Optimized Rat Models Better Mimic Patients with Irinotecan-induced Severe Diarrhea

Abstract

Irinotecan-induced severe diarrhea (IISD) not only limits irinotecan’s application but also significantly affects patients’ quality of life. Existing animal models often inadequately represent the dynamics of IISD development, progression, and resolution across multiple chemotherapy cycles, yielding non-reproducible and highly variable response with limited clinical translation. Our studies aim to establish a reproducible and validated IISD model that better mimics the pathophysiology progression observed in patients, enhancing translational potential. We investigated the impact of dosing regimens (including different dose and infusion time and two cycles of irinotecan administration), sex, age, tumor-bearing conditions, and irinotecan formulation on the IISD incidence and severity in mice and rats. Lastly, we investigated above factors’ impact on pharmacokinetics of irinotecan, intestinal injury, and carboxylesterase activities. In summary, we successfully established a standard model establishment procedure for an optimized IISD model with highly reproducible severe diarrhea incidence rate (100%) and a low mortality rate (11%) in F344 rats. Additionally, the rats tolerated at least 2 cycles of irinotecan chemotherapy treatment. In contrast, the mouse model exhibited suboptimal IISD incidence rates (60%) and an extremely high mortality rate (100%). Notably, dosing regimen, age and tumor-bearing conditions of animals emerged as critical factors in IISD model establishment. In conclusion, our rat IISD model proves superior in mimicking pathophysiology progression and characteristics of IISD in patients, which stands as an effective tool for mechanism and efficacy studies in future chemotherapy-induced gut toxicity research.

1. Introduction

Chemotherapy-induced diarrhea is a prevalent issue in cancer chemotherapy, particularly as one of the severe adverse effects of irinotecan or FOLFIRI (leucovorin calcium, 5-fluorouracil, and irinotecan) chemotherapy regimen. In both human and animal models, irinotecan-induced diarrhea is classified into early (acute) and late (or delayed-onset) diarrhea (Stein, 2010). In humans, acute diarrhea occurs within 24 hours post irinotecan administration and is clinically recognized as a component of cholinergic syndrome resulting in colonic hyperstimulation. Irinotecan has been shown to mimic the effects of acetylcholine by inhibiting acetylcholinesterase and its binding to muscarinic receptors. Hence, atropine monotherapy, as a competitive antagonist at anticholinergic receptors, is typically dosed as 0.25 to 1 mg intravenous (i.v.) or subcutaneous (s.c.) even prophylactically to prevent/treat irinotecan-induced acute diarrhea. Whereas, irinotecan-induced delayed-onset severe diarrhea (IISD) occurs via multiple mechanisms, affecting up to 40% of the treated patients (grades 3 and 4, according to NCI CTCAE 5.0) (Guichard, et al., 1998). IISD not only leads to prolonged hospitalization and delays in chemotherapy but also poses a life-threatening risk to patients in certain cases (institution, 2010), necessitating immediate resolution.

IISD has been established as a dose-dependent toxicity. The level of exposure of SN-38 (an active and toxic metabolite of irinotecan) to intestinal cells is directly correlated to the incidences and IISD. Various strategies to treat IISD, including anti-diarrheal therapies (e.g., loperamide), UGT1A1*28 genetic testing, UGT inhibitors or inducers, carboxylesterase inhibitors, etc., have been tested in preclinical and clinical studies. Although certain approaches have shown promising efficacy in treating or preventing IISD in preclinical studies, the majority of current treatment approaches fail to be translated from animal models to human applications due to poor methodology and the failure of the models to accurately mimic the human disease condition (Swami, et al., 2013). Therefore, to enhance the translation of novel therapies from precilinical studies to clinical solutions, there is a need for establishing improved animal models of IISD to delineate various pathways of irinotecan-induced gut toxicity and enable the assessment of molecular targets and dose-response of proposed interventions.

The animal models of IISD have provided insights into cellular mechanisms and clinical pharmacology of IISD. Mice, rats, and pig models of IISD were developed and offer both advantages and disadvantages, as summarized in Table 1 (Gnutzmann, et al., 2015). The major limitations of existing models include:(1) The incidence of severe diarrhea showed substantial variation across different labs, influenced by factors such as sex, strain, age, and body weight. (2) Irinotecan administration schedule is another critical determinant in the severity of late diarrhea. (3) In some studies, the distinction between irinotecan-induced acute diarrhea and IISD was not clear, rendering the anti-diarrhea efficacy study imprecise (Trifan, et al., 2002) (Kurita, et al., 2011). (4) Certain studies with high mortality rates > 50%, failed to extrapolate their findings to human applications (Gibson, et al., 2007). (5) The protocol or the procedure of building the IISD model lacks clarity and details in their procedure, which poses challenges for replication, particularly in defining criteria for diarrhea severity. For example, some authors either didn’t describe the difference between mild and severe diarrhea or only used vague words, such as muddy feces, watery feces, or mucous feces.

Table 1.

Summary of laboratory rat’s sex, age and bodyweight, irinotecan dose, administration routes and schedule’s impact on the IISD.

