We included 82 patients with T3-4, N0-2, M0 rectal cancer located within 8 cm of the anal verge who were treated with preoperative CRT using S-1 plus CPT-11 between 2009 and 2016. Prior to preoperative therapy, all patients underwent staging work-ups that included digital rectal examination, measurement of tumor marker levels (carcinoembryonic antigen and carbohydrate antigen 19-9), chest X-ray, abdominal and pelvic computed tomography (CT) and magnetic resonance imaging (MRI). MRI was performed on two occasions, as part of initial staging and following preoperative therapy. Testing for UGT1A1*6 and *28 polymorphisms under national insurance was finally given approval in November 2008 in Japan, and it became measurable at our institution in March 2009. UGT1A1 polymorphisms are assessed only in cases in which consent is obtained after consultation with a specialist in hereditary diseases[27]. The protocol for the present study was based on the SAMRAI-1 trial[28].

The patients were divided into two groups in accordance with the European Society for Medical Oncology guidelines to confirm the outcomes for these subgroups[29]: (1) “bad” rectal cancer [T3(b)c/T4 with peritoneal or vaginal involvement only, N1–2, MFI negative]; and (2) “ugly” rectal cancer (T4 with overgrowth to adjacent organs, pelvic side walls or sacrum, LLN positive, MFI positive).

Preoperative CRT consisted of S-1 (Days 1-5, 8-12, 22-26 and 29-33; 80 mg/m²/d), CPT-11 (Days 1, 8, 22 and 29; 60 mg/m²/d), and radiation (total 45 Gy, 1.8 Gy/d, 5 d per week for 5 wk). Six to eight weeks after completion of preoperative CRT, the patients were scheduled to undergo radical surgery.

All patients underwent total mesorectal excision or extended total mesorectal excision (total mesorectal excision with adjacent visceral resection) to achieve R0 resection. The surgical procedure included low anterior resection, intersphincteric resection and abdominoperineal resection. Intersphincteric resection was recommended in accordance with tumor stage and location, patient age, and preoperative anal function, and patients who did not meet those criteria were selected for abdominoperineal resection. Diverting ileostomy was routinely constructed for all patients with intestinal continuity. LLN dissection was performed when pretreatment MRI showed that the LLNs had a short-axis diameter > 7 mm. Postoperative complications were assessed according to the Clavien-Dindo classification[30]. Pathological response to CRT was evaluated according to the Japanese Classification of Colorectal Carcinoma of the Japanese Society for Cancer of the Colon and Rectum (8^th edition). Grade 0 was defined as no evidence of a therapeutic effect and Grade 3 was pCR[31]. We defined a good response as Grade 2 or 3 and poor response as Grade 0 or 1a/1b.

Hematological and nonhematological toxicity caused by preoperative CRT was evaluated according to the Common Terminology Criteria for Adverse Events, version 4.0[32]. Relative dose intensity was calculated as the ratio of the actual dose to the scheduled dose; S-1 (1600 mg/m²), CPT-11 (240 mg/m²) and full irradiation dose (45 Gy). Dose reductions of CPT-11 were not applied to the group of patients with UGT1A1 mutation.

Median follow-up was 51 mo (range, 17-116 mo). Postoperative adjuvant chemotherapy using 5-FU-based chemotherapy was recommended for all patients except those with ypT0/1 stage, high age, comorbidity, postoperative complications, and social factors. Patient surveillance was subsequently performed as follows: chest–abdominal CT every 6 mo, colonoscopy annually, and blood tests (including measurement of carcinoembryonic antigen and carbohydrate antigen 19-9 levels) at 3-mo intervals. Local recurrence was defined as the detection of a recurrent tumor within the pelvis, and recurrence was defined as the presence of recurrent disease outside the pelvis.

Local recurrence-free survival (LFS), relapse-free survival (RFS) and overall survival (OS) were estimated using the Kaplan–Meier method and compared using the log-rank test. The χ² test was also used to evaluate associations between UGT1A1 polymorphisms and toxicity and feasibility of treatment. We further evaluated clinical factors associated with LFS and RFS to determine the optimal clinical criteria of this regimen by Cox proportional hazard regression model. Independent variables with P < 0.1 in univariate analysis were entered into a multivariate analysis and P < 0.05 was considered statistically significant. Statistical analyses were performed using JMP version 12.0 software (SAS Japan Inc., Tokyo, Japan).

The patients’ clinical characteristics are shown in Table 1. Clinical T4 stage was diagnosed in 10 patients (12.2%). Clinical N stage was deemed positive in 46 patients (56.1%). MRI revealed tumor involvement of the MF in 29 patients (35.4%). EMVI was observed in 36 patients (43.9%). According to the risk category of rectal cancer, 50 patients (61.0%) were divided into the bad group and 32 (39.0%) into the ugly group.

The relative dose intensity was 90.1% for S-1, 92.9% for CPT-11 and 97.6% for RT. Toxicity data are shown in Table 2. Grade 3 or 4 hematological toxicity consisted of leukopenia (n = 15; 18.3%), neutropenia (n = 16; 19.5%) and febrile neutropenia (n = 3; 3.6%). Grade 3 or 4 nonhematological toxicity consisted of diarrhea (n = 22; 26.8%). For Grade 3 or 4 hematological toxicity, four of 16 neutropenia patients (25.0%) whose neutrophil count was reduced to < 500 cells/μL received granulocyte colony-stimulating factor. For Grade 3 or 4 nonhematological toxicity, four of 22 diarrhea patients (18.2%) were prescribed loperamide. All patients recovered after these conservative treatments.

