*3.2. Hepatic First-Pass E*ff*ect of Tofacitinib in Rats*

Figure 3A shows the mean arterial plasma concentration-time profiles following intravenous and intraportal administration of 10 mg/kg tofacitinib, and Table 3 shows the associated pharmacokinetic parameters. The mean arterial plasma concentrations of tofacitinib administered intravenously and intraportally showed polyexponential reductions. The AUCs were 417 and 242 µg·min/mL, respectively, demonstrating considerable hepatic first-pass metabolism of tofacitinib after absorption into the portal vein, with 42.0% of the intravenous dose metabolized in the liver before entering the systemic circulation. As a result, the CL and CLNR of tofacitinib were 67% and 60% faster, respectively, after intraportal administration. Furthermore, *V*ss was 50% higher after intraportal than after intravenous administration of tofacitinib (*p* < 0.05).

**Figure 3.** Mean arterial plasma concentration-time profiles of tofacitinib in Sprague-Dawley rats after (**A**) 30-min intravenous (*n* = 7) and intraportal (*n* = 7) infusions of 10 mg/kg tofacitinib and (**B**) 30-min intraportal (*n* = 6) infusion, and intraduodenal (*n* = 5) and intragastric (*n* = 5) instillations of 10 mg/kg tofacitinib. Bars represent standard deviations (SD).


**Table 3.** Pharmacokinetic parameters of tofacitinib after 30-min intravenous and intraportal infusion of the drug at dose of 10 mg/kg to male Sprague-Dawley rats.

Data are expressed as mean ± standard deviation (SD). \* *p* < 0.05; \*\* *p* < 0.01.

## *3.3. Gastric and Intestinal First-Pass E*ff*ects of Tofacitinib in Rats*

Figure 3B shows the mean arterial plasma concentration-time profiles following intragastric, intraduodenal, and intraportal administration of 10 mg/kg tofacitinib, and Table 4 shows the associated pharmacokinetic parameters. The AUCs of intragastrically and intraduodenally administered tofacitinib did not differ significantly (134 and 138 µg·min/mL), suggesting that the gastric first-pass effect of tofacitinib was negligible. In contrast, AUC was significantly lower after intraduodenal (138 µg·min/mL) than after intraportal (272 µg·min/mL) administration, indicating that the intestinal first-pass effect of tofacitinib was significant, with approximately 49.3% of the orally administered drug removed prior to entry into the portal vein.


**Table 4.** Pharmacokinetic parameters of tofacitinib after 30-min intraportal infusion, intraduodenal and intragastric instillation of the drug at dose of 10 mg/kg to male Sprague-Dawley rats.

Data are expressed as mean ± standard deviation (SD). <sup>a</sup> Intraportal infusion was significantly different (*p* < 0.05) from intraduodenal and intragastric instillation. <sup>b</sup> Intraportal infusion was significantly different (*p* < 0.001) from intraduodenal and intragastric instillation.

#### *3.4. Tissue Distribution of Tofacitinib in Rats*

Figure 4 shows the concentrations of tofacitinib in plasma (µg/mL) and in tissue samples (µg/g) and its tissue-to-plasma (T/P) ratios 30 min (distribution phase) and 2 h (elimination phase) after intravenous administration of 10 mg/kg tofacitinib. Tofacitinib was widely distributed in all rat tissues, with T/P ratios greater than 1.0 in every tissue except the brain, mesentery, and fat, both 30 min and 2 h after intravenous administration. At 30 min, tofacitinib was exclusively distributed in the kidneys, small intestine, and large intestine, with its concentrations remaining stable until 2 h after intravenous administration.

**Figure 4.** (**A**) Mean plasma and tissue/organ concentrations of tofacitinib and (**B**) tissue-to-plasma (T/P) ratios of tofacitinib 30 min (*n* = 4) and 2 h (*n* = 4) in Sprague-Dawley rats after 1-min intravenous infusion of 10 mg/kg tofacitinib. Data are expressed means ± standard deviations (SD). LI; large intestine, SI; small intestine.

#### *3.5. Biliary Excretion of Tofacitinib in Rats*

Following a 1-min intravenous infusion of 10 mg/kg tofacitinib, less than 1% of the intravenous dose (0.703 ± 0.303%) was excreted in bile of each of the three rats studied, suggesting that biliary excretion of tofacitinib is a minor elimination pathway.