Strain, sex, age, body weight	Dose (mg/kg), route, volume	Acute diarrhea	IISD start day	Incidence of Severe diarrhea grade ≥ 3, rat numbers	Recover start day	Criteria of diarrhea	Ref.
Wistar, male, 7–8 w,185–232 g	200, i.v. infusion for 0.5 hr, 3 mL/kg	Yes, disa PPe ared afte r 3 hr	2	Day 3 (100%), n=10.	4	0, normal; 1, soft feces or small black feces; 2, muddy feces; 3, watery, feces or mucous feces.	(Kurita, Kado, 2011)
	60, i.v. bolus, day 0–3 (4 doses)	No	3	Day 4 (60%), n=9.	5
Gunn, male 7–8w,206–254 g	20, i.v. bolus	Yes, disaPPear ed after 3 hr	2	Day 2 (100%), 2 rats died, n=9.	5
SD, female, 180–250 g	160, 170,180 for 2 days, i.v. infusion for 15 min (8 mL /15 min) 2.5 mL/min/kg	Yes, on days 1 and 2		Day 4, 160 mg/kg/day, 0%, n=3; 170 mg/kg/day, 100%, n=2.	7	0, normal stool; 1, slight wet/soft stool; 2, more than slight but less than moderate diarrhea; 3, moderate diarrhea: wet and unformed stool with moderate perianal staining of the coat; 4, moderate diarrhea; severe anal staining; 5, severe diarrhea: watery stool with severe Perianal staining of the coat.	(Waterhouse, et al., 2014)
SD, male, 180±20 g	150 for 2 days, i.v. bolus	NM	1	NM, n=6.	5	0–3	(Deng, et al., 2017)
SD, male, 288 ± 32 g	60, i.v. bolus , 4 doses	NM	4–5	5.3%, n=20.	7	1, normal: normal stool; 2, mild: slightly wet stool without staining of the coat; 3, moderate: wet and unformed stool with moderate perianal staining of the coat; 4, severe: watery stool with severe staining of the coat around the anus.	(Sezer, et al., 2009)
Wistar,6w, 177–188 g	20, i.v. bonus	No diarrhea		No diarrhea, n=3–8.	NM	0, normal; 1, soft feces or small black feces; 2, muddy feces; 3, watery feces; 4, mucous feces.	(Onoue, et al., 2008)
Gunn, 7 w, 186–230g	20, i.v. bonus	NM	4	62.5%, n=8.	6 ~ 8
Dark Agouti, female, 150–180 g#	200, i.p.	NM	3	Peaked on day 3, 62%, n=6.		0, none; 1, mild, staining around anus; 2, moderate, staining around anus and top of hind legs; 3, severe, staining over legs and abdomen, often with continual oozing.	(Bateman, et al., 2016)
Dark Agouti, female, 150–170 g#	200, i.p.	0–24 hr, 23%	3	Day 3, 6%,n=6.	5	0, no diarrhea;1 mild diarrhea; 2, moderate diarrhea; 3, severe diarrhea.	(Stringer, et al., 2009)
Dark Agouti , female, 150–170 #	200, i.p.	0– 24 hr, 23%	3	Day 3, no severe diarrhea, n=6.	5	1, mild diarrhea, staining of anus; 2, moderate diarrhea, staining top of legs and lower abdomen; 3, severe diarrhea, staining over legs and higher abdomen as well as continual anal leakage.	(Logan, et al., 2008, Andrea M. Stringer, 2007)
Gunn, male, 200–275 g	60~120, i.p.	NM	1	24hr after last dose, 20mg/kg×3, 50%, n=2; 40mg/kg×3, 100%, n=2; 80mg/kg, 0%, n=2.	NM	1, malformed stool; 2, watery stool with perianal staining; 3, severe perianal staining	(Tallman, et al., 2007)
SD, male,165– 185g	Total 240, i.v. bolus, 4–8 doses	On day 3 and 4 after 3hr	1–3	60mg/kg×4, 20%, n=10; 30mg/kg×2×4,0%, n=10; 30mg/kg×8day,0 %, n=5; 40mg/kg×6day; 0%, n=5.	NM	0, normal; 1, slight; 2, moderate; 3, severe.	(Kurita, et al., 2000)
SD, male, 220–280g#	120–480, i.v. bolus	NM	4	200mg/kg, QD×1, 0%; 120mg/kg, QD×2, 12.5 ± 6.9%; 135mg/kg, QD×2, 31.9 ± 7.8%; 150mg/kg, QD×2, 38.9 ± 7.6%; 80mg/kg, QD×4, 23.2 ± 12%; 120mg/kg, QD×3, 38.9 ± 7.6%; 100mg/kg, QD×4, 63.6 ± 7.6%; 120mg/kg, QD×4, 88.9 ± 1.6%; n=6–12.	NM	0, normal: normal stool or absent; 1, slight: slightly wet and soft stool; 2, moderate: wet and unformed stool with moderate perianal staining of the coat; 3, severe: watery stool with severe perianal staining of the coat.	(Trifan, Durham, 2002)
Breast cancerbearing dark agouti rats , female, 150g	100,150,20 0, i.p. bolus	100, no; 150, no; 200, yes.	4	100, 0%, recovery, n=4; 150, 50% died, n=4; 200, 100%, 100%died, n=4.	NM	0, none; 1, mild; staining around anus; 2, moderate; staining around anus and top of hind legs 3, severe; staining over legs and abdomen, often with continual oozing.	(Gibson, et al., 2003)
Male Wistar Han IGS rats;240– 440g;6–10weeks	200, i.p. bolus	No	3	85.2±12.3%(day3) , 0%,n=6	NM	Diarrhea was expressed as percent colonic fecal water content, >80% are loose and watery.	(Dahlgren, et al., 2022)

