Abstract
Background/Aim: Neoadjuvant chemoradiation/radiation therapy in locally advanced (LA) upper rectal adenocarcinoma management remains unclear. The aim of this study was to compare outcomes between neoadjuvant chemoradiation therapy (CRT) and upfront surgery (US). Patients and Methods: A total of 127 patients were retrospectively included from 5 centers (79 treated with US and 48 with CRT). CRT and US groups were compared in terms of postoperative complications and long-term oncological and functional results. Results: Total mesorectal excision (TME) was more frequent in CRT (58% vs. 20% in US, p<0.001). CRT was associated with more overall and severe postoperative complications (60% vs. 30%, p<0.001 and 17% vs. 1%, p=0.002, respectively), and was the only risk factor [OR=18.8 (2.2-160.2), p=0.007]. Five-year overall survival and 5-year recurrence-free survival were similar between CRT and US (96% vs. 91% p=0.256 and 85.4% vs. 85%, p=0.495). The functional results were similar between the two groups. Conclusion: CRT did not improve long-term oncological outcomes in patients with LA upper rectal adenocarcinoma, but increased postoperative complications compared with US.
- Upper rectal cancer
- neoadjuvant treatment
- chemoradiation therapy
- recurrence-free and overall survival
- long-term functional results
Management of locally advanced (LA) rectal cancer is based on a multimodal approach including radiation therapy, chemotherapy followed by total (TME) or partial mesorectal excision (PME) (1-5). Although management of low and mid LA rectal cancer is now well established and standardized, upper rectal cancer management remains debated. To date, very few studies have analyzed the impact of neoadjuvant treatment in patients with upper rectal cancer, and conflicting results have been reported (6-11). The main reasons are: 1) most of the trials reporting the effectiveness of TME and neoadjuvant treatment have included upper rectal cancer along with mid and low rectal tumors. Consequently, there is a lack of evidence regarding possible benefit of neoadjuvant treatment specifically in patients with LA upper rectal cancer (2, 12); 2) Most of the retrospective studies have compared long-term outcomes of sigmoid and/or upper rectal cancer to mid and/or low rectal cancer. This comparison could be considered irrelevant, because rectal cancer location is known to significantly affect oncological outcomes (13). Because of the great heterogeneity of the reported series, the ESMO Society for Medical Oncology states that upper rectal cancers do not benefit from preoperative neoadjuvant treatment and should be treated as colon cancer. However, many authors disagree and consider that patients with cT4 tumors of the upper rectum could potentially benefit from neoadjuvant chemoradiation therapy (CRT) or chemotherapy alone (14).
Due to absence of specific data, the aim of this multicentric study was to compare postoperative results and long-term oncologic outcomes of patients with LA upper rectal cancer receiving neoadjuvant CRT versus patients undergoing upfront surgery (US).
Patients and Methods
Study population. Between 2005 and 2015, all patients who underwent surgery for LA upper rectal adenocarcinoma in 5 tertiary referral colorectal centers in France were retrospectively reviewed from prospectively maintained databases (Beaujon, Pitié-Salpétrière, Caen, Besancon and Tours). All included patients had clinical and radiological LA cancer (T3-T4), with tumor located between 10 and 15 cm from the anal verge. In two centers (Caen and Tours), neoadjuvant CRT (or radiation therapy alone) was performed before surgical resection in all the patients (CRT group). In all three other centers (Beaujon, Pitié-Salpétrière and Besançon), upfront surgery was performed in all cases (US group). All patients who presented with metastatic disease or required associated surgical procedures (mainly other organ resections), who had previous surgery on the left sided colon or local excision of the rectum were excluded. Furthermore, patients who underwent abdominoperineal resection or Hartmann's procedure were also excluded. Data were collected from the medical records of each center and included demographic variables, primary tumor characteristics, intraoperative parameters and short- and long-term postoperative outcomes. Recurrence incidence and site were analyzed during follow-up.
Tumor evaluation and neoadjuvant treatment. Tumor evaluation was based on physical examination, total colonoscopy, endorectal ultrasound and/or pelvic magnetic resonance imaging (MRI) as well as computed tomography (CT-scan). Inferior tumor margin (between 10 and 15 cm from anal verge) was determined using MRI (n=88) and/or endorectal ultrasound through rigid rectoscopy (n=68). Preoperative tumor staging was also performed through MRI or endoscopic ultrasound (preoperative T and N staging). When administered, neoadjuvant therapy, through three-dimensional conformal radiotherapy, consisted of CRT (long-course treatment) or short-course radiotherapy. In patients receiving CRT, 50 Gray were administered in 25 fractions over 5 weeks, with concomitant chemotherapy using Capecitabin, Xeloda® (1600 mg/m2 per day of radiotherapy), and surgery was performed 7 to 8 weeks after completion of neoadjuvant therapy (15). In patients receiving short-course radiotherapy, 25 Gray were delivered in five fractions spanning over 5 to 7 days and surgery was performed a week later (16, 17).
