The anti-diabetic drug metformin protects against chemotherapy-induced peripheral neuropathy in a mouse model

Abstract

Chemotherapy-induced peripheral neuropathy (CIPN) characterized by loss of sensory sensitivity and pain in hands and feet is the major dose-limiting toxicity of many chemotherapeutics. At present, there are no FDA-approved treatments for CIPN. The anti-diabetic drug metformin is the most widely used prescription drug in the world and improves glycemic control in diabetes patients. There is some evidence that metformin enhances the efficacy of cancer treatment. The aim of this study was to test the hypothesis that metformin protects against chemotherapy-induced neuropathic pain and sensory deficits. Mice were treated with cisplatin together with metformin or saline. Cisplatin induced increased sensitivity to mechanical stimulation (mechanical allodynia) as measured using the von Frey test. Co-administration of metformin almost completely prevented the cisplatin-induced mechanical allodynia. Co-administration of metformin also prevented paclitaxel-induced mechanical allodynia. The capacity of the mice to detect an adhesive patch on their hind paw was used as a novel indicator of chemotherapy-induced sensory deficits. Co-administration of metformin prevented the cisplatin-induced increase in latency to detect the adhesive patch indicating that metformin prevents sensory deficits as well. Moreover, metformin prevented the reduction in density of intra-epidermal nerve fibers (IENFs) in the paw that develops as a result of cisplatin treatment. We conclude that metformin protects against pain and loss of tactile function in a mouse model of CIPN. The finding that metformin reduces loss of peripheral nerve endings indicates that mechanism underlying the beneficial effects of metformin includes a neuroprotective activity. Because metformin is widely used for treatment of type II diabetes, has a broad safety profile, and is currently being tested as an adjuvant drug in cancer treatment, clinical translation of these findings could be rapidly achieved.

Figure 1. Effect of metformin on cisplatin- or paclitaxel-induced mechanical allodynia in mice.

A). Treatment schedule; X: injection; O: no treatment. B) Mice (n = 10–12/group) were treated with cisplatin (cumulative dose 23 mg/kg i.p.) and metformin (200 mg/kg/dose i.p.) as depicted in panel A. Mechanical allodynia was quantified with von Frey hairs using the up and down method. C). Effect of delayed metformin treatment on cisplatin-induced mechanical allodynia. Mice (n = 8/group) received i.p. injections with cisplatin (cumulative dose 23 mg/kg i.p.) and delayed metformin (200 mg/kg/dose i.p.) as depicted in panel A and mechanical allodynia was monitored. D). Mice (n = 4–7/group) were treated with paclitaxel (10 mg/kg/every other day, i.p. for two weeks) and metformin (200 mg/kg i.p. daily from one day before until one day after paclitaxel) and mechanical allodynia was measured. E). Change in body weight after cisplatin and metformin treatment. ** p<0.01 vs. Saline+Saline group; ## p<0.01 vs. Saline+Cisplatin; && p<0.01 vs. Saline+Paclitaxel.

Figure 2. Effect of metformin treatment on cisplatin-induced sensory deficits.

A). Mice (n = 6–10/group) were treated with cisplatin and metformin as described in Figure 1. The time to respond to an adhesive patch on the hind paw was monitored. B). Effect of delayed metformin on cisplatin induced sensory deficits (n = 7–8/group). C). Effect of cisplatin treatment on rotarod performance. Mice (n = 6/group) were treated with cisplatin and their performance on the rotarod was monitored as an index of motor coordination. E). Mice (n = 7/group) were treated with lidocaine (5 µL, 4% in saline) or saline and 10 mins. later, the time to respond to an adhesive patch on the hind paw was measured. * p<0.05, ** p<0.01 vs. Saline+Saline; # p<0.05 vs. Saline+Cisplatin.

Figure 3. Effect of metformin on loss of intraepidermal nerve fibers induced by cisplatin.

Mice (n =  4–6 per group) were treated with cisplatin and metformin as in figure 1. Paw biopsies obtained from the hind paw at 5 weeks after the start of treatment were stained for intraepidermal nerve fibers (PGP9.5; red) and collagen (green). (A) saline/saline; (B) saline/cisplatin; (C) metformin/cisplatin. White arrows indicate the intraepidermal nerve fibers stained by PGP9.5 (in red), which can clearly be seen crossing the basement membrane (in green) and extending as long lines into the epidermis. (D) Quantification of intraepidermal nerve fiber density. ** p<0.01 vs. Saline+Saline; ## p<0.01 vs. Saline+Cisplatin.