Note: Irinotecan (kindly supplied by Pfizer, Kalamazoo, MI) was administered in a sorbitol/lactic acid buffer (45 mg/ml sorbitol/ 0.9mg/ml lactic acid, pH= 3.4). Male rats reach sexual maturity at about 6 to 10 weeks of age; females reach maturity at 8 to 12 weeks. Younger animals are more responsive to chemotherapy (Bateman, Weaver, 2016), but old rats were sensitive to the IISD. The intravenous LD50 in rat is 84 mg/kg (Pharma, 2010).

^#means the rats were given atropine to treat the acute diarrhea by subcutaneously 0.01 mg/kg, except the reference 15 were given 2 mg/kg.

To fill the existing knowledge regarding the mechanisms of IISD, our objective is to develop an improved rodent model of IISD. Subsequently, we will conduct a series of experiments to optimize the IISD model, providing a high-fidelity representation of the incidence and severity of IISD in humans, and ensure reproducibility and translational potential.

2. Materials and methods

2.1. Chemicals and reagents

Irinotecan HCl trihydrate was purchased from Teva Parenteral Medicine, Inc. (Irvine, NY, USA) in an injectable formulation (20 mg/mL) and JARI Pharmaceutical Co., Ltd. (Jiangsu, China) in a power form with 99% purity (LOT YL-101–180901). D-Sorbitol, lactic acid, 1,3-Propanediol were obtained from Sigma-Aldrich Co. (St. Louis, MO). The reference standards of irinotecan, SN-38, and SN-38-glucuronide were purchased from Toronto Research Chemicals Inc. (North York, ON, Canada). Acetonitrile, methanol, and formic acid (LC-MS grade) were procured from EMD (Gibbstown, NJ).

Irinotecan injection formulation preparation: Irinotecan HCl trihydrate was dissolved in 60 ± 5 °C water at a concentration of 20 mg/mL, followed by stirring at a speed of 200–400 rpm for 2 hr. Once the solution became clear, sorbitol 45 mg/mL and 0.9 g/mL lactic acid was added to the formulation, then the injection formulation was filtered with a 0.22 μm filter membrane in the cell culture hood (Thermo Scientific^™ 1377, USA). The pH of the injection formulation was 3.0–3.5. We prepared lab-made irinotecan injection formulation due to the high cost of commercially available prescription drug.

Materials and equipment for rat intravenous infusion: Harvard PHD 2000 Infusion Syringe Pump; appropriate size syringe (typically 10 mL); rat restrainer; heat lamp; heat pad; weight balance; isoflurane vaporizers for small animals and rodents; water from a Millipore Milli-Q water purification system (Bedford, MA, USA); 70% isopropyl alcohol. Disposable medical venous indwelling needles 26/ 24 gauze were purchased from the Daoran Trade (Jilin) Co., Ltd. (Jilin, China), and other supplements were obtained from commercial sources.

2.2. Establishment of the IISD Model

2.2.1. Animals

F344 rats at 8–10 weeks’ old (male and female) were purchased from Charles River Laboratories International, Inc. (Houston, TX). The young rat 4–6 weeks’ old (80–150 g, equal representation of both sexes) and the Pirc rats (female 44 weeks’ old) were bred in the animal facility at the University of Houston (protocol #17–005). The body weight of the adult male and female were from 150–200 g and 80–150 g (8–10 weeks) at the beginning of irinotecan administration, respectively. The rats were used for experiments after at least 1 week of acclimatization with free access to tap water and commercial animal chow. The animal room was maintained at a temperature of 22 ± 2°C and a relative humidity of 55 ± 15% with a 12-hr light-dark cycle. The animal protocols used in this study were approved by the University of Houston’s Institutional Animal Care and Uses Committee.

The mouse experiments were approved by the Institutional Animal Care and Use Committee of Texas Southern University (animal protocol # 9124). Female and male C57BL6 mice (body weight of 18–25 g, 4–6 weeks) were obtained from Jackson Lab (Bar Harbor, ME, USA). Animals were housed under standard conditions: room temperature 24 ± 2°C, humidity 50 ± 5% with a 12-hr light-dark cycle. To build the mouse model of IISD, C57BL6 mice were administered with irinotecan (50 mg/kg, i.p.) for 6 consecutive days according to our published paper (Sun, et al., 2020).

2.2.2. Model establishment procedures

(1). Dosing regimen.