Surgical procedures. Laparoscopic and open procedures were included alike. Surgical procedures consisted of TME or PME (with a constant aim to achieve 5-cm distal margin of mesorectum below the lower edge of the tumor, illustrating some of the remaining controversies regarding surgical management of the upper third of the rectum) (18) followed by termino-terminal or latero-terminal stapled anastomosis. Data regarding conversion to open procedure, diverting stoma (placement and time interval before reversal) and pelvic drain placement were collected for all patients.
Short-term outcomes. Short-term 90-day postoperative complications were ranked according to the international Clavien Classification (19). Any clinical (sepsis, peritonitis, emission of gas, pus, or feces from the pelvic drain, purulent discharge per anus, or rectovaginal fistula) and/or biological suspicion of anastomotic leakage led to an early CT-scan assessment. Severe postoperative complications were defined as Clavien-Dindo >2. Overall and intensive care unit (ICU) hospital stays were recorded for all patients.
Pathology results. Surgical specimens were analyzed using a standardized protocol (20). Tumors were staged using the TNM classification according to the 8th edition of American Joint Committee of Cancer (AJCC). Circumferential and distal margins were defined as positive (R+) when 1 mm or less (primary tumor nodes or tumor deposit within the mesorectum) and as negative (R0) when greater than 1mm. Total or partial mesorectal excision, differentiation grade as well as presence of lymphovascular invasion were stated in the pathology report.
Long-term oncological outcomes. Recurrence was defined, as a lesion deemed suspicious on cross-sectional imaging. Patients were followed up regularly and subject to surveillance protocols including thoracic and abdominal CT-scan. In all five centers, follow-up evaluation was performed every 3 to 4 months during the first 3 years, and every 6 months up to 5 years when no recurrence occurred. In case of recurrence, follow-up evaluation was carried on beyond 5 years.
Long-term functional results. Long-term functional genito-urinary and digestive outcomes as well as quality of life assessments were recorded using a French translation of the low anterior resection score (LARS, score below 20 indicating no LARS, between 21 and 29 indicating minor LARS and above 30 indicating major LARS) (21), the Wexner continence grading scale (22) and the SF36 health survey (23). All previously mentioned surveys and questionnaires were collected in outpatient clinic or by phone interview whenever possible. Greater scores were associated with impaired functional results and more severe symptoms.
Ethical approval. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.
Statistical analysis. Baseline characteristics of the total population, intraoperative and pathological characteristics as well as postoperative oncological and functional outcomes were analyzed. Subgroup analysis was also performed on CRT and US patients who underwent partial mesorectal excision. Categorical variables were compared using the χ2 test or Fischer's exact test when appropriate. Quantitative variables were compared using the T student's test or Mann–Whitney test whenever appropriate. The Kaplan–Meier method was used to estimate recurrence-free survivals (RFSs) and OS, which were compared using the Log-rank test. Univariate and multivariate logistic regression analyses were performed to determine independent risk factors for severe postoperative complications. All statistical analyses were performed using SPSS version 20.0 (SPSS Inc., Chicago, IL, USA) and statistical significance was accepted at the 0.05 level.
Demographic and preoperative characteristics.
Results
Study population. During the study period, a total of 127 patients with LA upper rectal cancer were included, of which 79 (62%) underwent upfront surgical resection (US group) and 48 (38%) received neoadjuvant treatment followed by surgical resection (CRT group). There were 81 (64%) men and the median age was 65.9 years (range=41-94 years). Symptoms led to tumor diagnosis in 86 (68%) patients. Overall, there were 28 (22%) patients classified as ASA 3-4, and median tumor distance from anal verge was 13.0 (10.0-15.0) cm. On preoperative evaluation, patients in the CRT group presented with T2N1 (n=2, 4%), T3 (n=35, 73%) or T4 (n=11, 23%) stage tumors whereas patients in the US group presented with T3 (n=66, 84%) or T4 (n=13, 17%) stage tumors (p=0.112). There were 35 (72.9%) CRT patients and 49 (62%) US patients who presented with N positive preoperative staging (p=0.208). Neoadjuvant treatment consisted of preoperative CRT (long-course setting) in 40 (83%) patients and of radiation therapy without chemotherapy (short-course setting) in 8 (17%) patients. Patients' demographic and preoperative characteristics are presented in Table I.
Intraoperative course. Overall, 49 (39%) patients underwent laparoscopic resection (28 (58%) in the CRT group vs. 21 (27%) in the US group, p<0.001). Total mesorectal excision was performed in 44 (35%) patients (28 (58%) in the CRT group vs. 16 (20%) in the US group, p<0.001). Overall, 71 (56%) patients underwent latero terminal anastomosis with no significant differences between the two groups (p=0.425). Diverting stomas were performed in 90 (71%) patients (43 (90%) in the CRT group vs. 47 (60%) in the US group, p<0.001). Intraoperative findings are presented in Table II.
