The impact of sunitinib N-oxide as a photodegradation product of sunitinib

During treatment with sunitinib, dosage adjustment according to the monitored blood concentration of sunitinib and SU12662 is considered useful. On the other hand, the appearance of hand-foot skin reaction (HFSR) cannot be explained by blood sunitinib concentration alone. Although light exposure greatly affects skin disorders associated with medication use, the photodegradation of sunitinib has not been studied in detail. Here, we investigated the photodegradation products of sunitinib using LC-MS and examined cytotoxic activities using an MTT assay. N-desethyl sunitinib and sunitinib N-oxide were identified as photodegradation products, and their concentrations increased under irradiation in a time-dependent manner. Although the IC50 value of N-desethyl sunitinib in the HEK 293 cell line (11.6 µmol/L) was similar to that of sunitinib (8.6 µmol/L), the IC50 value of sunitinib N-oxide (121.9 µmol/L) was over 10 times higher than that of sunitinib. In addition, N-desethyl sunitinib and sunitinib N-oxide were found in blood obtained from a patient taking sunitinib (24.7 and 2.3 ng/mL, respectively). Because the appearance of adverse drug reactions associated with sunitinib can be reduced by using α-tocopherol nicotinate, which has a strong antioxidant effect, we believe that sunitinib N-oxide might strongly promote the development of HFSR.Keyword : sunitinib, sunitinib N-oxide, photodegradation product, light


Introduction
Sunitinib, a multi-targeted tyrosine kinase inhibitor, is used as a first-line drug treatment for metastatic renal cell carcinoma (RCC) and is considered to be one of the key drugs for treating RCC [1][2][3]. The antitumor activity of sunitinib depends on its concentration in the blood and, in a meta-analysis of clinical trials of patients taking sunitinib for metastatic RCC and gastrointestinal stromal tumor, patients with a high cumulative area under the concentration-time curve (AUCcum) of total sunitinibreflecting the total amount of sunitinib and SU12662, an active metabolite of sunitinibhad a significantly longer time to tumor progression and overall survival [4]. In addition, some adverse drug reactions (ADRs) such as anorexia and fatigue were reported to be correlated with total sunitinib concentration, and the mean total sunitinib concentration is also reported to be higher in patients with bleeding events than in those without them [5].
Thus, during treatment with sunitinib, dosage adjustment according to the monitored concentration of sunitinib and SU12662 in blood is considered useful [6][7][8][9]. Noda and colleagues [5] have also reported that some dose-limiting toxicities of sunitinib, such as hand-foot skin reaction (HFSR), hypertension, and blood toxicity, developed irrespective of the total sunitinib concentration. In addition, the frequency of HFSR is significantly higher in Japanese people, although there is no significant difference in the blood concentration of sunitinib between Japanese and Western populations [3,10]. Thus, the appearance of HFSR cannot be explained solely by a high blood concentration of sunitinib.
Although HFSR due to sunitinib has been considered to be caused mainly by damage to the capillary endothelium due to physical pressure and VEGF inhibition, the precise mechanisms remain unclear [11]. The causes of skin disorders linked to the use of various drugs have been studied, and several factors and mechanisms have been elucidated. As an example, light-dependent photosensitivity is reported to be caused by the interaction of an active substance generated by light exposure with constituent components in the body [12]. In addition, skin disorders linked to a new quinolone antibacterial agent are caused by photoreactive substances generated by exposure of a drug or its metabolites in the skin to light [13]. As mentioned above, light exposure has been considered to strongly promote the appearance of skin disorders associated with medications [14].
Sunitinib and SU12662 have been reported to be converted from the Z form to the E form by photoinduced isomerization upon exposure to light in water [15]. On the other hand, because the sum of the E and Z forms after irradiation of sunitinib did not coincide with the original drug amount in that study, it is conceivable that other

Materials
Sunitinib and sorafenib were purchased from LC Laboratories (Woburn, MA). Industries (Osaka, Japan). All other reagents were obtained from commercial sources, and those used for analysis were graded for high-performance liquid chromatography, liquid chromatography-mass spectrometry (LC-MS), or analytical use.

Sample preparation
Sunitinib, sunitinib N-oxide, N-desethyl sunitinib, and sorafenib were dissolved in methanol and diluted to 1.0 mg/mL with methanol as stock solutions. Samples were stored in light-proof bottles at −20°C.