To establish the IISD model, groups of F344 rats (n = 6–40, sex equal) received various doses and schedules of irinotecan via either i.v. infusion or i.v. injection in the tail vein (Table 3). The administration of 1150 mg/m²/day for two consecutive days resulted in the highest incidence rate of severe diarrhea (100%) with a low mortality rate (11%) in F344 rats. Therefore, this specific dose and administration schedule was chosen as an illustrative example to delineate the model establishment procedures, with other doses and administration schedules following a similar protocol. In the rats model of IISD, female and male rats were given irinotecan for two consecutive doses of 1150 mg/m²/day by i.v. infusion through the tail vein under mild anesthesia with isoflurane on day 1 and day 2. The parameters set for the Harvard PHD 2000 Infusion Syringe Pump were: infusion rate of 5 mL/hr, syringe diameter of 14.5 mm.

Table 3.

Effect of CPT-11 dose and administration schedule on the severity of late diarrhea and animal survival. To establish a rat diarrhea model, groups of 6–40 rats received single or multiple daily doses of irinotecan( formulation pH=3.0~3.5) at the indicated concentrations. Scoring of late diarrhea was conducted once daily. The severity of the diarrhea was scored using the scale in Table 2. Diarrhea results are reported as mean ± SE.

Species, age, body weight	Dose (mg/m2)	Dosing schedule	Dosing route	Incidence of IISD (%)		Relative B.W. loss on day 5 (%)	Mortality rate (%)
				Grade 3	Grade 4
F344, 6–8 weeks, male 180±15 g, female 120 ±10 g	30	QD×1,n=6	i.v., bolus	0	0	0	0
	120	QD×1,n=6	i.v., bolus	0	0	0	0
	800	QD×1,n=6	i.v. infusion	0	0	10±6	0
	1600	QD×2,n=6	i.v. infusion	16.6	0	15±5	0
	1150	QD×1,n=6	i.v. infusion	33.4	0	14 ±8	0
	2300	1150, QD×2, n=6	i.v. infusion	66.6	33.4	20±8	16.6
	2300a	1150, QD×2, n=6	i.v. infusion	66.6	33.4	22±7	16.6
	2300	1150, QD×2, n=40	i.v. infusion	62.5	37.5	22±7	11.1
	2500	1250, QD×2, n=18	i.v. infusion	50	50	20±9	22
	1500b	QD×1,n=4	i.v. infusion	no	no	no	100
	4600C	2 rounds, n=12	i.v. infusion	58	42	25±9	42
F344, 4–5 weeks, male 95±4 g, female 60 ±5 g	2300	1150, QD×2,n=8	i.v. infusion	25	0	Increased	0
Pirc rats, 40 weeks, female, 300±30 g	2300	1150, QD×2,n=4	i.v. infusion	N	100	N	100
FVB mice, 6–10weeks, female 18±5 g	900	150, QD×6,n=10	i.p.	30	30	20±9	100
Male and female C57BL6 mice, 6–10 weeks, 18±5g	900	150, QD×6,n=10	i.p.	30	20	20±8	100

Note: Relative body weight at day 5 was calculated for each animal relative to the animal’s weight at the beginning of the study (day 0). Infusion volume 0.8ml~3ml was based on the body weight of the rats, e.g. the bodyweights of young female F344 rats were 80 g, the infusion volume was 0.8ml; the Pirc female F344 rats were 300 g, the infusion volume was 3ml.

^ameans the infusion rate was 8 mL/hr and the irinotecan formulation pH was 5.0–5.5 (adjusted by 20mM HEPES buffer solution).

^bmeans the rats died after i.v. infusion.

^Cmeans the interval days was 21 days between the two rounds of chemotherapy. We didn’t give the rats atropine to treat acute diarrhea. The mortality% = the number of rats who died after two consecutively doses of irinotecan /total number of rats who received two consecutively doses of irinotecan *100%.

In the two cycles of irinotecan administration study, following a three-week recovery period from the first cycle of irinotecan administration (two consecutive doses of irinotecan 1150 mg/m²). The rats were administered another two consecutive doses of irinotecan 1150 mg/m² to mimic the dosing regimen of human’s chemotherapy (n=18).

(2). Anesthesia.

The determination of an optimal isoflurane concentration for anesthesia is a critical step in the model. Parameters for the Isoflurane Vaporizer for Small Animals & Rodents were set as: an O₂ flow rate of 0.5–1 L/min for rat anesthesia, and the vaporizer was set to 3% isoflurane. Upon initiating the intravenous infusion, the isoflurane concentration was adjusted to 1%, since a 3% concentration significantly reduced rat’s body temperature, blood pressure, and heart rate (Yang, et al., 2014), which would increase the mortality rate in our model.

(3). Procedure of Tail Vein Catheterization.

Briefly, the injection solution was drawn into a 10 mL syringe, then connected to a 26-gauze i.v. catheter. The air bubble in the catheter tubing was expelled as the infusion of the air bubble into the vein could cause lethal conditions, such as pulmonary embolism. Mature rats often require cleaning of the tail with 70% alcohol. The lateral tail veins were identified by slightly elevating and gently rotating the tip of the tail. The veins are located superficially just under the skin. Holding the tail under slight tension, the needle was inserted about 5–10 mm into the vessel with the bevel facing upwards and nearly parallel to the vein. The initial puncture should be within the caudal 1/3 of the tail. confirmation of the needle’s location could be achieved by gently pulling the plunger, with the possibility of observing a blood flash, though not always guaranteed. However, aspiration was avoided to prevent vein collapse. Upon successful insertion, characterized by minimal resistance (limit puncture attempts to three), the needle should be carefully removed and the syringe was then connected to the infusion pump.

