A phase II study of neoadjuvant capecitabine, oxaliplatin, and irinotecan (XELOXIRI) in patients with locally advanced rectal cancer

Abstract Purpose Addition of perioperative multi‐agent chemotherapy to the treatment strategy for locally advanced rectal cancer (LARC) may be a promising option. We conducted a phase II study to evaluate the safety and efficacy of capecitabine combined with oxaliplatin and irinotecan (XELOXIRI) as triplet neoadjuvant chemotherapy in patients with LARC. Methods Patients received neoadjuvant irinotecan and oxaliplatin and capecitabine and then underwent total mesorectal excision. The primary study endpoint was the pathological complete response (pCR) rate. Results Between June 2013 and December 2016, 55 patients were enrolled in the study. Forty‐two (77.8%) of 54 completed the study protocol. The pCR rate was 7.7% (95% CI 3.0% to 18.2%). The 3‐year local recurrence rate was 3.9%, the 3‐year disease‐free survival (DFS) rate was 77.3, and the 3‐year overall survival rate was 96.0%. Conclusion XELOXIRI neoadjuvant chemotherapy appears to be feasible and efficacious for patients with LARC. Although neoadjuvant XELOXIRI alone did not yield our expected pCR rate, the local recurrence rate, 3‐year DFS, and measures of safety met current standards.


| INTRODUC TI ON
In cases of locally advanced rectal cancer (LARC), local recurrence and/or distant metastasis often occur despite R0 resection. Thus, outcomes are unsatisfactory when treatment is limited to total mesorectal excision (TME), which is a standard surgical procedure for rectal cancer. 1 For this reason, LARC is usually treated not only by surgery but also perioperatively by combined radiation and chemotherapy.
In Western countries, the widely accepted standard treatment for LARC consists of neoadjuvant long-course chemoradiation or short-course hypofractionated radiotherapy followed by TME and adjuvant chemotherapy. In Japan, the standard treatment for rectal cancer remains upfront surgery with lateral pelvic lymph node dissection followed by adjuvant chemotherapy. Although neoadjuvant chemoradiotherapy (NCRT) has been shown to significantly decrease the locoregional recurrence rate in cases of LARC, significant effects on distant recurrence and overall survival (OS) rates have not been found. 2,3 In addition, NCRT is associated with long-term bladder and sexual dysfunction, fecal incontinence, and myelosuppression. 4  Several studies have been conducted to evaluate the efficacy and safety of NAC alone for LARC, [5][6][7] and clinical trials of NAC (CAPOX plus bevacizumab [BEV], perioperative CAPOX), aimed at improving disease-free survival (DFS) and OS of patients with rectal cancer, have been conducted. 8,9 However, results of these trials were unsatisfactory.
Meanwhile, a triplet combination of fluoropyrimidine, oxaliplatin, and irinotecan, either the FOFOXIRI or XELOXIRI regimen, administered to patients with metastatic colorectal cancer was shown to yield a higher response rate than that of doublet chemotherapy. 10,11 Few studies have investigated efficacy of a triplet regimen administered as NAC in patients with LARC. Considering triplet combinations as more powerful than other regimens, a clinical trial of the triplet combination XELOXIRI was initiated, and members of our study group have reported results of phase I of the trial. 12 Herein, we report the results of phase II, conducted to confirm the safety of XELOXIRI administered as NAC in patients with LARC and to evaluate the efficacy of XELOXIRI administered as NAC.

