UGT1A1 polymorphisms in rectal cancer associated with the efficacy and toxicity of preoperative chemoradiotherapy using irinotecan

2.1 Patients and the preoperative chemoradiotherapy protocol

Between 2009 and 2016, 82 patients with untreated resectable rectal cancer (T3-T4, N0-N2, M0) received preoperative S-1 plus irinotecan chemoradiotherapy before radical surgery. Among these patients, 46 patients agreed to undergo assessment of UGT1A1 polymorphisms. We retrospectively collected the clinical data of these 46 patients. These patients had histologically confirmed adenocarcinoma and an ECOG performance status of 0-2. Disease was staged according to the UICC staging system. We used physical examinations, barium enemas, magnetic resonance imaging (MRI), and CT to evaluate the patients according to the TNM staging system. We defined lower rectal cancer as tumors located below or including the peritoneal reflection. Preoperative chemoradiotherapy consisted of 80 mg/m² per day S-1 (days 1-5, 8-12, 22-26, and 29-33), 60 mg/m² per day irinotecan (days 1, 8, 22, and 29), and 45 Gy radiation (1.8 Gy/day, 5 days per week for 5 weeks). Dose reductions were not applied to the group of homozygous patients. Six to eight weeks after completing chemoradiotherapy, the patients were scheduled for total mesorectal excision.

These protocols were based on the S-1 combined preoperative neoadjuvant multimodality therapy with radiation and irinotecan for locally advanced rectal cancer (SAMRAI-1) trial26 and were approved by the institutional review board at Hyogo College of Medicine (No. 2756).

2.2 Classification of UGT1A1 polymorphisms

UGT1A1 polymorphisms were assessed after consulting with a specialist of hereditary diseases. Blood samples were obtained from patients scheduled to undergo irinotecan treatment, and UGT1A1*28 and UGT1A1*6 were analyzed using the Invader assay (SRL Inc., Kobe, Japan). Polymorphisms were classified into three groups: WT (*1/*1), heterozygous (*28/*1, *6/*1), and homozygous (*28/*28, *6/*6, *28/*6). Double heterozygosities (*28/*6) were classified in the homozygous group because of the reported possible serious toxicities associated with double heterozygosities.27

2.3 Surgical procedure

All patients underwent total mesorectal excision, which included low anterior resection followed by double-stapling technique for reconstruction, intersphincteric resection, and APR. Diverting ileostomy was routinely constructed for all cases, except those with APR. If the involvement of adjacent organs was diagnosed before preoperative chemoradiotherapy, en bloc resection of primary tumors was carried out. Lateral lymph node (LLN) dissection was carried out when lateral lymph nodes were >7 mm determined by MRI before preoperative chemoradiotherapy.

2.4 Association between UGT1A1 polymorphisms and clinicopathological features

The investigated clinical features included age, gender, tumor location, tumor size, pretreatment clinical T stage, N stage and UICC stage, operation type, CEA level, and mGPS. Pathological features included ypT stage, ypN stage, ypUICC stage, histology, circumferential resection margin, downstage, and tumor regression grade. Tumor regression grade by chemoradiotherapy was evaluated according to the 8th edition of the Japanese classification of colorectal carcinoma by the Japanese Society for Cancer of the Colon and Rectum.28 Briefly, grade 0 was defined as no evidence of effect; grade 1 was defined as viable tumor cells remaining in more than one-third of the tumor area; grade 2 was defined as viable tumor cells remaining in less than one-third of the tumor area; and grade 3 (ypCR) was defined as no viable tumor cells remaining. We defined a good response as grades 2 or 3 and a poor response as grades 0 or 1.

2.5 Association between UGT1A1 polymorphisms and toxicity or relative dose intensity

Hematological and non-hematological toxicities from preoperative chemoradiotherapy were evaluated according to the Common Terminology Criteria for Adverse Events, version 4.0. Relative dose intensity (RDI) was calculated as the ratio of the actual dose to the scheduled dose: S-1 (1600 mg/m²), irinotecan (240 mg/m²), and irradiation (45 Gy).