(4). Animal care during the infusion.

Airway obstructions should be prevented during the experiments. The drug infusion or isoflurane can stimulate the lung and salivary gland secretion of liquid, hence, a cotton swab was used to clear the airway liquid. If the rat exhibited signs of respiratory distress (e.g., gasping or labored breathing), or if their gums, ears, or feet displayed a blue tinge (cyanosis), the drug infusion and anesthesia were temporarily halted until their skin color returned to pink (usually take 2–3 min for the recovery).

(5). Monitoring of diarrhea.

All rats were monitored for body weight change and fecal conditions throughout the experimental period post-irinotecan or vehicle administration. Diarrhea within 24 hours after the last irinotecan administration was classified as acute diarrhea, and that after 24 hours was classified as IISD. The severity of diarrhea was scored according to Table 2, with representative images of different diarrhea grades shown in Figure 1.

Table 2.

Criteria for diarrhea score in the rat model of IISD

Diarrhea Score	Diarrhea Characteristics
0	Normal (between 3/4 inches long and 1/4 inches thick, pellet-shaped/ American footballshape, > 3 droppings)
1	Loose stool (slight, slightly wet, and soft stool, > 3 droppings, size becomes smaller)
2	Watery diarrhea with shape (wet and loose stool with moderate perianal staining of the coat/ paper, size become smaller, significant feces staining )
3	Slimy diarrhea (unformed stool, staining of the coat/ paper, or no feces in 4 hr observation)
4	Severe watery diarrhea with visible bleeding, staining over legs and abdomen, dead after 48 hr of irinotecan administration

Figure 1.

Common fecal shape for diarrhea score evaluation. Proposed the common fecal scoring charts of rats using a 4-point scale. Grades 3 and 4 were considered as severe diarrhea. The paper towel methods collected stools without bedding chips-induced contamination and score error. Grade 0/normal (row 1) showed normal shape and size, grade 1 or 2/wet (middle) characterized by changing shaped stool with watermark (row 2), and grade 3 or 4 watery/bloody (bottom row 3 and 4) stools showed liquid consistency with no solid pieces, visible blood in the stool or paper, or fecal staining over legs and abdomen.

2.3. Histopathology assessment

Tissue samples, including the liver, ileum, and colon, were collected three days after irinotecan administration. The liver and ileum were thoroughly rinsed in saline and then fixed in Molecular Fixative (Somagen, Edmonton, AB). These fixed tissues were sent to Baylor College of Medicine - Human Tissue Acquisition & Pathology (Houston, TX, USA) for the processes of embedding, sectioning, and staining with hematoxylin and eosin (H&E).

2.4. Pharmacokinetics (PK) study of irinotecan in different treatment group rats

To assess the impact of dosage, drug formulation pH, infusion time, and the age of rat on the incidence of diarrhea, we conducted the irinotecan PK studies using our previously published methods (Basu, et al., 2016). Briefly, irinotecan injections ranging from 30 to 1150mg/m² were administered to the rats (4–6 weeks young rats and8–10 weeks adultn = 6) via tail vein i.v. infusion. For the dose of 30,120,1150,1500 mg/m² in the Table 3 was single dose, the others dose was twice consecutively. In all of the pharmacokinetics experiment, we collected the blood samples (approximately 30–50 μL) by tail snipping at specific time point after single dose infusion or injection. Plasma was collected after centrifugation (8,000 rpm, 3 min) and kept at −80 °C until analysis. The PK plasma samples were prepared by spiking blank solvent (50% methanol, 20 μL) into the collected PK plasma samples (20 μL) and extracting with 200 μL of acetonitrile: methanol (50:50, v:v) containing an internal standard (Baohuoside). The plasma sample were analyzed by UHPLC-MS/MS according to our published assay (Basu, Zeng, 2016). The pharmacokinetic parameters were calculated with WinNonlin 6.3 software, based on a non-compartmental model.

2.5. Measuring the carboxylesterase (CE) activities using irinotecan as a probe substrate

The assays for CE activities were carried out following previously published methods (Sun, Zhu, 2020). To determine the SN-38 formation rate, which represents the CE activities, liver and small intestine S9 fractions (2 mg protein/mL) prepared from 4 weeks and 10 weeks of rats were incubated with irinotecan (1, 5, and 50 μM) at 37°C in sodium phosphate buffer (50 mM, pH = 7.4). Samples were taken periodically (0.2 mL) and the reaction was stopped by adding 0.2 mL of acetonitrile containing 0.1% of formic acid and 50 nM of daidzin (I.S.). After vortexing and centrifugation at 15,000 g for 15 min, 10 μL of the supernatants were injected into UHPLC-MS/MS (Basu, Zeng, 2016). Samples incubated with boiled (100 °C for 5 min) S9 fractions were used as the negative control.

