Observational study on the efficacy and safety of FOLFOXIRI therapy for locally advanced rectal cancer: The Kanagawa Yokohama Colorectal Cancer Study Group (KYCC)1802

Abstract

Purpose

Although chemoradiotherapy is considered an efficacious treatment option for patients with locally advanced rectal cancer, the associated radiological toxicities and late anal dysfunction are concerning issues. Herein, we examined the efficacy and safety of folinic acid, 5-fluorouracil, oxaliplatin, and irinotecan (FOLFOXIRI) as upfront therapy for marginally unresectable local rectal cancers without distant metastasis.

Methods

This multicenter, prospective, observational study was designed by the Yokohama Colorectal Cancer Study Group. The primary endpoint was the conversion rate. Secondary endpoints were the R0 resection rate, response rate, pathological response rate, postoperative complication rate, relapse-free survival, local progression-free rate, and circumferential resection margin-negative rate.

Results

Twenty patients were enrolled in this study. The study achieved its primary endpoint; the R0 resection rate was 80% (95% confidence interval, 56.3 to 94.3). Major grade ≥ 3 adverse effects included neutropenia in 7 (35%) patients, anemia in 3 (15%), fatigue in 2 (10%), enterocolitis in 2 (10%), febrile neutropenia in 1 (5%), leukopenia in 1 (5%), diarrhea in 1 (5%), fever in 1 (5%), and urinary tract infection in 1 (5%).

Conclusion

FOLFOXIRI therapy was well tolerated and showed comparable efficacy results, providing potential R0 resection in patients with marginally unresectable locally advanced rectal cancer without distant metastasis.

Trial registration

UMIN Clinical Trials Registry (UMIN000040275).

Supplementary Information

The online version contains supplementary material available at 10.1007/s00384-025-04873-2.

Introduction

The standard surgical procedures for resectable rectal cancer include total mesorectal excision (TME) and tumor-specific mesorectal excision (TSME). The purpose of TME is complete circumferential resection of the mesorectum, whereas TSME involves complete partial tumor-specific mesorectal resection. However, if TME or TSME procedures are the first choice of treatment for locally highly invasive advanced rectal cancer, the local recurrence rate will be high, owing to tumor exposure on the surgical resection surface, and surgical resection (R0) may be difficult [1]. Therefore, according to the ESMO and NCCN guidelines, chemoradiation should be performed before TME/TSME. One randomized phase III trial comparing preoperative chemoradiotherapy (CRT) and radiotherapy (RT) in patients with locally advanced T4 rectal cancer reported conversion rates of 89% and 81% in the CRT and RT groups, respectively. Moreover, R0 resection rates were reported to be 84% and 68%, respectively [2]. A comparative study of preoperative CRT and RT for primary unresectable rectal cancer reported conversion rates of 74% and 64% in the CRT and RT groups, respectively, and R0 resection rates of 59% and 50%, respectively [3]. However, chemoradiation has been reported to result in poor quality of life, including anal function [4–6]. In contrast, an increasing number of clinical trials have examined preoperative chemotherapy without radiation therapy, suggesting that preoperative folinic acid, 5-fluorouracil, and oxaliplatin therapy is non-inferior to preoperative chemoradiation for locally advanced rectal cancer in terms of overall survival and local recurrence rates and that radiation therapy can be omitted [7].

Recently, the use of 5-fluorouracil, irinotecan, and oxaliplatin for advanced colorectal cancer has been reported to correlate with prolonged prognoses [8, 9]. The simultaneous use of all three drugs has been reported to cause marked shrinkage of the primary tumor. Folinic acid, 5-fluorouracil, oxaliplatin, and irinotecan (FOLFOXIRI) therapy is considered an extremely successful treatment for unresectable and recurrent colorectal cancer with minimal disease progression [10]. However, there have been no reports on the therapeutic efficacy of FOLFOXIRI therapy aimed at R0 curative intent conversion of unresectable R1 or R2 locally advanced rectal cancer without distant metastasis. We hypothesized that preoperative FOLFOXIRI therapy to achieve tumor shrinkage increases the R0 resection rate and that the suppression of micrometastases significantly contributes to the recurrence rate and survival.

In this prospective, observational study, we evaluated the safety and efficacy of FOLFOXIRI therapy for highly noncurative locally advanced rectal cancer.

Patients and methods

Patients

This multicenter, prospective, observational study was designed by the Kanagawa Yokohama Colorectal Cancer Study Group. Written informed consent was obtained from each patient before enrollment. Seven participating hospitals enrolled patients between January 2021 and October 2022.