Short-term postoperative outcomes and pathology results. Postoperative mortality was nil and overall, 53 (42%) patients had postoperative morbidity (21 (44%) in the CRT group 23 (29%) vs. in the US group, p<0.001). Clavien-Dindo III-IV postoperative complications occurred in 8 (17%) patients in the CRT group and 1 (1%) patient in the US group, p=0.002). Median in-hospital stay was 10 (6-46) days for both groups. Overall, stoma reversal was performed in 125 (98%) patients, after a median delay from the primary operation of 86 (8-433) days. Short-term postoperative data are presented in Table II. Pathology results of both groups are detailed in Table III. In the CRT group, 27 (56%) patients presented with pathological T3-4 stage tumor. Overall, 49 (39%) patients were considered to have a positive lymph node status, and 51 (41%) lymphovascular invasion, with no significant differences between the two groups (p=0.089 and p=0.366, respectively). Median pathology tumor diameter was 27 (0-65) mm in the CRT group and 48.5 (20-120) mm in the US group, p<0.001). Overall positive margin rate was 17% (n=22) with no significant difference between the two groups (p=0.263). Univariate and multivariate analysis showed that neoadjuvant treatment was the only risk factor for the development of severe postoperative complications [OR=18.8 (2.2-160.2), p=0.007] (Table IV). Subgroup analysis of patients with preoperative cT4 tumors (n=11 in the CRT group and n=13 in the US group) did not show any significant differences between the two groups in terms of baseline characteristics, intraoperative and postoperative data (Table V).
Intraoperative parameters and postoperative outcomes.
Pathology results.
Late postoperative outcomes.
Long-term oncological outcomes and functional results (Table VI). Adjuvant chemotherapy was administered in 12 (25%) CRT patients and 43 (54%) US patients (p=0.002). Overall, recurrence occurred in 22 (17%) patients (8 (17%) in the CRT group vs. 14 (18%) in the US group, p=0.809) after a median follow-up of 62.7 (6-259) months. Median time to recurrence was 78.1 months (range= 6-259 months) in CRT patients and 45.6 months (range=6-125 months) in US patients (p<0.001). In the overall population, 3- and 5-year OS were 96.9% and 92.9%, and 3- and 5-year RFS were 88.2% and 85.0%. Three- and 5-year OS were 97.9% and 95.8% in the CRT group and 96.2% and 91.1% in the US group (p=0.256) (Figure 1). Three- and 5-year RFS were 89.6% and 85.4% in the CRT group and 87.3% and 84.8% in the US group (p=0.495) (Figure 2). Four (3%) patients required permanent stoma of which 2 (2%) were performed postoperatively, due to unsatisfactory functional results. Quality of life assessments were available for analysis in 31 (65%) CRT patients and 16 (20%) US patients. LARS and Wexner assessments were similar between the two groups.
Partial mesorectal excision subgroup analysis. Overall, 83 (65%) patients underwent partial mesorectal excision (20 in the CRT group and 63 in the US group). There were no significant differences between the two groups in terms of demographic and preoperative characteristics. CRT patients were more often treated laparoscopically (55% vs. 30%, p=0.044), with a more frequent diverting stoma placement (90% vs. 57%, p=0.027). Furthermore, CRT patients presented with more overall (55% vs. 27%, p<0.001) and severe (35% vs. 2%) postoperative complications and had an increased in-hospital stay (13 vs. 10 days, p=0.042). In the US group, there were more pathology T3-4 stage tumors (100% vs. 60%, p<0.001) and pathology tumor diameter was more important (50 vs. 30 mm, p<0.001). Recurrence and overall survival rates were similar between the CRT and the US groups (15% vs. 14%, p=1.000 and 10% vs. 9%, p=1.000, respectively). Long-term functional results were also similar between the two groups.
Discussion
Our study showed that CRT was associated with impaired short-term results and similar long-term oncological and functional outcomes compared with patients who underwent US. These results were further confirmed by multivariate analysis, which revealed CRT as an independent risk factor for severe postoperative complications. Two subgroup analyses on partial vs. total mesorectal excision in patients with preoperative cT4 showed no differences between the CRT and US groups.
Neoadjuvant CRT and TME have become cornerstones in LA rectal cancer management. However, the CRT beneficial effect remains unclear regarding upper rectal cancer and outcomes are inconsistent in this specific subset of patients, with most studies not separating upper from mid/low rectal cancer (2, 12, 24, 25). Even recent trials consider rectal cancer as a single entity, regardless of distance from anal verge (26). Because data regarding impact of CRT in patients with LA upper rectal cancer are therefore still lacking, the present study sought to analyze the effect of neoadjuvant treatment in this specific subset of patients, in terms of short-term postoperative outcomes, long-term oncological and functional results.