2.6. Statistical analyses

WinNonlin 8.3 (Pharsight, Mountain View, CA) was used to analyze the pharmacokinetics parameters of irinotecan, SN-38 and SN-38G with the non-compartmental model, including maximum concentration (C_max), time to reach maximum concentration (T_max), half-life (T_1/2), and area under the curve (AUC_0-t). The PK parameters were analyzed using the one-way ANOVA, and comparisons between the means of different dose, pH and were analyzed by Post Hoc Tests (Tukey). Other statistical analysis was performed by using an unpaired Students’ t-test. A p ≤0.05 was considered statistically significant.

3. Result and discussion

3.1. Irinotecan dose and administration schedule determined the incidence of IISD and animal mortality rate.

Based on the published data in Table 1, we optimized the IISD model, standardized rat model and provided our procedure in detail. The new model was proved to be highly reproducible, having been tested on over 200 rats and used to evaluate the anti-diarrhea activity of various compounds and herbs. The IISD was a dose-limiting toxicity since the incidence of diarrhea increased with the increased dose (Table 3).

The optimized model was developed using 8–10 weeks’ old F344 rats (equal representation of both sexes), given two consecutive doses of irinotecan at 1150 mg/m². This regimen yielded a 100% IISD incidence with a minimum mortality rate of 11%, offering a practical model for other researchers. Besides, this rat model could tolerate two cycles of chemotherapy, but the mortality rate was increased to 42% due to the IISD. Besides, rats did not exhibit severe delayed diarrhea (occurring 2 days post-irinotecan administration) when the dosage was lower than 800 mg/m² or 150 mg/kg. Conversely, dosages exceeding 1500 mg/m² or 250 mg/kg resulted in 100% mortality.

Compared to the rats’ IISD model, the incidence of IISD in the mouse was considerably lower at 60%, even after six consecutive doses of irinotecan. And all mice with IISD, unfortunately, perished on days 9 or 10. Besides, i.p. injection of the cytotoxic drug irinotecan may directly injure the intestine (Sgaragli, et al., 1993, Al Shoyaib, et al., 2019). Moreover, fecal collection and diarrhea score evaluation in mice were challenging due to the small size of their feces. Lastly, while the Mongolian gerbil animal is a typical choice for studying inflammatory bowel disease, this model was not used because the mouse couldn’t recover from the first cycle of irinotecan and the mechanism of diarrhea involved is different between IISD and IBD (Bleich, et al., 2010). For pigs, building the IISD model was expensive and inconvenient for lab use. Therefore, the rat model of IISD was superior compared to using mice and pigs.

The diarrhea assessment in rats is typically determined by evaluating fecal consistency, occult blood, and shape, which is referred to the Bristol stool chart of humans. The comprehensive fecal evaluation chart for rats was lacking. To address this gap, we first proposed a standardized fecal scoring chart for rats using a 4-point scale, which was effective for distinguishing severe diarrhea (Figure 1 and Table 2). To ensure accurate scoring, the paper towel methods were used for feces collection, minimizing contamination from bedding chips and sampling errors (Yim, et al., 2021). The only limitation of this method was time-consuming, requiring half a day to evaluate the diarrhea scores of approximately 30 rats on days 4 and 5.

3.2. The characteristic of the IISD in F344 wild-type rats with an optimized dose of 1150 mg/m².

The severe diarrhea of rats was significantly occurred due to irinotecan injection, reaching the highest score on day 4 and day 5 (grade 3 or 4). This led to a 100% incidence of IISD in rats, with 85% of rats recovering in 2 weeks after irinotecan administration (Figure 2, n = 24). Concurrently, the rats experienced significant weight loss after Irinotecan administration, with a maximum loss of 20% body weight on days 4 or 5. But the rat could regain the body weight to their initial levels after one week.

Figure 2.

The time-course change of diarrhea score, incidence, body weight and histology after irinotecan administration in rats(1150 mg/m²/day on two consecutive days). CPT-11 induced acute and late-onset diarrhea in male and female rats (A, n=6, equal sex). Red color indicates male rats; the blue color indicates female rats. The time-course changes of the incidence of diarrhea (B) and rat’s body weight in IISD model (C). The time-course histology study of the rat’s ileum and liver after irinotecan treatment (D).

To further study the irinotecan-induced intestine injury, we conducted a temporal histology study. The analysis revealed dramatic remodeling in the morphology of the ileum epithelium from day 3, including shortening of villi, epithelium vacuolation, dilation of the blood vessels, formation of multiple crypt abscesses, crypt loss, leukocyte infiltration, thinning of the intestinal walls, with the situation deteriorating the most on day 5 (Figure 2C, n=3). Despite these severe intestinal alterations, irinotecan administration did not cause significant liver injury.

3.3. The influence of sex, age, tumor-bearing conditions and irinotecan injection formulation on IISD

3.3.1. The impact of sex and body weight on the IISD model.

In the animal study, adult male rats of the same age typically weighed more than the female rats. Given the same dosage of 200 mg/kg irinotecan, male rats received a higher drug amount due to their heavier body weight. For example, the 8-week-old male rats had body weights of 180± 15 g and the females weighed 120 ± 10 g, and the amount of irinotecan given to the male rats was 1.48-fold higher than that given to the females. As a result, when dosages were calculated based on the body weight of the same age of male and female rats, the mortality rate for male rats was double compared to that of female rats after the administration of Irinotecan (Table 3).

