*3.5. Measurement of Vmax, Km, and CLint of Tofacitinib in Hepatic Microsomes*

In rats with G-ARF and C-ARF, the *V*max values for the disappearance of tofacitinib in the hepatic microsomal protein decreased by 9.52 and 28.7%, respectively, but were not significantly different compared to that in control rats (Figure 6). *K*<sup>m</sup> values were comparable among the three groups (Figure 6). However, CLint for the disappearance of tofacitinib in the hepatic microsomal protein was significantly lower (54.4% decrease) in rats with C-ARF compared to that in control rats (Figure 6), suggesting that disappearance of tofacitinib could be slower in C-ARF rats. CLint was also lower (31.1% decrease) but not significantly different in G-ARF rats compared to that in control rats. Taken together, our data indicate that G-ARF or C-ARF affect hepatic function to inhibit the expression of CYP3A1/2 and CYP2C11, resulting in slower metabolism of tofacitinib.

**Figure 6.** Measurement of *V*max*, Km,* and CLint for the disappearance of tofacitinib in hepatic microsomes of control, G-ARF, and C-ARF rats (*n* = 3 per group): This experiment was performed three times, and data are expressed as mean ± standard deviation. Bars represent standard deviation. \* *p* < 0.05. *V*max: maximum velocity; *K*m: apparent Michaelis–Menten constant, the concentration at which the rate is one-half of *V*max for the metabolism of tofacitinib; CLint: intrinsic clearance; G-ARF: gentamicin-induced acute renal failure; C-ARF: cisplatin-induced acute renal failure.

#### **4. Discussion**

To establish acute renal failure, gentamicin and cisplatin were chosen. Gentamicin, a representative aminoglycoside antibiotic, induces moderate and reversible acute renal failure [32], while cisplatin, a chemotherapeutic drug, causes more severe and irreversible acute renal failure [33]. Both drugs accumulate in the renal tubule; produce reactive oxygen species (ROS) such as superoxide anion, hydrogen peroxide, and hydroxyl radical; and result in the induction of tubular necrosis and/or apoptosis [34–36]. The impaired renal function caused by these drugs was fully demonstrated, but liver damage did not seem to be serious in our preliminary study. Induction of acute renal failure by gentamicin and cisplatin was confirmed not only by a significant decrease in weight gain and CLCR but also by significant increases in 24-h urine output and plasma levels of urea nitrogen and creatinine than those in control rats. Renal biopsy also demonstrated the induction of acute renal failure.

The contribution of gastrointestinal excretion (including biliary excretion) as unchanged tofacitinib to CLNR of the drug seems to be nearly negligible. The GI24 h values were negligible in control, G-ARF, and C-ARF rats, i.e., less than 0.195% of the intravenous dose in the three groups (Table 1). This lower GI24 h did not appear to be caused by chemical or enzymatic degradation of tofacitinib in the rat's gastrointestinal tract; tofacitinib was stable when incubated for 24 h in various buffers of pH 2–10 [22] and in the gastric juice of rats (pH 3.5) (data not shown). Furthermore, according to a report by Lee and Kim [12], when 10 mg/kg tofacitinib was intravenously administered to rats (*n* = 3) after bile duct cannulation, biliary excretion of unchanged tofacitinib for 24 h was 0.703% of the intravenous dose, a nearly negligible contribution to CLNR of tofacitinib. Therefore, the CLNR value shown in Table 1 may represent the metabolic clearance of tofacitinib.

The AUCs of tofacitinib were not dose proportional after intravenous doses over 20 mg/kg and oral doses over 50 mg/kg were administrated [12]. A dose of 10 mg/kg of tofacitinib was chosen for the intravenous study, and 20 mg/kg of tofacitinib was selected for the oral study. After intravenous administration of tofacitinib to G-ARF and C-ARF rats, its AUCs were significantly higher, possibly as a result of significantly slower CL than in control rats. The lower CL of tofacitinib was attributable to significantly decreased CL<sup>R</sup> and CLNR of the drug in rats with G-ARF and C-ARF than in controls. The lower CL<sup>R</sup> may have been due to both significantly lower *Ae*0–24 h and higher AUCs in G-ARF and C-ARF rats. The higher AUC in rats with G-ARF and C-ARF was due to lower *Ae*0–24 h and lower CLNR than those in control rats. The lower *Ae*0–24 h in G-ARF and C-ARF rats could have been due to impaired kidney function. Tofacitinib did not show a urine flow rate-dependent timed-interval CL<sup>R</sup> in rats; a straight line was not found between 1/timed-interval CL<sup>R</sup> of tofacitinib and 1/urine flow rate among the three groups [37]. Greater urine output did not result in higher *Ae*0–24 h of tofacitinib, indicating that tofacitinib was not predominantly reabsorbed in the renal tubule. The significantly greater urine output in rats with G-ARF and C-ARF was because reabsorption of water was decreased in the renal tubule due to decrease in protein expression of aquaporins caused by gentamicin and cisplatin [38]. However, *Ae*0–24 h of some drugs showed a urine flow rate-dependent timed-interval CL<sup>R</sup> in rats; the lower the urine output, the lower the *Ae*0–24 h [37], which resulted in a straight line between 1/timed-interval CL<sup>R</sup> and 1/urine flow rate in both control and uranyl nitrate-induced acute renal failure (U-ARF) rats [23,39].

