Fasting protects against the side effects of irinotecan treatment but does not affect anti-tumour activity in mice

Abstract

Background: The main limitation to the use of irinotecan in the treatment of colorectal cancer is the severity of side effects, including neutropaenia and diarrhoea. Here, we explored the effects of 3 days of fasting on irinotecan-induced toxicities, on plasma, liver and tumour pharmacokinetics and on anti-tumour activity in mice.

Experimental approach: Male BALB/c mice received C26 colon carcinoma cells subcutaneously. They were randomized 1:1 into equally sized ad libitum fed and fasted groups after which they were treated with irinotecan. Weight and adverse side effects were recorded daily. At the end of the experiment, tumours were resected and weighed, and concentrations of irinotecan and its active metabolite SN-38 were determined in plasma and tumour.

Key results: Fasting prevented the diarrhoea and visible signs of discomfort induced by irinotecan. Ad libitum fed animals developed leucopenia compared with untreated controls, whereas fasted mice did not. Irinotecan suppressed tumour growth equally in both treated groups, compared with untreated controls. Levels of the active irinotecan metabolite SN-38 9 (calculated as AUC values) were significantly lower in fasted mice in both plasma and liver, but not in tumour tissue.

Conclusions and implications: Fasting protected against irinotecan-induced side effects without interfering with its anti-tumour efficacy. Fasting induced a lower systemic exposure to SN-38, which may explain the absence of adverse side effects, while tumour levels of SN-38 remained unchanged. These data offer important new approaches to improve treatment with irinotecan in patients.

Figure 1

Protective effects of fasting against irinotecan toxicity. A total of 24 mice, six mice per group, were observed for adverse side effects for 10 days after the first irinotecan injection; irinotecan was given i.p. on days 1, 3 and 5 after fasting (shown by *). (A) Effect of fasting on body weight. Fasted (F) mice lost about 20% of their body weight during 3 days' fasting. During irinotecan treatment, ad libitum (AL) fed mice showed 20% weight loss, while fasted mice gained weight and reached their starting weight at day 5. (B–E) Effects of fasting on activity (B); coat (C); diarrhoea (D) and the posture of the mice (E). Data shown are means ± SEM (n = 6). **P = 0.003, *** P < 0.001, significantly different from (F + irino) group; one‐way ANOVA with Bonferroni's correction. (F) Effect of fasting on circulating leukocyte numbers, measured on day 8 after the first irinotecan injection. Fasting alone did not affect leukocyte levels. The mice fed ad libitum and treated with irinotecan (AL+ irino) showed leukopenia compared with fasted mice treated with irinotecan (F + irino). *** P < 0.001, significantly different as indicated; one‐way ANOVA with Bonferroni's correction.

Figure 2

Effect of irinotecan treatment on tumour growth in fasted and ad libitum fed groups. At the end of the experiment, irinotecan‐treated tumours were significantly smaller in both the fasted (F) and ad libitum (AL) fed groups compared with untreated controls. Data shown are means ± SEM (n = 6). ***P < 0.001, significantly different as indicated; one‐way ANOVA with Bonferroni's correction. Fasting alone had no significant effect on tumour weight.

Figure 3

Pharmacokinetics of irinotecan (circles) and SN‐38 glucuronide (triangles) and SN‐38 (squares) in plasma, tumor and liver. Concentration‐Time curves are presented for ad libitum fed groups (black symbols), fasted weight‐adjusted dosed (white circles) and fasted flat‐fixed dosed (grey squares). Fasted weight‐adjusted dosed and fasted flat‐fixed dosed animals show a trend towards higher plasma irinotecan AUC values, while irinotecan AUC values in the liver and tumor are lower in fasted animals. SN‐38 AUC values are lower in both fasted groups in plasma and liver, and SN‐38G AUC values are lower in plasma. SN‐38 AUC values in tumor tissue show large variation.

Figure 4

Statistical validation of pharmacokinetic differences. Data shown are the AUC values for each experimental group. No significant differences were found for irinotecan AUC values between ad libitum (AL) fed, weight‐adjusted dosed fasted (F) and flat‐fixed dosed fasted (F flat) groups in plasma (A + B), liver (D + E) and tumour (F + G). SN‐38 AUC values in plasma (B) were significantly lower in weight‐adjusted dosed fasted animals (*P = 0.02), and SN‐38G AUC values in plasma (C) were significantly lower in weight‐adjusted dosed fasted animals and fasted flat‐fixed dosed animals *P = 0.04, **P = 0.01. SN‐38 AUC values in the liver (E) were significantly lower in both fasted groups **P = 0.003. No significant differences were seen in SN‐38 AUC values for tumour tissues; P = 0.27. n = 3; differences as indicated; Student's t test.

Figure 5

CES2 mRNA expression in liver and tumour tissue from ad libitum fed and fasted mice. CES2 expression at 1 (t = 1), 8 (t = 8) and 12 h (t = 12) after irinotecan injection in liver and tumour tissue from ad libitum (AL) fed and fasted (F) animals did not reveal any significant differences.