However, when dosage calculation was based on body surface area rather than body weight, the mortality rate for male rats was comparable to females. This is because chemotherapy dosing based on body surface area is a better indicator of physiologic parameters that are responsible for drug disposition than body weight (Beumer, et al., 2012). If we calculate the dose based on body surface area, we found that male rats were 15% overdosed compared to female doses, because of the significant difference in the body weight between the female (120 ± 10 g) and male (180 ± 15, Table 3). This finding suggested that body surface area-based dosing was more suitable for IISD rat studies, as it eliminated sex-based disparities.

Therefore, we adjusted the dose based on body surface area. We used the dose of 200 mg/kg derived from the results observed from 8–10 weeks-old female rats, which had a body weight of 150±20 g (equivalent to 1150 mg/m²). We then applied this same dosage of 1150mg/m² to male rats that weighed 250±20 g (equivalent to 180 mg/kg). The incidence of severe diarrhea didn’t change, but the mortality rate of male rats was decreased to be comparable to that observed in female rats. This adjustment was further confirmed by comparing the PK profile when we dosed the rats based on body surface area. There was no significant difference in the PK parameters of irinotecan, SN-38, and SN-38G between female rats and male rats.

Thus, these observed differences were likely due to body weight differences rather than sex differences, and these could be eliminated when dosing based on body surface area.

3.3.2. The influence of multiple cycles of chemotherapy on the IISD.

The mortality rate among rats increased to 45% after the second cycle of irinotecan treatment (n=11), compared to 11.1% in the first cycle of irinotecan treatment (n=18). In the model group, the incidence of severe diarrhea reached 100% by day 5, but the second cycle of irinotecan administration extended the diarrhea duration time of the rats compared to the diarrhea duration time in the first cycle of irinotecan administration. The diarrhea score of the rats was all 0 on day 7 after the first cycle of irinotecan administration, but the rats still had grade 1 and 2 diarrhea by day 10 in the second cycle.

3.3.3. The effect of rat age on the IISD model.

4-week-old young F344 rats (Figure 3A, B, C, D) were used to build the IIISD model. In the four-week-old rats model of IISD, severe diarrhea was only observed in 25% of male rats (Figure 3C), and all rats recovered by day 5. Irinotecan administration affected their rate of weight gain compared to the control group rats (Figure 3B, D). Overall, young rats were less sensitive to the IISD, and the plasma AUC of SN-38 was much lower than that of the adult 10 weeks’ old rats (Figure 4A). Furthermore, the AUC ratio of SN-38 to CPT-11 in adult rats was 5.5-fold higher than in 4 weeks’ young rats (0.198 VS.0.036, Figure 4B).

Figure 3.

The diarrhea incidence and body weight change for young rats and tumor-bearing rats after two consecutive doses of 1150mg/m² irinotecan administration. The 4 weeks old rats (n=4) for both female (A, B) and male (C, D) were more resistant to the IISD. Pirc rats were more sensitive to toxicity compared to adult rats (E, F).

Figure 4.

The pharmacokinetics profile of Irinotecan in female 10 weeks (adult, n=6) and 4weeks female rats (n=3), the AUC ratio of SN-38 to irinotecan in plasma (B), and the carboxylesterase activity in the small intestine (C) and liver (D) using different concentration of irinotecan as a substrate.

This discrepancy stems from the difference in CE activity between 4 weeks’ rats and 8–10 weeks’ rats. To test this hypothesis, the CE activities of the liver and small intestine were measured using different concentrations of CPT-11 as a probe substrate. The results showed that the CE activities of the liver and small intestine in adult rats were 2–5 folds higher than the CE activities in 4-week rats. Therefore, the lower CE activities of young rats led to lower blood exposure to SN-38, which further reduced the incidence and severity of IISD. Consistent with our previous study, Mrp2 knockout mice showed low plasma exposure to SN-38 due to the low CE activities (Sun, Zhu, 2020).

3.3.4. The impact of tumor-bearing on the IISD model.

We also evaluated tumor-bearing’s impact on the IISD model using 44-week Pirc rats with polyposis growth in the colon (Figure 3E, F). The other reason why we chose F344 rats is that Pirc rats are a kind of classical intestinal-cancer-associated adenomatous polyposis coli gene-mutant rats, that mimic human colon cancer with an F344 background. In the Pirc rats IISD model, 2 out of 4 rats already had bloody fecal or wet feces before irinotecan treatment, and all rats showed IISD on day 3 and died on day 6 after two consecutive injections of irinotecan. Therefore, the tumor-bearing rat had a higher incidence of IISD compared to wildtype F344 rats.

Given that Pirc rats commonly display bloody or watery feces as they age (Irving, et al., 2014), we should take into consideration the effect of variability in polyp burden and fecal condition observed among individual rats on our experiment. moreover, we should evaluate the IISD model in tumor-bearing rats by using younger Pirc rat at 16 weeks in our future studies, which have less tumor burden and diarrhea occurrence comparing to the 44 weeks’ rats. Dose adjustment will be needed for these Pirc rats.