Eligibility

The inclusion criteria were as follows: (1) histologically proven rectal adenocarcinoma; (2) patients scheduled for FOLFOXIRI therapy as first-line treatment for rectal cancer; (3) the lowest part of the tumor located ≤ 10 cm from the anal verge (AV); (4) CRM-positive determined using magnetic resonance imaging or computed tomography or invasion to other organs (cT4b tumors); (5) no distant metastasis or lesion with distant metastasis (M1) that could be controlled by this treatment and surgery; (6) patients aged > 20 years and < 81 years; (7) Eastern Cooperative Oncology Group performance status 0 or 1; (8) no history of chemotherapy, rectal resection, pelvic lymph node dissection, or pelvic RT, including treatment for other cancer types; and (9) major organ functions were preserved.

The exclusion inclusion criteria were as follows: (1) concurrent overlapping cancers or iatrogenic overlapping cancers with a disease-free interval of 5 years or less (with the exception of early-stage cancers that can be cured by local treatment); (2) active infection (including hepatitis B antigen-positive cases); (3) fever of 38 °C or higher at enrollment; (4) any serious complications; (5) receiving continuous systemic administration of steroids or immunosuppressive agents; (6) any other cases deemed inappropriate for inclusion by the treating physician.

The CRM criteria were based on those of the MERCURY Study [11]. A positive CRM was defined on magnetic resonance imaging examination when 1) the distance between the tumor and fascia propria of the rectum was < 1 mm, 2) the distance between the tumor and puborectalis muscle was < 1 mm, or 3) the tumor extended into the anal canal and had invaded beyond the internal anal sphincter. A similar finding on the computed tomography scan was defined as CRM-positive.

Chemotherapy

FOLFOXIRI (oxaliplatin, irinotecan, and folinic acid on day 1, followed by 5-fluorouracil over 46-h continuous infusion) was administered every 2 weeks for 2 to 14 cycles, depending on the patient’s tolerance or the doctor’s decision. As this was an observational study, there were no regulations regarding the dose or frequency of administration, and it was administered according to institutional standards.

Histopathological examination

Histopathological regression grades were assessed according to the Japanese Society for Cancer of the Colon and Rectum: Grade 0, no regression; Grade 1a, minimal effect (necrosis less than one-third of the lesion); Grade 1b, mild effect (necrosis less than two-thirds but one-third or more of the lesion); Grade 2, moderate effect (necrosis more than two-thirds of the lesion); and Grade 3, no tumor cells (pathological complete response) [12].

Endpoints and statistical analysis

The primary endpoint was the conversion rate. The secondary endpoints were the R0 resection rate, response rate, pathological response rate, postoperative complication rate, relapse-free survival, local progression-free rate, and circumferential resection margin (CRM)-negative rate.

The analysis population in this study will be those patients enrolled in the study who actually received at least one dose of chemotherapy (FOLFOXIRI therapy).

In the primary analysis, the estimated value of the conversion rate was calculated, with the exact 95% confidence interval also determined. Additionally, a hypothesis test for the population proportion was conducted, assuming the null hypothesis that the Conversion Rate is 75%.

Results

The baseline clinical characteristics of the 20 patients are summarized in Table 1. From January 2021 to October 2022, 20 patients were enrolled in this study, including 13 (65%) males and 7 (35%) females, with a median age of 71 (47–75) years. The main occupied site was up to 5 cm AV in 12 patients (60%), up to 10 cm AV in 6 patients (30%), and up to 15 cm AV in 2 patients (10%). The histological types were well-differentiated tubular adenocarcinoma in 11 cases (55%), moderately differentiated tubular adenocarcinoma in 8 cases (40%), and mucinous adenocarcinoma in 1 case (5%). The depth of tumor was T3 in 3 cases (15%), T4a in 4 cases (20%), and T4b in 13 cases (65%). T4b consisted of 6 cases (30%) of the levator ani muscle, 2 cases (10%) of the sigmoid colon, 1 case (5%) of the vagina, 1 case (5%) of the prostate, 1 case (5%) of the ureter, 1 case (5%) of the retroperitoneum, and 1 case (5%) of the seminal vesicle. Thirteen cases (65%) were RAS mutants, 4 cases (20%) were wild-type, and 3 cases (15%) were unmeasured. In contrast, 0 cases (0%) were BRAF mutants, 17 cases (85%) were wild-type, and 3 cases (15%) were unmeasured. MSI was unmeasured in all cases.

Table 1.

