|Home | About | Journals | Submit | Contact Us | Français|
In patients with diabetes treated with sunitinib symptomatic hypoglycaemia has been reported. To explore the mechanism of this adverse effect we performed a prospective study to investigate the effect of sunitinib on insulin concentration, insulin clearance and insulin sensitivity.
We studied the early effects of sunitinib on insulin sensitivity and insulin clearance with a hyperinsulinaemic euglycaemic clamp (insulin infusion rate 60 mU m−2 min−1; steady‐state 90–120 min) in patients with renal cell carcinoma before and 1 week after the start of sunitinib 50 mg day−1. Insulin sensitivity index (SI) was defined as steady‐state glucose disposal divided by the steady‐state plasma insulin.
Ten patients (one with diabetes, treated with metformin) were included in the study protocol. Steady‐state insulin concentrations during the clamp increased after 1 week of sunitinib (from 128.9 ± 9.0 mU l−1 to 170.8 ± 12.8 mU l−1, P < 0.05; 95% CI on difference − 64.3, −19.6). The calculated insulin sensitivity index decreased from 0.22 ± 0.04 before to 0.18 ± 0.02 μmol kg−1 min−1 per mU l−1 insulin (P < 0.05; 95% CI on difference 0.07, 0.08). As the insulin infusion rate was similar for both clamps, the increased steady‐state insulin concentration indicates reduced insulin clearance.
Sunitinib affects insulin clearance which could possibly lead to overexposure to insulin in patients using insulin or insulin‐secretion stimulating agents.
The development of tyrosine kinase inhibitors (TKIs) targeting the vascular endothelial growth factor (VEGF) has improved cancer treatment. However their use can be hampered by side effects. Sunitinib, a VEGFR‐TKI, is currently used for the treatment of metastatic renal cell carcinoma (mRCC), gastrointestinal stromal cell tumours (GIST) and pancreatic neuroendocrine tumours.
The most reported side effects of sunitinib are hypertension, fatigue, hand‐foot syndrome and diarrhoea 1, 2. Several case reports and retrospective studies mention the occurrence of hypoglycaemia during treatment with sunitinib in patients with diabetes mellitus treated with a sulfonylurea derivate or insulin 3, 4, 5, 6. To explore the mechanism of this adverse effect we performed a prospective study to investigate the effect of sunitinib on insulin concentration, insulin clearance and insulin sensitivity.
After approval of the protocol (NCT01227213) by the Institutional Review Board of the Radboud University Medical Center, a total of 10 mRCC patients with an indication to start sunitinib were recruited. All participants provided written informed consent. Patients were eligible if they had a life expectancy of more than 12 weeks, a WHO performance status of 0–2 and no evidence of severe or uncontrolled diseases other than renal cell carcinoma. Patients treated with corticosteroids or oral anti‐coagulants were excluded. All studies were performed at the Radboud University Medical Center according to institutional and Good Clinical Practice guidelines. In the week before and 1 week (7–10 days) after starting treatment with sunitinib (50 mg day−1), a 120 min hyperinsulinaemic euglycaemic clamp was performed.
