significantly higher for those patients who underwent adjuvant chemotherapy (78.5% vs. 63.1%). However, this difference was no longer significant when the patients were stratified according to the pathology stage or nodal status (Fig. 3). The overall survival rate was significantly higher in the chemo group (P ¼ 0.013, Fig. 2b). In the multivariate Cox regression analysis, the following factors emerged as independent prognostic factors associated with decreased DFS in these patients undergoing curative resection after neoadjuvant CCRT: short distal resection margin (HR, 1.036; 95% CI, 1.008–1.064; P ¼ 0.010), advanced AJCC stage (stage I: HR, 8.445; 95% CI, 1.076– 66.263, P ¼ 0.042; stage II: HR, 14.562; 95% CI, 1.933–109.701, P ¼ 0.009; stage III: HR, 35.577; 95% CI, 4.881–259.290, P < 0.001),
and no postoperative chemotherapy (HR, 2.306; 95% CI, 1.101–4.829; P ¼ 0.027) (Table IV).
DISCUSSION Neoadjuvant CCRT has contributed to improved clinical outcomes for locally advanced rectal cancer and is now accepted as a standard treatment [1–4]. With advances in imaging technology, more patients clinically diagnosed with T3–4 or N þ rectal cancer are undergoing neoadjuvant CCRT before radical resection in many centers. Compared with postoperative CCRT, because of the risk of toxicity, the chemotherapeutic regimen of neoadjuvant CCRT is generally reduced in dosage and is theoretically more focused on radiosensitization than a systemic effect. As a result, the patient is recommended to complete the ‘remnant’ adjuvant chemotherapy schedule after radical resection. Adjuvant chemotherapy is now the standard treatment for patients who have completed neoadjuvant CCRT and radical resection [2,3]. However, the benefit of adjuvant chemotherapy following neoadjuvant CCRT is assumed on the basis of the results of trials of adjuvant therapy for colon cancer [11], and solid evidence is lacking from large-scale trials about whether adjuvant chemotherapy benefits to all patients with rectal cancer after neoadjuvant CCRT [12].
TABLE II. Parameters Related to Neoadjuvant Chemoradiation and Operations Performed on Patients Chemo (n ¼ 441) CCRT chemo regimen 5-FU FL Xeloda Etc. Radiation dose (cGy) Radiation fraction Interval from CCRT to operation (days) Operation type Low anterior resection Abdominoperineal resection Etc. Resection margin Proximal (cm) Distal (cm) Circumferential (positive : negative) Number of harvested lymph nodes
P-value 0.146
216 28 188 11 4400 22 56
(50.0%) (6.5%) (43.5%) (2.3%) (4400–4400) (22.0–22.0) (54.0–59.0)
388 (88.0%) 42 (9.5%) 11 (2.5%) 11.5 (7.5–16.0) 2.0 (1.4–3.4) 326 : 30 10.0 (7.0–15.0)
CCRT, concurrent chemoradiation; FU, fluorouracil; FL, fluorouracil and leucovorin.
Journal of Surgical Oncology
Nonchemo (n ¼ 35) 13 (39.4%) 5 (15.2%) 15 (45.5%) 0 4400 (4400-4475) 22 (22.0–24.3) 56 (51.8–61.5)
0.736 0.593 0.665 0.182
28 (80.0%) 5 (14.3%) 2 (5.7%) 12.0 (6.3–17.6) 2.0 (1.5–4.0) 21 : 2 9.0 (6.0–13.0)
0.707 0.577 1.000 0.274
442
Jung et al.
TABLE III. Details of Recurrence and Mortality in Patients
Recurrence Recurrence pattern Local Systemic Both Duration to recurrence (months) Mortality Duration to mortality (months) Follow-up after operation (months) *
Chemo (n ¼ 441)
Nonchemo (n ¼ 35)
P-value
90 (20.4%)
12 (34.3%)
0.016* 0.277
26 (5.9%) 51 (11.6%) 13 (2.9%) 48.3 31 (7.0%) 53.9 48.4
4 (11.4%) 6 (17.1%) 1 (2.9%) 40.9 6 (17.1%) 45.9 42.1
0.013 0.014* 0.045 0.057
*Analyzed with Kaplan–Meier curves and the log-rank test.