Associations between toxicity/feasibility and UGT1A1 polymorphisms were investigated (Table 3). Forty-eight of 82 patients (58.5%) were assessed for UGT1A1 polymorphism, and 25 (52.1%) were wild type and 23 (47.9%) were mutant type. Patients with the mutant type had more Grade 3 or 4 hematological toxicity than those with the wild type had (P < 0.05). However, there was no significant difference in the incidence of nonhematological toxicity, including diarrhea, in either genotype (P = 0.65). There was no significant difference in CPT-11 dose intensity according to UGT1A1 polymorphisms despite the significant differences observed in hematological toxicity (P = 0.26).

Thirty-one patients (37.8%) underwent low anterior resection, 43 (52.4%) intersphincteric resection and eight (9.8%) abdominoperineal resection. Five patients (6.1%) underwent combined adjacent organ resection and eight (9.8%) LLN dissection.

The postoperative complications are shown in Table 4. Grade 3 pelvic infection was confirmed in nine patients (11.0%) and five (6.1%) developed Grade 3 ileus. Among the patients undergoing sphincter-preserving surgery, seven (9.5%) had Grade 3 anastomosis leakage. During follow-up, six patients could not undergo stoma takedown because of pelvic infection with anastomotic leakage (n = 4) and local recurrence (n = 2).

Pathological findings are listed in Table 5. Thirteen patients (15.9%) achieved complete tumor regression with tumor regression grade 3 (pCR). T downstaging was seen in 41 patients (50.0%) and N downstaging in 36 (43.9%). R0 resection was performed in 79 of 82 patients (96.3%) and R1 resection in three (3.7%), with microscopic residual tumor in the anus levator muscle (n = 2) and pelvic plexus on the pelvic sidewall (n = 1). No patient had R2 resection. Patients with UGT1A1 mutations showed a significantly better response to CRT (including CPT-11) than those without mutations (Table 3).

Twenty-six patients (31.7%) received 5-FU-based adjuvant chemotherapy: UFT plus leucovorin (n = 19), mFOLFOX6 (n = 4), S-1 (n = 2) and capecitabine (n = 1). The reasons for not receiving adjuvant chemotherapy were: ypT0/1 stage (n = 18), high age (n = 11), comorbidity (n = 2), postoperative complications (n = 12), social factors (n = 6), and others (n = 7).

After a median follow-up of 51 mo, 5-year LFS, 5-year RFS and 5-year OS rates were 90.1%, 72.5% and 91.3%, respectively (Figure 1). Local recurrence was seen in six patients: LLNs (n = 4) and other sites (n = 2). Distant recurrence was detected in 20 patients: lung (n = 15), liver (n = 6), para-aortic region (n = 2), inguinal region (n = 1) and bone (n = 1). Some patients had overlapping metastases. LFS did not differ significantly between the bad and ugly groups (96.0% vs 76.2%; P = 0.10); however, RFS was significantly poorer in the ugly group (38.5% vs 87.8% in bad group; P < 0.01).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Long-term outcomes of patients with locally advanced rectal cancer treated with preoperative chemoradiotherapy using tegafur/gimeracil/oteracil plus irinotecan.

We investigated the risk factors for LFS and RFS (Table 6). Multivariate analysis showed that no risk factors for LFS were detected, including previously described risk factors such as T4 stage, MFI, EMVI and LLN swelling. However, MFI and EMVI were associated with poor RFS for locally advanced rectal cancer (OR: 5.82, 95%CI: 1.68-20.2, P < 0.01; OR: 3.42, 95%CI: 1.02-11.5, P = 0.04).

Prospective studies have investigated the optimal treatment strategies for management of locally advanced rectal cancer, and have concluded that preoperative 5-fluorouracil-based chemoradiotherapy (CRT) at 45–50.4 Gy is a standard treatment. However, local recurrence rate remains about 10%; mainly for highly advanced cases.

Multidisciplinary treatments were planned to overcome highly advanced rectal cancer, such as extended surgery, higher radiation doses, and concurrent use of second drugs, such as oxaliplatin or CPT-11.

The aim of this study was to investigate the safety, therapeutic effect, and outcome of preoperative CRT using S-1 plus irinotecan for locally advanced lower rectal cancer.

Between 2009 and 2016, 82 patients underwent total mesorectal excision after preoperative CRT. Preoperative CRT consisted of S-1 (80 mg/m²/d), CPT-11 (60 mg/m²/d), and radiation (total 45 Gy). The median follow-up was 51 months (range: 17-116 mo).

This regimen was well tolerated in terms of toxicity. Associations between toxicity/feasibility and UGT1A1 polymorphisms were investigated. Compared with patients with wild-type UGT1A1, those with mutant type had more Grade 3 or 4 hematological toxicity (P < 0.05). With regard to oncological outcome, mesorectal fascia invasion and extramural vascular invasion were associated with poor relapse-free survival for locally advanced rectal cancer. However, Cox regression analysis did not detect any risk factors for local recurrence-free survival.

This regimen had favorable oncological outcomes for highly advanced rectal cancer.

This was a small retrospective study performed in a single institution. A randomized multicenter study is needed to investigate the influence of dose setting by UGT1A1 polymorphism for preoperative CRT using irinotecan.