| Patients and eligibility criteria
Eligibility criteria were as follows: (1) age over 20 years; (2) Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; (3) histologically proven adenocarcinoma of the rectum; (4) resectable clinical stage II (T3 or T4 with N0) or III (any T and N1-3), with a positive node defined as ≥0.7 mm in a short-axis diameter on imaging and the primary tumor located either above or below the peritoneal reflection and the inferior tumor margin located within 12 cm of the anal verge; and (5) adequate organ function defined by a white blood cell count ≥4000 × 10 9 /L, hemoglobin ≥9.0 g/dL, platelet count ≥100 × 10 9 /L, total bilirubin ≤1.5 mg/dL, serum transaminases ≤150 IU/L, and serum creatinine ≤1.5 mg/dL. Exclusion criteria were as follows: (1) major surgery, radiation therapy, and prior chemotherapy within 4 weeks of inclusion in the trial; (2) a history of interstitial pneumonia, watery diarrhea, evidence of active infection (including positivity for hepatitis B surface antigen), or presence of a severe, systemic disease (e.g. heart failure, hepatic failure, ulcer, uncontrolled diabetes mellitus); (4) presence of another malignancy within the previous 5 years; (5) pregnancy or breast feeding; (6) a homozygous UGT1A1*6/*6 or UGT1A1*28/*28 genotype, or heterozygous UGT1A1*6/*28 genotype; and (7) peripheral neuropathy grade ≥2 according to National Cancer Institute-Common Toxicity Criteria (NCI-CTC) v. 4.0.
Four institutions participated in this study. The ethics committee at each of the participating institutions approved this phase II study, and all patients provided written informed consent before any specific study procedure was undertaken.

| Study design and treatment
Patients enrolled in the study described having received intravenous oxaliplatin (85 mg/m 2 ) and intravenous irinotecan (150 mg/m 2 ) on day 1 and oral capecitabine (1000 mg/m 2 ) twice daily on days 1-7 followed by 7 days of rest, according to the recommended dose established in the previously reported phase I study. 12

| Dose adjustments
Adverse events were assessed according to the Common Toxicity Criteria of the National Cancer Institute (version 4.0). If a patient developed a grade 4 hematological and/or grade 3 or higher nonhematological adverse event, the dose capecitabine and/or oxaliplatin and/or irinotecan was reduced by 25% for all subsequent cycles.
In patients who developed allergic reactions or laryngeal spasm syndrome, duration of the oxaliplatin infusion was increased to 4-6 h.
Treatment was delayed for up to 3 weeks if a patient's absolute neutrophil count was lower than 1500/μl, or their platelet count was lower than 75,000/μl. Patients who required more than 3 weeks recovery from an adverse reaction were excluded from the study.

| Pathological analysis
Standard pathological analysis was performed on all resected specimens at each of the participating institutions. Tumor regression was graded on the basis of a previously described classification system. 13 This was done to determine the effect of NAC, with effects graded as follows: grade 0, no effect; grade 1a, tumor cell necrosis, or degeneration in less than 1/3 of the entire lesion; grade 1b, tumor cell necrosis, degeneration and lytic change in more than 1/3 but less than 2/3 of the entire lesion; grade 2, prominent tumor cell necrosis, degeneration, lytic change and/or disappearance of tumor cells in more than 2/3 of the entire lesion but presence of some viable tumor cells; and grade 3, absence of viable tumor cells.

| Endpoints and statistical analysis
The primary study endpoint was the pathological complete response  local recurrence rate, the R0 resection rate, and safety in terms of adverse events.
The pCR rate has been reported to range from 10% to 20% among patients who have undergone NCRT. [14][15][16][17] We based the sample size of the study described herein on an expected pCR rate of 18% and a threshold pCR rate of 5% to detect differences at a two-sided alpha error of 0.05 and a statistical power of 0.2. As such, the planned sample size was 48 patients, allowing for a 10% dropout rate.
DFS was calculated from the start of treatment to disease progression or to death from any cause, and OS was calculated from the start of treatment to death from any cause, using the Kaplan-Meier method. DFS and OS were compared between subgroups, and differences were analyzed by log-rank test. All statistical analyses were performed with JMP 13 (SAS Institute,). the study protocol is given in Figure 1.

| Patient characteristics
Clinical characteristics of the 55 enrolled patients are summarized in Table 1  Pathological stages are shown in relation to patients' baseline T and N stages for the 52 patients who underwent resection in Table 3.