Several studies have suggested that the oncological outcomes after neoadjuvant CCRT and radical resection for locally advanced rectal cancer correlate with the treatment response [6,13,14]. It is well known that the patient with a complete response following radical resection have an excellent prognosis [13,15,16]. Patients with a good response (ypT0–2N0) have also shown superior DFS compared with the general expectation based on the clinical stage before treatment in many studies [6,8,13,17,18]. Because of this observation, some investigators insist that the risk for these patients must be reevaluated and adjuvant chemotherapy should be used selectively depending on the postoperative stage. Several retrospective series have demonstrated excellent prognosis for yp stage I or node-negative patients independent of whether they receive postoperative chemotherapy [8–10,17], and these studies have concluded that adjuvant chemotherapy might be spared in these favorable subgroups. A multicenter analysis of
Adjuvant Chemotherapy After Neoadjuvant Chemoradiation and Curative Resection for Rectal Cancer: Is it Necessary for all Patients? KYUNG UK JUNG, MD,1 HEE CHEOL KIM, MD, PhD,2* JOON OH PARK, MD, PhD,3 YOUNG SUK PARK, MD, PhD,3 HEE CHUL PARK, MD, PhD,4 DOO HO CHOI, MD, PhD,4 YONG BEOM CHO, MD, PhD,2 SEONG HYEON YUN, MD, PhD,2 WOO YONG LEE, MD, PhD,2 1 AND HO-KYUNG CHUN, MD, PhD 1
Department of Surgery, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea 2 Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea 3 Department of Medicine, Division of Hematology-Oncology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea 4 Department of Radiation Oncology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Background: The benefit of adjuvant chemotherapy for patients with locally advanced rectal cancer who have received neoadjuvant concurrent chemoradiation therapy (CCRT) and undergone curative resection remains unclear. Methods: This study was a retrospective review of prospectively collected data. Patients with locally advanced rectal cancer who underwent curative surgery after neoadjuvant CCRT between January 2006 and March 2011 were identified. Four hundred forty-one patients who completed adjuvant chemotherapy (chemo group) were compared with 35 patients who did not receive any adjuvant treatment (nonchemo group). Results: The 5-year disease-free survival (DFS) was significantly higher in the chemo group (78.5% vs. 63.1%, P ¼ 0.016). After stratification of the patients according to nodal status, these differences were no longer significant, but there were trends toward inferior DFS in the nonchemo group in all survival curves. In multivariate Cox regression analysis, no adjuvant chemotherapy (HR, 2.306; 95% CI, 1.101–4.829; P ¼ 0.027) emerged as an independent prognostic factor associated with decreased DFS. Conclusions: Adjuvant chemotherapy was significantly associated with increased DFS among patients who had undergone neoadjuvant CCRT and radical resection for locally advanced rectal cancer. Adjuvant chemotherapy should be considered in every patient after neoadjuvant CCRT irrespective of the final pathology stage.
J. Surg. Oncol. 2015;111:439–444. ß 2014 Wiley Periodicals, Inc.
KEY WORDS: neoadjuvant CCRT; rectal cancer; postoperative chemotherapy
INTRODUCTION
PATIENTS AND METHODS
The current standard treatment for locally advanced rectal cancer is the combination of concurrent chemoradiation therapy (CCRT) and surgical resection. With better local control and less toxicity proven by a large-scale randomized controlled trial [1], the international consensus for cancer treatment generally recommends neoadjuvant CCRT rather than postoperative one for patients with clinically T3–4 or N þ rectal cancer [2–4]. In principle, patients undergo planned postoperative chemotherapy after neoadjuvant CCRT and radical resection. This strategy has been applied to all patients regardless of their surgical pathology results [2,3]. Even if the surgical specimen obtained from radical resection reveals a complete response without any viable tumor cells, the patient is expected to complete the adjuvant chemotherapy schedule over 4–6 months according to their clinical stage, which was estimated before neoadjuvant treatment. However, several reports have shown that the clinical outcomes are more strongly associated with the pathology staging of the surgical specimen after neoadjuvant CCRT rather than the pretreatment clinical staging [5,6], and it is unclear whether patients with a good response benefit from adjuvant chemotherapy. Several studies have been published on this issue, but with somewhat conflicting results [7–10]. The aim of this study was to assess the influence of postoperative chemotherapy on the oncological outcome according to the pathology stage in patients with rectal cancer undergoing neoadjuvant CCRT and curative resection.
Patients and Data Collection
ß 2014 Wiley Periodicals, Inc.