CRT vs. US: subgroup analysis of patients with clinical T4 rectal cancer.
Patients who received CRT presented with more postoperative sepsis (21% vs. 5%, p=0.017), and more severe postoperative complications (17% vs. 1%, p<0.001). Anastomotic fistula rates were also higher in the CRT group although the difference was not statistically significantly different (10% vs. 3%, p=0.107). The latter result could be explained by the insufficient number of patients but remains consistent with other published results showing increased pelvic collections and sepsis associated with CRT without increased anastomotic fistula rates (27). Nonetheless, conclusions regarding short-term postoperative outcomes are difficult to draw since many reports comparing CRT and US in LA rectal cancer, have found no differences between the two groups (28, 29).
Because significantly more CRT patients underwent TME compared with US patients (58% vs. 20%, p<0.001), the extent of mesorectal excision (total or partial) had to be taken into consideration in the analysis and results' interpretation. Current results suggested, through partial mesorectal excision subgroup analysis, that the extent of mesorectal excision did not alter short- or long-term postoperative outcomes compared with overall population. Indeed, in the PME subgroup, CRT patients presented with 55% and 35% overall and severe postoperative complications, compared with 27% and 2% US patients, respectively. Furthermore, in multivariate analysis, TME was not significantly associated with the occurrence of severe postoperative complications. Nonetheless, due to the small number of patients, this point requires further confirmation in larger series.
Overall survival based on preoperative chemoradiation therapy. Log-rank test (p=0.256). Blue: Upfront surgery (US); Green: chemoradiation therapy (CRT) group.
Recurrence-free survival based on preoperative chemoradiation therapy. Log-rank test (p=0.495). Blue: Upfront surgery (US); Green: chemoradiation therapy (CRT) group.
Univariate and multivariate analysis: predictive factors of severe postoperative complications.
In terms of long-term oncological outcomes, current results did not show any significant difference between the groups in terms of recurrence rates. The reported 5-year recurrence rates in this setting range between 5 and 10% (1, 3, 5). It is therefore not surprising that the number of expected recurrences in our study was low. More patients might be required in order to demonstrate any beneficial effect of neoadjuvant CRT in LA upper rectal cancer. The previous statement is also valid regarding overall survival, which was also similar between the two groups in current study.
All patients in the US group and the majority of patients in the CRT group had a T3-4 initial evaluation. Only 2 (4%) CRT patients had T2N+ initial tumor staging. It was therefore reasonable to consider tumor stages similar in both CRT and US groups, with 11 (46%) cT4 patients in the CRT group and 13 (54%) in the US group. Aside from the extent of mesorectal excision, tumor stage was taken into consideration in multivariate analysis and only preoperative CRT was found to be an independent risk factor for severe postoperative complications [OR=18.8 (2.2-160.2), p=0.007].
The current study has several limits. First, although patients from 5 tertiary referral centers were included during a study period spanning over 10 years, the total number of patients was relatively low (mainly because only patients with LA upper rectal cancer were included). Second, although follow-up was satisfactory, long-term functional outcomes analyses suffered from important missing data (patients lost to follow-up or deceased). Third, CRT consisted of three-dimensional conformal radiotherapy whereas a few recent studies have reported reduced toxicity through intensity-modulated radiation therapy (30, 31), although higher level of proof is still required (32). Nonetheless, to our knowledge, the current study is the first to report CRT impact in patients presenting with LA upper rectal cancer and analyze short- and long-term oncological and functional outcomes in this subset of patients. In the current study, the functional postoperative outcomes were similar between the two groups, which is not in accordance with previously published results associating CRT with impaired rectal and sexual functions (27). However, these results need confirmation on larger cohorts of patients.
In conclusion, the present study suggests that neoadjuvant CRT in patients presenting with LA upper rectal cancer is associated with impaired short-term outcomes and with no benefit in terms of long-term oncological or functional outcomes, compared with upfront surgery.
Footnotes
Authors' Contributions
Concept/design of the work: M. Ouaissi, Y. Panis, A. Alves, M. Karoui; Data acquisition: N. Tabchouri, Y. Eid, G. Manceau, A. Frontali, Z. Lakkis, B. Heyd, S. Chapet; Data analysis and interpretation: N. Tabchouri, M. Ouaissi, Y. Panis; Drafting: N. Tabchouri, M. Ouaissi; Critical revision of the manuscript: Y. Panis, M. Karoui, E. Salamé, T. Lecomte, G. Calais, M. Ouaissi; Final approval: Y. Panis, M. Ouaissi.
Conflicts of Interest
The Authors have no conflicts of interest to declare.
- Received April 13, 2020.
- Revision received April 24, 2020.
- Accepted April 27, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