The CL<sup>R</sup> values of tofacitinib were estimated from free (unbound to plasma proteins) fractions in plasma; the values thus estimated were 5.99, 1.83, and 0.0856 mL/min/kg for control, G-ARF, and C-ARF rats, respectively, based on 20.7% plasma protein binding of tofacitinib measured by equilibrium dialysis [22]. The CL<sup>R</sup> values of tofacitinib were faster than their respective CLCRs in control and G-ARF rats, but CLRs of tofacitinib and CLCR in C-ARF rats were comparable each other, suggesting that tofacitinib is mainly excreted in urine via active secretion for control and G-ARF-rats [12,18,24] and in glomerular filtration for C-ARF rats. This was also supported by control and U-ARF rats; CL<sup>R</sup> of metformin and chlorzoxazone were faster than CLCR in control rats, but CL<sup>R</sup> of both drugs and CLCR were comparable in U-ARF rats, and thus both drugs were mainly excreted in urine via active secretion for control rats and glomerular filtration for U-ARF rat [23,39]. As shown in the results, both gentamicin and cisplatin induced acute renal failure by tubular necrosis through ROS production, but cisplatin induced more severe renal failure and seemed to completely inhibit the function of active secretion [40].

Based on the AUC difference between intravenous and intraportal administration of tofacitinib to rats, the first-pass metabolism of tofacitinib by the liver after reaching the portal vein was approximately 42.0% [12]. Therefore, tofacitinib has a characteristic with an intermediate hepatic extraction ratio and its hepatic clearance could be changed by both the hepatic CLint and the hepatic blood flow rates [41]. Thus, a significantly lower CLNR of tofacitinib when administered intravenously to rats with G-ARF and C-ARF could have been due to a significantly lower CLint for the elimination of tofacitinib in the liver. The reduced protein expressions and activities of the hepatic CYP3A1/2 and CYP2C11 subfamily could have been responsible for the lower hepatic CLint in G-ARF and C-ARF rats. Similar results with regard to changes in hepatic CYP3A1/2 and/or CYP2C11 isozymes have been reported in acute renal failure rats induced by glycerol, bilateral ureter ligation, and nephrectomy [42]. Surgically induced chronic kidney disease rat models also showed lower levels of hepatic CYP3A and CYP2C subfamilies compared to sham-operated control rats [43]. Consistent with our result in renal failure rat models, the CYP3A subfamily also decreased in patients with end-stage renal failure [44].

After intravenous administration of tofacitinib to control, G-ARF, and C-ARF rats, its *V*ss values were not significantly different among the three groups of rats, since the free fractions of tofacitinib in the plasma from control, G-ARF and C-ARF rats were comparable (data not shown). Similar results were also reported for cyclosporin [45] in G-ARF rats and tacrolimus [46] in C-ARF rats, whereas the *V*ss of metformin [23]; omeprazole [26]; and DA-1131, a new carbapenem [47] in U-ARF rats, significantly increased compared to that in control rats, which was due to the increase in free fraction of the drugs.

After oral administration of tofacitinib to rats with G-ARF and C-ARF, the AUC values were also significantly higher than in control rats. The absorption of tofacitinib from the gastrointestinal tract was almost complete among all three groups; GI24 h values were less than 1.27% of oral dose for control, G-ARF, and C-ARF rats. Therefore, absorption is not a factor for the higher AUCs in rats with G-ARF and C-ARF. However, decreased absorption after oral administration of azosemide [48]; oltipraz [49]; and YJA-20379-8, a new proton pump inhibitor [50], to rats with U-ARF has been reported. Although the expression of CYP3A1/2 and CYP2C11 in the intestine markedly increased in rats with G-ARF and C-ARF compared to those in control rats, AUC of tofacitinib increased in rats with G-ARF and C-ARF. The CLint for the disappearance of tofacitinib in the intestinal microsome was not measured in this study because the active site of CYP3A1/2 and CYP2C11 in the intestine was sensitive and very unstable [51]. It has been reported that CYP3A activity in the intestine was increased in renal failure models induced by cisplatin, glycerol, bilateral ligation, or nephrectomy [42]. The AUC increase in rats with G-ARF and C-ARF might be because active secretion of tofacitinib was reduced by tubular necrosis caused by gentamicin and cisplatin. Thus, tofacitinib accumulated in the body, resulting in significantly lower *Ae*0–24 h of oral dose and significantly lower CL<sup>R</sup> along with increased AUCs in rats with G-ARF and C-ARF than in control rats (Table 2). In addition, considering that approximately 21.3% of the oral dose was metabolized in the liver of control rats after oral administration of tofacitinib [12], the hepatic first-pass effect seemed to decrease in rats with G-ARF and C-ARF after absorption of tofacitinib into the portal vein. This was likely due to decreased hepatic enzyme activity, and protein expression of CYP3A1/2 and CYP2C11 in rats with G-ARF and C-ARF, which also contributed to the increase in AUC of tofacitinib after oral administration of the drug.

Consistent with our data in renal-failure rat models, results of a previous study showed that CYP3A subfamily decreased in patients with end-stage renal failure [44]. In patients with severe renal impairment, the plasma concentration of tofacitinib was significantly increased and, thus, the AUC of tofacitinib in these patients was higher than twice that in normal healthy subjects, suggesting that the reduction of tofacitinib dosage is recommended in patients with severe renal failure [18]. Because the renal excretion of tofacitinib was different between human and rats (approximately 30% of oral tofacitinib in human [9] and 6.21% of oral dose in rats), it is difficult to clearly conclude the clinical significance of the rat's results, but it seems clear that the increase of AUC in the renal failure state was due to slower hepatic metabolism and smaller urinary excretion of the drug. Our study could be applied to drug–drug interactions in clinical practice when administered in combination with CYP3A4 and CYP2C19 inhibitors, such as itraconazole and erythromycin, which may result in an increase in plasma concentration of tofacitinib by reduced nonrenal elimination of tofacitinib. Therefore, it is necessary to consider the dose reduction of tofacitinib when AUC increases twice or more.