2.6 Associations with pathological response

We further analyzed predictors of pathological good response by selecting six candidate factors including pretreatment clinical T stage, pretreatment clinical N stage, histology, CEA before chemoradiotherapy, mGPS before chemoradiotherapy, and UGT1A1 polymorphisms. mGPS was classified using C-reactive protein levels (CRP: >0.5 mg/dL) and serum albumin levels (<3.5 mg/day).29, 30

2.7 Statistical analysis

Subjects with UGT1A1 polymorphisms are presented as a frequency distribution. Patient characteristics and pathological outcomes are presented as the median and range for continuous variables and by frequency distribution for categorical variables. For comparisons of patient characteristics and outcomes between the UGT1A1 groups, analysis of variance models for continuous variables and Fisher's exact test for categorical variables were used. RDI for each preoperative chemoradiotherapy are presented as median and quartiles and were compared between the UGT1A1 groups using analysis of variance models. Occurrence of grade 1-4 (any grade) toxicity and grade 3-4 is presented as frequency distributions for each UGT1A1 group. Associations between grade 3-4 toxicity and the UGT1A1 groups were evaluated using Fisher's exact test. The frequency distributions of the pathological response (good or poor) for each patient characteristic are shown. Associations between pathological responses and patient characteristics were evaluated using odds ratio and Fisher's exact tests, where the categories “heterozygous” and “homozygous” in the UGT1A1 groups were combined. Allele frequencies of UGT1A1 *6 and *28 were evaluated by an exact test for Hardy-Weinberg equilibrium. A two-sided P-value < .05 was considered statistically significant without adjusting for multiplicity for exploratory analysis. All statistical analyses were carried out using SAS software (SAS Institute, Cary, NC, USA).

3.1 UGT1A1 polymorphisms

We examined UGT1A1 polymorphisms in the 46 patients in the present study. The distribution of UGT1A1*6 and UGT1A1*28 polymorphisms is shown in Table 1. Twenty-six patients (56.5%), 15 patients (32.6%), and five patients (10.9%) were classified into the WT, heterozygous (nine *1/*6 and six *1/*28), and homozygous groups, respectively. The homozygous group consisted of two homozygous patients (one *6/*6 and one *28/*28) and three double heterozygous patients (*6/*28). The allele frequencies of UGT1A1 *6 and *28 were calculated as 15.3% and 12.0%, respectively. Genotype frequencies did not significantly deviate from the Hardy-Weinberg equilibrium (P = 0.38).

3.2 Associations between clinicopathological features and UGT1A1 polymorphisms

Clinical features are listed in Table 2. There were no significant differences in clinical features between the UGT1A1 polymorphism groups, except for mGPS. Pathological features are listed in Table 3. Among the patients treated with preoperative chemoradiotherapy, 71.7% were downstaged according to the UICC criteria. Five patients (10.9%) were grade 1a (all in the WT group), two (4.3%) were grade 1b (all in the WT group), 27 (58.7%) were grade 2 (13 in the WT group, 11 in the heterozygous group, and three in the homozygous group), and 12 (26.1%) were grade 3 (six in the WT group, four in the heterozygous group, and two in the homozygous group). There were no significant differences in pathological features between the UGT1A1 polymorphism groups.

3.3 Association between toxicity and UGT1A1 polymorphisms

Toxicities are listed in Table 4. We observed a significant difference in the incidence of leukopenia and neutropenia in the UGT1A1 polymorphism groups. Grade 3-4 leukopenia and neutropenia did not occur in WT patients. Patients in the heterozygous and homozygous groups had more grade 3-4 neutropenia and leukopenia than those in the WT patient group (P < .05). However, there was no significant difference in the incidence of other factors, including diarrhea, liver function, and renal function, among the three groups.

3.4 Association between RDI and UGT1A1 polymorphisms

Relative dose intensities are listed in Table 4. Each UGT1A1 group showed >90% RDI in those receiving S-1, 88% in those receiving irinotecan, and 99% in those receiving radiation therapy. There was no significant difference in RDI according to UGT1A1 polymorphism despite the significant differences observed in hematological toxicities.

3.5 Associations with pathological responses

Associations between representative candidate factors and pathological responses are listed in Table 5. Only UGT1A1 polymorphisms were significantly associated with good tumor response by preoperative chemoradiotherapy (P = .014) (Table 5). All patients in the heterozygous and homozygous groups showed a good response (grade 2/3) following preoperative chemoradiotherapy.