Baseline clinical characteristics of 20 patients

	n	%
Sex
Male	13	65
Female	7	35
Age (years)
Median	71
Range	47–75
ECOG performance status
0	13	65
1	7	35
Distance from the AV (cm)
< 5	12	60
≥ 5, < 10	6	30
≥ 10	2	10
Histological grade
Well differentiated	11	55
Moderately differentiated	8	40
Poorly differentiated	0	0
mucinous	1	5
Clinical T stage
T3	3	15
T4a	4	20
T4b	13	65
Classification of T4b
Levator ani muscle	6	30
Sigmoid colon	2	10
Vagina	1	5
Prostate	1	5
Ureter	1	5
Retroperitoneum	1	5
Seminal vesicle	1	5
RAS status
Mutation	13	65
Wild type	4	20
Unmeasured	3	15
BRAF status
Mutation	0	0
Wild type	17	85
Unmeasured	3	15

ECOG Eastern Cooperative Oncology Group, AV anal verge

During chemotherapy, major adverse events (grade ≥ 3)(CTCAE v5.0) were experienced by 12 of 20 patients (60%): neutropenia in 7 (35%), anemia in 3 (15%), fatigue in 2 (10%), enterocolitis in 2 (10%), febrile neutropenia in 1 (5%), leukopenia in 1 (5%), diarrhea in 1 (5%), fever in 1 (5%), and urinary tract infection in 1 (5%) patient (Table 2).

Table 2.

Overview of adverse effects during chemotherapy(CTCAE v5.0)

	G1	G2	G3	G4	Any grade (%)	≥ G3 (%)
Leukopenia	0	7	0	1	40	5
Neutropenia	0	6	5	2	65	35
Anemia	2	1	3	0	30	15
Thrombocytopenia	1	0	0	0	5	0
Febrile neutropenia	0	0	1	0	5	5
AST	1	0	0	0	5	0
ALT	2	0	0	0	10	0
Total bilirubin	1	0	0	0	5	0
Nausea	2	2	0	0	20	0
Vomiting	1	0	0	0	5	0
Fatigue	4	2	2	0	40	10
Diarrhea	2	0	1	0	15	5
Appetite loss	4	4	0	0	40	0
Peripheral neuropathy	6	1	0	0	35	0
Fever	1	0	1	0	10	5
Urinary tract infection	0	0	1	0	5	5
Edema	1	0	0	0	5	0
Enterocolitis	0	0	2	0	10	10
Interstitial pneumonia	0	1	0	0	5	0

AST aspartate aminotransferase, ALT alanine aminotransferase

A median of 6 (2–14) courses was administered in all 20 patients. Nine patients were treated with bevacizumab. Four patients received CRT (Cape + RT50.4 Gy) after FOLFOXIRI. Two patients (10%) could not be operated on: one had tumor progression, and the other could not continue due to performance status decline. Subsequently, 18 patients (90%) underwent surgery. Of these, 16 (80%) underwent R0 resection (95% confidence interval, 56.3 to 94.3), and 2 (10%) underwent R1 resection. Of the R0 surgeries, 2 (10%) were histologic response Grade 3, 4 (20%) were Grade 2, 4 (20%) were Grade 1b, and 6 (30%) were Grade 1a. Excluding the 4 cases with additional CRT, R0 resection was performed in 13 of 16 cases (81.3%) (95% confidence interval, 54.3 to 95.9)(Fig. 1).

Fig. 1.

Flowchart of 20 enrolled patients. PS, performance status; FOLFOXIRI, folinic acid, 5-fluorouracil, oxaliplatin, and irinotecan

The details of the 18 surgical cases are presented in Table 3. The surgical procedures included low anterior resection in 9 patients, super-low anterior resection in 1 patient, intersphincteric resection in 1 patient, abdominal-perineal resection in 6 patients, and pelvic surgery in 1 patient. Lateral dissection was performed in 8 patients. Ileostomy or colostomy was performed in all patients.

Table 3.

Details of 18 surgical cases

Operation time	456 min	(248–654)
Blood loss	151 mL	(5–4835)
	n	%
Operative procedures
Open	5	27.8
Lap	5	27.8
TaTME	2	11.1
Ro	6	33.3
Type of surgery
LAR	9	50
sLAR	1	5.5
ISR	1	5.5
APR	6	33.3
Pelvic	1	5.5
Lateral lymph node dissection
Yes	8	44.4
No	10	55.5
Stoma
Preoperative	3	16.6
Temporary	9	50
Permanent	6	33.3
No	0	0
Anastomosis
Yes	11	61.1
No	7	38.9
Anastomotic leakage
Yes	0	0
No	11	100

TaTME transanal total mesorectal excision, Lap laparoscopic, LAR low anterior resection, sLAR super-low anterior resection, APR abdominal-perineal resection, ISR intersphincteric resection

Complications of Clavien–Dindo classification Grade 3 or higher were observed in 3 patients (16.7%): surgical site infection, Grade 3a; ileus, Grade 3a; perineal wound dehiscence, Grade 3b. In addition, 2 cases (11.1%) of Grade 2 pelvic infection were observed. No cases of anastomotic leakage were observed.