This study was a retrospective review of a prospectively collected database maintained since 1996. This database includes data for all patients with colorectal cancer undergoing surgery in Samsung Medical Center, Seoul, Korea.
Abbreviations: CCRT, concurrent chemoradiation therapy; CT, computed tomography; MRI, magnetic resonance imaging; AJCC, American Joint Committee on Cancer; 5-FU, 5-fluorouracil; RT, radiotherapy; BMI, body mass index; DFS, disease-free survival; HR, hazard ratio; CI, confidence interval; EORTC, European Organization for Research and Treatment of Cancer. Conflicts and source of funding: None * Correspondence to: Hee Cheol Kim, MD, PhD, Department of Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, #50, Irwon-dong, Gangnam-gu, Seoul, 135-710, Republic of Korea. Fax: 82-2-3410-6980. E-mail: [email protected] Received 7 June 2014; Accepted 9 October 2014 DOI 10.1002/jso.23835 Published online 9 December 2014 in Wiley Online Library (wileyonlinelibrary.com).
440
Jung et al.
Between January 2006 and March 2011, 661 patients with locally advanced rectal cancer (radiological T3–4 or N þ) underwent some type of surgical procedure after neoadjuvant CCRT in Samsung Medical Center. Only patients with newly diagnosed primary rectal cancer were included, and patients with recurrence or a history of any kind of treatment with curative intent (radical resection, local excision with CCRT, only CCRT in early rectal cancer, wait and see after CCRT, etc.) were excluded. Among these 661, we excluded sequentially seven patients with anal cancer, 30 with local excision, 25 with definitely palliative resection (remnant disease after surgery although the initial purpose of the resection was curative), and 22 with stage IV disease at the time of the operation after neoadjuvant CCRT (Fig. 1). After excluding these patients, 26 patients who had a history of cancer or who had developed cancer in another body site during the follow-up period were excluded. Among the remaining 551 patients, 441 underwent postoperative chemotherapy and completed the entire schedule as initially planned (chemo group). Thirty-five patients did not receive any adjuvant chemotherapy (nonchemo group) after surgery. Fifty-one patients received more than one cycle of chemotherapy, but did not complete the initially planned schedule. We could not verify whether 24 patients had completed the adjuvant chemotherapy schedule. The clinical and pathological features and oncological outcomes were compared between the two groups. Confirmation of the pathology obtained by colonoscopic biopsy was performed for all patients before treatment. Pretreatment workup for deciding the clinical stage included a digital rectal examination, transrectal ultrasonography, abdominopelvic computed tomography (CT), rectal magnetic resonance imaging (MRI), and chest CT or X-ray. Clinical and pathology classification and staging were determined
according to the American Joint Committee on Cancer (AJCC) TNM staging system. Preoperative chemotherapy included 5-fluorouracil (5FU), usually as a bolus administration of 5-FU at 500 mg/m2/day during weeks 1 and 5 of radiotherapy (RT), or 850 mg/m2 of oral Xeloda twice a day during the entire period of RT. Preoperative RT was delivered using the 3- or 4-field technique, and the median dose was 4400 cGy in 22 fractions. The patients underwent radical surgery after 6–8 weeks after completion of neoadjuvant CCRT. All operations were performed by an experienced colorectal surgical team, according to the principles of radical surgery including the total mesorectal excision technique. All patients were considered candidates for postoperative chemotherapy. They underwent 5-FU based chemotherapy within 3–6 weeks after surgery except when the patient’s general health was too poor to withstand the toxicity of chemotherapeutic agents or the patient refused chemotherapy. Our follow-up policy was every three months in the first year after surgery, every six months for the next two years, and every year thereafter. The examinations performed included physical examination, carcinoembryonic antigen, colonoscopy, abdominopelvic CT, chest CT or X-ray, and positron emission tomography scanning. Rectal and liver MRI was used selectively when there was a suspicion of recurrence. Determination of recurrence was made by a multidisciplinary committee comprising colorectal surgeons, hepatobiliary surgeons, thoracic surgeons, medical oncologists, radiologists, and pathologists. Disease-free survival (DFS) in any patient was defined as the interval between the date of surgery and the date of the first detection of recurrence. All time-to-event data were censored at the last known follow-up without evidence of recurrence. Because we didn’t calculated overall survival but disease free survival, the death of unknown or other cause were not counted as event and censored at the time of death.