The experiments were performed in the morning after an overnight fast. Patients took their medication in the morning of the experiment with a cup of water. Subjects were studied in the supine position in a temperature‐controlled room (23–24 °C). Before the start of the experiment blood samples were obtained for the measurement of fasting insulin and glucose. Two intravenous cannulae were inserted. One was positioned retrogradely into a dorsal vein of the hand that was placed in a plexiglass box, ventilated with heated air, for sampling of arterialized venous blood 7. The second cannula was inserted in an antecubital vein of the contralateral arm for infusion of insulin and glucose. During the clamp insulin (Insulin aspart; Novorapid®; NovoNordisk, Bagsvaerd, Denmark; diluted in NaCl 0.9% to a concentration of 1 U ml−1, with the addition of 2 ml whole blood per 50 ml) was infused at a rate of 60 mU min−1m−2 body surface area. Arterialized venous plasma glucose determinations were performed at 5 min intervals using an enzymatic‐amperometric method (Biosen C‐line GP+; EKF‐diagnostic GmbH, Barleben, Germany). Plasma glucose was clamped at glucose 5.0 mmol l−1 (i.e. 90 mg dl−1), to reach an euglycaemic state, by a variable infusion of glucose 20% solution. The validity of the glucose clamp measurements of insulin sensitivity depends on achieving steady‐state conditions. The t 1/2 of insulin in plasma is 4–6 min, so after 30 min insulin levels have already reached steady‐state. However glucose concentration and infusion should also reach steady‐state with a coefficient of variation of <5%. Therefore the period of 90–120 min of the clamp is as steady‐state. At 90 and 120 min after the start of the clamp, venous blood was sampled and after centrifugation the supernatant was stored at −80 °C until the measurement of insulin in all samples of all patients by radio‐immunoassay at the same time. Plasma insulin was assessed by an in‐house radioimmunoassay (RIA) using 125I‐labelled human insulin and anti‐human insulin antiserum raised in guinea pig. Bound and free tracer were separated by sheep anti‐guinea pig antiserum and precipitation by means of polyethylene glycol (PEG). The interassay coefficient of variation (CV) for the insulin measurement was 4.7% and the intra‐assay coeffecient of variation was 9.7% at a level of of 34 mU l−1.
For calculation of the whole body glucose disposal (M value) during the euglycaemic clamp the following formula was used:
The insulin sensitivity index [μmol kg−1 min−1 per mU l−1] is a measure of insulin sensitivity in relation to the plasma insulin concentration during the steady‐state of the euglycaemic clamp 8. The insulin sensitivity index was calculated by dividing the M value by the mean insulin concentration during the euglycaemic clamp.
Differences in parameters before and after the start of sunitinib were statistically analyzed using a parametric Student's t‐test for paired observations. Data are presented as mean ± SEM. A value of P < 0.05 was considered to be statistically significant.
Between March 2012 and January 2013 10 patients signed informed consent and were included in the study. All patients had mRCC and sunitinib was the choice of treatment as decided by their oncologist. Mean age was 59 (range 43–72) years. One patient had diabetes and used metformin. Sunitinib treatment for 1 week had no significant effect on fasting glucose, fasting insulin, BMI or blood pressure (Table 1). Creatinine increased significantly after the start of sunitinib. However several studies show that the pharmacokinetics of insulin aspart are not affected by impaired renal function 9, 10.
During the hyperinsulinaemic euglycaemic clamp, glucose concentrations were stable but mean concentrations were not completely similar (before 4.84 ± 0.10 mmol l−1 and after 1 week of sunitinib 5.06 ± 0.09 mmol l−1, P < 0.05). The mean coefficient of variation for blood glucose during steady‐state was <5% in both experiments. Sunitinib treatment did not increase whole body glucose disposal (M value) during the clamp (26.5 ± 3.7 μmol kg−1 min−1 before start vs. 28.5 ± 3.1 μmol kg−1 min−1, P > 0.05).
Mean plasma insulin concentrations (Table 1) obtained during the steady‐state of the clamp were significantly higher after 1 week of sunitinib treatment (128.9 ± 9.0 mU l−1 vs. 170.8 ± 12.8 mU l−1; P < 0.02)(95% CI on difference − 64.3, −19.6) (Figure 1), indicating reduced clearance of insulin as insulin was infused at a fixed rate.
Before start of sunitinib, the patients were insulin resistant compared with previously studied lean healthy volunteers 11, as shown by a calculated insulin sensitivity index of 0.22 ± 0.04 μmol kg−1 min−1 per mU l−1(95% CI on difference 0.07, 0.08). After 1 week of sunitinib, the calculated insulin sensitivity index decreased to 0.18 ± 0.02 μmol kg−1 min−1 per mU l−1 (P < 0.05) suggesting a decrease in insulin sensitivity.