Analysis of the photodegradation products of sunitinib
Sunitinib stock solution was diluted to 500 ng/mL with 50% methanol and exposed to room or UV (185 nm) light at room temperature. Samples were collected 0, 24, 48, and 72 h after the start of the irradiation and photodegradation products were detected and quantified using time-of-flight (TOF) MS. The structures of these products were determined using quadrupole MS/MS (qMS/MS).

Analysis of photodegradation products
In TOF MS analysis of sunitinib exposed to UV light for 72 h, two clear signals were found at m/z 415.213 and 371.189 (Fig. 1). In qMS/MS analysis, signals at m/z 326, 283, and 255 were found as fragment ions of m/z 415.2. Because the pattern of these fragment ions coincided with that obtained from m/z 399.2 of sunitinib, a photodegradation product found as the m/z 415.2 ion was identified as a sunitinib oxide, which was generated by oxidation of the nitrogen atom side of sunitinib, to which two ethyl groups bind (Fig. 2). Similarly, clear signals at m/z 326, 283, and 255 were found as fragment ions of m/z 371.2, and a photodegradation product found as the m/z 371.2 ion was identified as a deethylate of sunitinib that was generated by deethylation of the tertiary amine of sunitinib (Fig. 2). In addition, in LC-qMS/MS analysis, the retention

Discussion
The risk of developing HFSR, one of the dose-limiting toxicities of sunitinib, has not been correlated with the blood concentration of sunitinib, and so elucidation of the factors leading to the onset of HFSR is required. We focused on the skin reactions accompanying light irradiation, considered an important clinical problem for many medicines, and investigated the photodegradation products of sunitinib. We found for the first time that sunitinib N-oxide is generated by UV irradiation of sunitinib and that it is also present in the blood of a patient taking sunitinib. Although sunitinib N-oxide was found in plasma and urine as a micro-decomposition product in an in vivo rat study, its contribution to drug efficacy and adverse and pharmacological effects has not been studied [16].
Recently, α-tocopherol nicotinate was reported to reduce the appearance of ADRs related to sunitinib [17]. The authors found that ADR was reduced due to the hydrogen peroxide trapping efficacy of α-tocopherol nicotinate. However, because αtocopherol nicotinate has a strong antioxidant effect, we believed that the ADRs of sunitinib could be suppressed via an antioxidant-mediated decrease in sunitinib N-oxide generation. Because we have not assessed the impact of sunitinib N-oxide on the appearance of HFSR, we need to fully explore the influence of sunitinib N-oxide in the human body and the relationship between sunitinib N-oxide level and HFSR onset in future research.
In an MTT assay, the cytotoxic activity of sunitinib N-oxide was found to be lower than that of sunitinib and N-desethyl sunitinib. These data suggested that the effect of sunitinib N-oxide on humans may be different from that of sunitinib and N-desethyl sunitinib. However, we could not rule out the possibility that the differences in intracellular uptake of each substance may have affected the MTT assay results, and we also did not evaluate the pharmacological effects of sunitinib N-oxide in detail. At least for this point, the effect on sunitinib N-oxide on ADRs remains to be clarified and requires further in-depth study.
In the analysis of sunitinib and sunitinib-related compounds using LC, the retention time of sunitinib N-oxide was very similar to that of sunitinib. Because they both have the same basic structure, it may be difficult to distinguish these compounds by UV detection and accurate long-term separation analysis is needed to separately quantify these compounds using a UV detector [18]. Because the activity of sunitinib and sunitinib N-oxide was indicated to be different in our study, we believe that LC-MS/MS but not LC-UV/Vis is suitable for the separate routine clinical analysis of sunitinib and sunitinibrelated compounds.
In conclusion, we found that sunitinib N-oxide was generated by UV irradiation of sunitinib and could be detected in the blood of a patient taking sunitinib. Although the pharmacological effects of sunitinib were not clarified, we believe that sunitinib N-oxide might strongly affect the appearance of ADRs because it has been reported that the ADRs induced by sunitinib can be ameliorated by antioxidant treatment. We aim to study the distribution and pharmacological effects of sunitinib N-oxide and assess its influence on the development of ADRs of sunitinib.