Statistics The data were analyzed using PASW Statistics software (version 20.0; IBM SPSS Statistics, Armonk, NY, USA). Continuous variables are presented as the median (interquartile range) and were analyzed with the Mann–Whitney U test. Categorical variables are presented as the number (percentage) and were analyzed with the chi-squared and the Fisher’s exact tests. Kaplan–Meier curves were constructed to evaluate the DFS. Comparisons between groups were performed using the log–rank test. To separate out the effects of explanatory variables (basic demographics - age, sex and body mass index; pathologic parameters—histology, stage; oncologic result of operation—distal resection margin, circumferential resection margin; adjuvant chemotherapy), multivariable survival analysis for DFS was performed using a backward stepwise Cox proportional-hazards regression model. The results are presented as the hazard ratio (HR) and 95% confidence interval (CI). Statistical results were considered significant at P < 0.05.
RESULTS
Fig. 1. Schema of the patient selection flow. Journal of Surgical Oncology
Chemo (n ¼ 441) and nonchemo (n ¼ 35) patients showed similar demographics and clinicopathological characteristics except for age and body mass index (BMI) (Table I). The chemo group was significantly younger than the nonchemo group (54.0 vs. 64.0 years, P ¼ 0.002). BMI was significantly larger in the chemo group than in the nonchemo group (23.8 vs. 21.7 kg/m2,P ¼ 0.020). The gender of patients, location and size of the tumor, histology, and postoperative pathology stage did not differ between the two groups. The chemotherapy regimen during neoadjuvant CCRT, radiation dose and fraction, interval between CCRT and the operation, operation type, and margin status were also similar in both groups (Table II).
CTx After Neoadjuvant CCRT
441
TABLE I. Demographic and Clinicopathological Characteristics of Patients
Age (years) Gender Male Female BMI (kg/m2) Location of cancer (from AV, cm) Size of cancer (long diameter, cm) Histology of cancer* WD þ MD PD þ MUC þ SRC ypT stage (0 : 1 : 2 : 3 : 4) ypN stage (0 : 1 : 2) yp Stage (0 : I : II : III)
Chemo (n ¼ 441)
Nonchemo (n ¼ 35)
P-value
54.0 (46.5–62.5)
64.0 (47.0–72.0)
0.002 0.144
285 (64.6%) 156 (35.4%) 23.8 (21.9–25.5) 4.0 (3.0–6.0) 2.1 (1.0–3.0)
27 (77.1%) 8 (22.9%) 21.7 (20.7–25.0) 4.0 (3.0–7.0) 2.5 (0.7–3.5)
403 (91.6%) 37 (8.4%) 82 : 19 : 116 : 206 : 18 301 : 100 : 40 75 : 107 : 118 : 141
32 (91.4%) 3 (8.6%) 9 : 4 : 4 : 17 : 1 23 : 9 : 3 9 : 8 : 6 : 12
0.020 0.735 0.695 1.000 0.095 0.922 0.454
AV, anal verge; BMI, body mass index; WD, well differentiated; MD, moderatelyl differentiated; PD, poorly differentiated; MUC, mucinous; SRC, signet ring cell. * *Data from one patient was missing.
Table III shows the details of recurrence and mortality for the patients. The overall median follow-up period was 47.8 months. The median follow-up period was nonsignificantly shorter in the nonchemo group than in the chemo group (42.1 and 48.4 months, respectively). Ninety patients (20.4%) in the chemo group and 12 patients (34.3%) in the nonchemo group exhibited some kind of recurrence during the follow-up. The median duration to the detection of recurrence was significantly shorter in the nonchemo group (40.9 vs. 48.3 months, P ¼ 0.013). The Kaplan–Meier survival curve showed significantly different DFS between the groups (P ¼ 0.016, Fig. 2a). The 5-year DFS rate was significantly higher for those patients who underwent adjuvant chemotherapy (78.5% vs. 63.1%). However, this difference was no longer significant when the patients were stratified according to the pathology stage or nodal status (Fig. 3). The overall survival rate was significantly higher in the chemo group (P ¼ 0.013, Fig. 2b). In the multivariate Cox regression analysis, the following factors emerged as independent prognostic factors associated with decreased DFS in these patients undergoing curative resection after neoadjuvant CCRT: short distal resection margin (HR, 1.036; 95% CI, 1.008–1.064; P ¼ 0.010), advanced AJCC stage (stage I: HR, 8.445; 95% CI, 1.076– 66.263, P ¼ 0.042; stage II: HR, 14.562; 95% CI, 1.933–109.701, P ¼ 0.009; stage III: HR, 35.577; 95% CI, 4.881–259.290, P < 0.001),
and no postoperative chemotherapy (HR, 2.306; 95% CI, 1.101–4.829; P ¼ 0.027) (Table IV).