Discussion

In upfront surgery for locally advanced rectal cancer, it is difficult to secure the CRM, and radical surgery is challenging. Therefore, preoperative CRT is the standard treatment in Europe and the United States, and an increasing number of facilities in Japan have therefore adopted preoperative CRT in recent years. In contrast, RT deteriorates patient quality of life, including anorectal function decline and sexual dysfunction [4–6]. This was observed at 24 months in 51.3% of patients with RT and 36.5% of patients without RT, including patients with occasional fecal incontinence [4]. In addition, preoperative RT has been reported to cause erection and ejaculation problems in men and vaginal dryness and pain during intercourse in women [6]. It has also been reported that the incidence of toxic effects is higher in the postoperative period[7]. Furthermore, the possibility of late complications such as secondary cancer and myelosuppression cannot be ruled out.Therefore, it is desirable to omit RT if possible.

The advantages of preoperative chemotherapy include an improved R0 resection rate owing to tumor shrinkage, early treatment of micrometastases, and more compliant and potent chemotherapy than postoperative chemotherapy. The disadvantages of preoperative chemotherapy include the possibility that radical resection can become difficult if the treatment is ineffective, and that adverse events due to anticancer drugs make it difficult to continue treatment, decrease the patient's ability to tolerate surgery, and increase perioperative complications.

Falcon et al. previously compared fluorouracil, leucovorin, and irinotecan and FOLFOXIRI, which are considered standard chemotherapy regimens for the primary treatment of advanced recurrent colorectal cancer [10]. The report showed that the FOLFOXIRI group outperformed the fluorouracil, leucovorin, and irinotecan group in progression-free survival and overall survival (median progression-free survival, 6.9 vs. 9.8 months; hazard ratio, 0.63; P = 0.0006; median overall survival, 16.7 vs. 22.6 months; hazard ratio, 0.70; P = 0.032). The response rate was also significantly higher in the folinic acid, 5-fluorouracil, and oxaliplatin group (34%) than in the FOLFOXIRI group (60%). Therefore, FOLFOXIRI therapy was selected as the treatment that would maximize the benefits of preoperative drug therapy. In our study, the actual conversion rate was 90%, and the R0 resection rate was 80%. Therefore, the data were comparable to those obtained with the aforementioned preoperative CRT [2, 3].

As FOLFOXIRI comprises a four-drug combination of folinic acid, 5-fluorouracil, irinotecan, and oxaliplatin, adverse events are concerning. In this study, adverse events of grade ≥ 3 due to drug therapy were observed at a high rate (60%). Among them, neutropenia was observed in 35% of cases. However, there was only one case of febrile neutropenia, and no other serious complications were observed. Adverse drug reactions in previous FOLFOXIRI-based clinical trials showed neutropenia in the range of 50–73%, which is comparable to that in our study [10, 13, 14]. Other symptoms of diarrhea and vomiting were not significantly different. Therefore, despite its toxicity, FOLFOXIRI therapy was considered well tolerated. The incidence of perioperative complications of Clavien–Dindo classification Grade 3 or higher was 16.7% (3 of 18 patients). Three patients underwent open abdominal-perineal resection or total pelvic exenteration, and all were at high risk of complications owing to prolonged surgery. Therefore, preoperative FOLFOXIRI therapy is considered acceptable.

This multicenter, prospective, observational study has some limitations. The dosage and other factors were determined according to the standard dosage of the facility. The use of bevacizumab in combination was tailored for each patient. There were no cases of perforation, and no particular increase in postoperative complications such as delayed wound healing in patients treated with bevacizumab. The number of FOLFOXIRI administrations was not determined. The median number of courses administered was 6, and the number of courses varied from case to case.

We believe that radiotherapy should not be added if possible. However, some patients may require additional therapeutic benefit to achieve R0 resection, in which case the addition of chemoradiotherapy may be of great benefit. Radiotherapy may be added to the treatment strategy when drug therapy alone is ineffective [7, 15].We believe that locally advanced rectal cancer requires a flexible response to each case.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

NS, M Sato, and M Shiozawa developed the study design and concept. HO, M Sato, KI, MA, KK, MN, TG, AH, and HM collected patient data. HO, M Shiozawa, MA, KK, MN, TG, AH, and HM participated in patient treatment. HO wrote the manuscript draft. NS and MS critically revised the manuscript. All authors read and approved the final manuscript.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval

All procedures involving human participants were conducted 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

Written informed consent was obtained from each patient before enrollment.

Conflicts of interest

The authors declare no competing interests.