3.3.5. The impact of infusion rate and injection formulation pH on the IISD model.

We compared the PK profile of the Irinotecan by different infusion rates and injection formulation pH. The result showed that the PK parameters of Irinotecan didn’t change with increased infusion rate and pH (Table 3 and Table 5). Moreover, neither the infusion rate nor injection formulation pH affected the incidence of IISD.

Table 5.

The pharmacokinetics parameters of irinotecan by different infusion time and injection formulation pH (n=3–6, female, 10-week)

Dosage, formulation pH	Drug	T1/2 (hr)	Tmax (hr)	Cmax (μmol/L)	AUC (μmol/L*hr)
800 mg/m2 , (pH=3.0~3.5)	CPT-11	2.75 ± 0.32	0.55 ± 0.00	61.2 ± 6.15	189.3 ± 20.7
	SN-38	10.92 ± 7.19	0.88 ± 0.38	3.44 ± 1.07	13.94 ± 3.98
	SN-38G	36.4 ± 17.84	4.38 ± 3.32	0.21 ± 0.045	0.74 ± 0.11
1150 mg/m2 (pH=3.0~3.5)	CPT-11	3.36 ± 1.22	0.65 ± 0.00	82.5 ± 15.6*	263.7 ± 28.9*
	SN-38	15.27 ± 14.61	0.90 ± 0.00	2.68 ± 0.72	11.55 ± 2.33
	SN-38G	6.93± 0.00	3.53 ± 4.06	0.23 ± 0.064	0.94 ± 0.19
1150 mg/m2 (pH=5.0~5.5)	CPT-11	3.76 ± 0.40	0.82 ± 0.13	68.4 ± 2.8	220.6 ± 15.0
	SN-38	12.11 ± 4.13	0.73 ± 0.13	1.83 ± 0.42*	6.16 ± 1.37*
	SN-38G	20.09 ± 23.66	1.65 ± 2.33	0.45 ± 0.13*	1.84 ± 0.35*

Note: The infusion time for 1150 mg/m² and 800 mg/ m² was 0.4 hr and 0.3 hr, respectively.

^*means p<0.05, a significant difference compared to the group with a dosage of 800 mg/m².

4. Conclusion

We conducted a systemic evaluation of various factors that affect the success rate in establishing the IISD model. Our finding highlighted the critical impact of irinotecan dose, administration schedule, tumor-bearing status and age of rats. Notably, juvenile rats were more resistant to the IISD due to the lower blood SN-38 exposure induced from lower CE activity compared with adult rats. The only limitation of the rat IISD model was the severe body weight loss (10–20%) after irinotecan administration, which only had moderate (6%) in chemotherapy patients. Given the robust correlation between plasma SN-38 exposure and IISD, the pharmacokinetics parameters of SN-38 play a crucial role in the optimization of the IISD model.

Compared to the mice and pig’s model, this rat IISD model better mimics human pathophysiology progress and offers higher translational potential. The typical characteristics of the rat IISD model based on 200 rats across 6 batches of F344 rats, including the same administration route of irinotecan and similar diarrhea characteristic to those observed in patients (Bailly, 2019). Both rats and humans showed IISD 48 hours after irinotecan administration, marked by visible bloody diarrhea. Remarkably, the rats could recover from the first cycle of irinotecan administration and tolerate two cycles of irinotecan administration, which is not achievable in mice, thereby better mimicking the pathophysiology progress of patients with IISD than ever before. Moreover, the tumor-bearing Pirc rat can also be applied to study the IISD and mimicked the sign and symptoms of cancer in patients. Lastly, to improve the objectivity in diarrhea score determination, practical diarrhea-scoring criteria was also proposed. In conclusion, the rat model of IISD stands as a valuable tool for researchers encountering difficulties in evaluating the pathophysiology progress of the IISD and assessing the anti-diarrhea activity of various agents with high reproducibility.

Table 4.

The pharmacokinetics parameters of irinotecan 1150 mg/ m² in F344 rats (n=3 for male and n=3 for female,10 week-old)

Treatment type	Drug	T1/2 (hr)	Tmax (hr)	Cmax (μmol/L)	AUC (μmol/L*hr)
Female pH = 5.0~5.5	CPT-11	3.6 ± 0.50	0.82 ± 0.14	68.8 ± 3.00	220.9 ± 10.30
	SN-38	13.95 ± 0.40	0.82 ± 0.14	1.89 ± 0.11	6.45 ± 0.30
	SN-38G	27.18 ± 28.61	0.65 ± 0.00	0.38 ± 0.16	1.59 ± 0.31
Male pH = 5.0~5.5	CPT-11	3.93 ± 0.26	0.82 ± 0.14	67.93 ± 3.13	220.3 ± 21.38
	SN-38	10.27 ± 3.86	0.65 ± 0.00	1.78 ± 0.64	5.86 ± 2.10
	SN-38G	5.93± 0.00	2.65 ± 3.25	0.52 ± 0.06	2.09 ± 0.12

Note: The PK study for two consecutive doses 1500 mg/m² of irinotecan couldn’t be conducted due to a high mortality rate among rats (83%, 5 out of 6 rats) following irinotecan administration. The combination of stress from the blood sample collection and drug toxicity may contribute to the increased mortality rate.

Data Availability Statement

The authors declare that all the data supporting the findings of this study are contained within the paper.