| Safety (adverse events)
Adverse events are shown in relation to NAC and to adjuvant chemotherapy in

| Three-year DFS and 3-year OS
Five of the 54 patients who underwent NAC died, three from cancer, one from cerebral infarction (as noted above), and one from myocardial infarction. Kaplan-Meier curves for DFS are shown in Figure 2. As noted above, the response rate reported for the triplet combination of fluoropyrimidine, oxaliplatin, and irinotecan was higher than that for doublet chemotherapy in patients with metastatic colorectal cancer. 10,11 Few studies have investigated the activity of any triplet regimen as NAC in patients with LARC.
The BACCHUS trial 24 is the only randomized clinical trial in which a triplet-chemotherapy plus BEV in patients with rectal cancer has been investigated. 24 The trial, comparing neoadjuvant FOLFOX + BEV with FOLFXIRI + BEV for LARC, was conducted in the UK, but unfortunately it was terminated early because of poor accrual.
Neoadjuvant FOLFOXIRI + BEV therapy achieved a promising pCR rate (20%) and the therapy was well-tolerated.
Preoperative use of molecularly targeted drugs is expected to shrink tumors and improve outcomes. However, use of such drugs may increase occurrence of postoperative complications such as anastomotic leakage. 7 Therefore, the study we describe herein was conducted as a phase II study investigating the efficacy of a triplet NAC regimen that does not include a molecular-targeted drug. Our study was of XELOXIRI used for NAC in patients with LARC. Although a triplet regimen such as FOLFOXIRI and XELOXIRI is a common regimen for metastatic or recurrent colorectal cancer and its safety and tolerability have been shown, 11,12 these have not yet been established in the neoadjuvant setting. This triplet regimen has a particularly high rate of grade 3 or higher neutropenia and diarrhea.
In this XELOXIRI study, the rate of neutropenia (≥ grade 3) is 25.9%, which is higher than that of the XELOX regimen of 12.5%-17%. 7,8 However, it is lower than that of the FOLFOXIRI regimen (50%) and the FOLFIRI regimen (28%) for metastatic or recurrent colorectal cancer. 10 The rate of diarrhea (≥ grade 3) is 11.1% in this study, which is also higher than that of the XELOX regimen of 2.4%-3.1%, 7,8 although it is lower than the FOLFOXIRI regimen (17%-30%) and the doublet regimen (11%-14%). 11,12 In addition, in our study, hematologic toxicity (≥ grade 3) was higher compared with the CRT trials were also less common than with other treatments. [7][8][9][10][11][12]25 (Table S1). 22 The difference might be due to the effectiveness of triplet XELOXIRI, the high adjuvant chemotherapy completion rate, and the high prevalence of lateral lymph node dissection.
In recent years, the development of total neoadjuvant therapy including doublet chemotherapy has progressed, and largescale randomized trials have shown improvements in pCR and DFS but not in OS. [27][28][29] Data from our study indicate that clinical trials of total neoadjuvant therapy, including triplet XELOXIRI, are warranted.

| Study limitations
Our data should be interpreted in light of our study limitations. The study was conducted as a single-arm trial, and the patient group was fairly small. Because of the limited sample size, we were unable to investigate which type of LARC is most susceptible to XELOXIRI used as NAC. Notably, a large phase II/III study (PROSPECT trial) that compares CRT with neoadjuvant FOLFOX chemotherapy is being conducted in the United States. We anticipate that additional trials will be conducted to confirm the efficacy of triplet NAC protocols and any related adverse events.

| CON CLUS ION
Although our study did not show improvement in pCR with XELOXIRI used as NAC for LARC, the local recurrence rate, 3-year DFS, and safety were acceptable. The study was conducted as a prospective single-arm clinical trial that included a fairly small group of patients.
Thus, there remains a need for a larger trial to investigate the effectiveness of triplet NAC for LARC.

AUTH O R CO NTR I B UTI O N S
Authors making substantial contributions to conception and design, and/or acquisition of data, and/or analysis and interpretation of

ACK N OWLED G EM ENT
We sincerely thank the participating patients for their contribution to this study and staff from the pharmacology, radiology, and nursing departments that cooperated with our study. We also thank Prof.
Tina Tajima of St. Marianna University School of Medicine for editing the English for this paper.

CO N FLI C T S O F I NTE R E S T
I.T., T.M., and Y.D. are Editorial Board Members of AGS.

E TH I C A L A PPROVA L
This study was conducted in accordance with the Declaration of