The present study demonstrates for the first time that sunitinib reduces insulin clearance in patients with mRCC. This conclusion is based on the higher steady‐state insulin concentrations during the clamp, where insulin is infused in a fixed infusion rate when subjects were treated with sunitinib.
The mechanism of the reduced insulin clearance during sunitinib treatment is not yet known. Insulin clearance is a complex phenomenon and depends on metabolic degradation, which is thought to be largely receptor mediated 12, 13. The liver is the primary site for insulin clearance, removing 40–80% during the first portal passage 13. The first step in insulin clearance is binding to its receptor on the cell membrane activating the tyrosine kinase pathway leading to internalization of insulin followed by degradation in lysosomes 14, 15, 16, 17. In vitro studies have indicated that insulin internalization and insulin action are linked 18. Thus insulin‐induced insulin receptor activation not only mediates insulin actions but is also involved in its clearance. Sunitinib has shown binding affinity with the insulin receptor and inhibits the insulin‐like growth factor type 1 receptor inducing ubiquitination and thereby degradation of this receptor 19, 20. Therefore, it is likely that sunitinib can inhibit the kinase activity of the insulin receptor and consequently hamper insulin internalization and degradation.
Besides the effect on insulin clearance we observed a significant decrease in the calculated insulin sensitivity index, suggesting a decline in insulin sensitivity. However comparisons of insulin sensitivity during hyperinsulinaemic euglycaemic clamps are only valid if the same conditions are reached for all patients 21. In our subjects the same insulin infusion rate was used throughout the experiment, however insulin concentrations were significantly higher after the start of sunitnib and moreover glucose clamp levels differed significantly. Therefore we cannot draw firm conclusions from these measurements regarding insulin sensitivity. It is intriguing that the potential effect of sunitinib on insulin sensitivity is congruent with its effect on insulin clearance, supporting the notion that both effects are mediated by the interaction of sunitinib with the insulin receptor.
Some limitations to our study should be addressed. First, we did not observe an effect of sunitinib on fasting blood glucose and fasting insulin. However in our population of patients not treated with insulin or sulfonylurea derivatives, normal regulation of endogenous insulin secretion must have compensated for the reduced insulin clearance by reduced pancreatic release of insulin. A similar study in patients with diabetes would be interesting to confirm the effect of sunitinib on insulin clearance in this population. However this is not feasible considering the size of the patient population and the sample size needed. Second, C‐peptide concentration, as a measure of endogenous insulin production, was not measured. Under clamp conditions, C‐peptide concentrations as marker of endogenous insulin production are typically suppressed 22, 23. Third, creatinine increased significantly after the start of sunitinib. However several studies have shown that the pharmacokinetics of insulin aspart are not affected by this level of impaired renal function 9, 10. Fourth given the need to start active and standard treatment in patients with progressive metastatic disease, it was not deemed proportional to perform a placebo controlled study or to delay the start of treatment to exclude a sequence/time effect with the first aim of understanding the cause of reported hypoglycaemia.
In summary, treatment for 1 week with sunitinib reduced insulin clearance in patients with mRCC. In patients using insulin or insulin‐secretion stimulating agents the effect of sunitinib on insulin clearance could in theory result in overexposure to insulin and thereby induce hypoglycaemia. Further research into the effect of sunitinib on insulin levels in clinical practice should be performed to see whether the observed effect is of clinical relevance to patients.
All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare AT had financial support by a personal grant from the Dutch Cancer Society and The Netherlands Foundation of Cardiovascular Excellence for the submitted work, there are no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years and no other relationships or activities that could appear to have influenced the submitted work.
This study was presented as a poster at the European Society of Medical Oncology meeting in 2014.
Thijs A. M. J., Tack C. J., van der Graaf W. T. A., Rongen G. A., and van Herpen C. M. L. (2016) The early effect of sunitinib on insulin clearance in patients with metastatic renal cell carcinoma. Br J Clin Pharmacol, 81: 768–772. doi: 10.1111/bcp.12797.