DISCUSSION Neoadjuvant CCRT has contributed to improved clinical outcomes for locally advanced rectal cancer and is now accepted as a standard treatment [1–4]. With advances in imaging technology, more patients clinically diagnosed with T3–4 or N þ rectal cancer are undergoing neoadjuvant CCRT before radical resection in many centers. Compared with postoperative CCRT, because of the risk of toxicity, the chemotherapeutic regimen of neoadjuvant CCRT is generally reduced in dosage and is theoretically more focused on radiosensitization than a systemic effect. As a result, the patient is recommended to complete the ‘remnant’ adjuvant chemotherapy schedule after radical resection. Adjuvant chemotherapy is now the standard treatment for patients who have completed neoadjuvant CCRT and radical resection [2,3]. However, the benefit of adjuvant chemotherapy following neoadjuvant CCRT is assumed on the basis of the results of trials of adjuvant therapy for colon cancer [11], and solid evidence is lacking from large-scale trials about whether adjuvant chemotherapy benefits to all patients with rectal cancer after neoadjuvant CCRT [12].
TABLE II. Parameters Related to Neoadjuvant Chemoradiation and Operations Performed on Patients Chemo (n ¼ 441) CCRT chemo regimen 5-FU FL Xeloda Etc. Radiation dose (cGy) Radiation fraction Interval from CCRT to operation (days) Operation type Low anterior resection Abdominoperineal resection Etc. Resection margin Proximal (cm) Distal (cm) Circumferential (positive : negative) Number of harvested lymph nodes
P-value 0.146
216 28 188 11 4400 22 56
(50.0%) (6.5%) (43.5%) (2.3%) (4400–4400) (22.0–22.0) (54.0–59.0)
388 (88.0%) 42 (9.5%) 11 (2.5%) 11.5 (7.5–16.0) 2.0 (1.4–3.4) 326 : 30 10.0 (7.0–15.0)
CCRT, concurrent chemoradiation; FU, fluorouracil; FL, fluorouracil and leucovorin.
Journal of Surgical Oncology
Nonchemo (n ¼ 35) 13 (39.4%) 5 (15.2%) 15 (45.5%) 0 4400 (4400-4475) 22 (22.0–24.3) 56 (51.8–61.5)
0.736 0.593 0.665 0.182
28 (80.0%) 5 (14.3%) 2 (5.7%) 12.0 (6.3–17.6) 2.0 (1.5–4.0) 21 : 2 9.0 (6.0–13.0)
0.707 0.577 1.000 0.274
442
Jung et al.
TABLE III. Details of Recurrence and Mortality in Patients
Recurrence Recurrence pattern Local Systemic Both Duration to recurrence (months) Mortality Duration to mortality (months) Follow-up after operation (months) *
Chemo (n ¼ 441)
Nonchemo (n ¼ 35)
P-value
90 (20.4%)
12 (34.3%)
0.016* 0.277
26 (5.9%) 51 (11.6%) 13 (2.9%) 48.3 31 (7.0%) 53.9 48.4
4 (11.4%) 6 (17.1%) 1 (2.9%) 40.9 6 (17.1%) 45.9 42.1
0.013 0.014* 0.045 0.057
*Analyzed with Kaplan–Meier curves and the log-rank test.
Several studies have suggested that the oncological outcomes after neoadjuvant CCRT and radical resection for locally advanced rectal cancer correlate with the treatment response [6,13,14]. It is well known that the patient with a complete response following radical resection have an excellent prognosis [13,15,16]. Patients with a good response (ypT0–2N0) have also shown superior DFS compared with the general expectation based on the clinical stage before treatment in many studies [6,8,13,17,18]. Because of this observation, some investigators insist that the risk for these patients must be reevaluated and adjuvant chemotherapy should be used selectively depending on the postoperative stage. Several retrospective series have demonstrated excellent prognosis for yp stage I or node-negative patients independent of whether they receive postoperative chemotherapy [8–10,17], and these studies have concluded that adjuvant chemotherapy might be spared in these favorable subgroups. A multicenter analysis of