#### **4. Discussion**

The present study found that the dose-normalized AUC of tofacitinib was dependent on the administered dose. Plots of AUC versus dose for intravenous and oral tofacitinib yielded slopes of 2.74 and 4.65, respectively (Supplementary Figure S2). Several factors may account for the observed dose-dependent characteristics of tofacitinib. First, *V*ss values did not differ significantly among the four intravenous doses, suggesting that the tofacitinib distribution process did not affect its dose-dependency. Thus, the contribution of *V*ss to the dose-dependency of tofacitinib was negligible. Second, the contribution of CL<sup>R</sup> to CL was not significant. *Ae*0–24 h was less than 11.0% for all intravenous doses and less than 17.2% for all oral doses, with no significant differences among doses, suggesting that the contribution of renal excretion to the dose-dependent characteristics of tofacitinib is also low.

The renal extraction ratios (CLR/renal plasma flow rate) for urinary excretion of unchanged tofacitinib were estimated in rats based on its CLR, a reported renal blood flow rate of 36.8 mL/min/kg [20], and a hematocrit of approximately 45% [21]. The estimated renal extraction ratios following intravenous administration of 5, 10, 20, and 50 mg/kg tofacitinib were 13.3%, 13.8%, 13.8%, and 4.07%, respectively. These findings indicate that, in rats, tofacitinib has a low renal extraction ratio and that little tofacitinib is excreted via the kidneys. Therefore, most of the administered tofacitinib was eliminated via nonrenal pathways (CLNR).

The CLNR of tofacitinib was affected by gastrointestinal (including biliary) excretion of unchanged drug and metabolic clearance. The contribution of gastrointestinal excretion to CLNR was negligible, with no tofacitinib detected in the gastrointestinal tract 24 h after intravenous administration. The observed CLNR of tofacitinib may therefore represent metabolic clearance of the drug, suggesting that changes in its CLNR in rats may be due to changes in its metabolism. The increases in the dose-normalized AUC after intravenous and oral administration of tofacitinib may have been due to saturation of its metabolism, in agreement with the inverse relationship between slower CLNR and higher intravenous dose. Oral tofacitinib also showed a dose-dependent AUC in humans, as evidenced by dose-dependent increases in dose-normalized AUCs [8,22].

Although the *F* values of tofacitinib differed in humans (74%) [7] and rats (29.1–33.8%), based on calculations using the same intravenous and oral doses, *F* values were not 100% in either species. Because it is difficult to measure first-pass metabolism in humans, first-pass metabolism in the liver and gastrointestinal tract was measured in rats. The *F* and GI24 h of 10 mg/kg tofacitinib administered orally were 29.1% and 3.16%, respectively. The level of unchanged drug in the gastrointestinal tract (3.16%) may be due in part to the gastrointestinal (including biliary) excretion of absorbed drug. For comparison, the mean "true" fraction of unabsorbed dose (*F*unabs) following oral administration could be estimated using the equation [23];

$$\text{GII}\_{24\text{ h,oral}} = \text{F}\_{\text{unabs}} + (\text{F} \times \text{GI}\_{24\text{ h,intravenous}}) \tag{3}$$

where GI24 h, oral and GI24 h, intravenous are the percentages of oral and intravenous doses, respectively, remaining in the gastrointestinal tract after 24 h. Because GI24 h, intravenous in the present study was negligible, *F*unabs was almost equal to GI24 h, oral, indicating that gastrointestinal (including biliary) excretion of absorbed tofacitinib contributed little to the total drug recovered from the gastrointestinal tract after oral administration. Thus, approximately 96.8% of orally administered tofacitinib (10 mg/kg) was absorbed from the gastrointestinal tract in rats. Because only 3.16% of oral tofacitinib was not absorbed from the gastrointestinal tract at 24 h and the *F* value was 29.1%, approximately 67.7%·[100% − (3.16% + 29.1%)] of orally administered tofacitinib may have been eliminated by first-pass metabolism.

After intravenous administration of tofacitinib, the CL values of 14.5–36.3 mL/min/kg based on plasma data were considerably lower than the reported cardiac output of 296 mL/min/kg based on blood data [20] and a hematocrit of approximately 45% [21] in rats. These findings suggested that the first-pass effects of tofacitinib in the lungs and heart were negligible.

The AUCs were similar after intragastric and intraduodenal instillation of 10 mg/kg tofacitinib, suggesting that gastric first-pass effects on tofacitinib were negligible. However, the AUC after intraduodenal instillation of 10 mg/kg tofacitinib was 50.7% of that after intraportal administration, suggesting that approximately 49.3% of orally administered drug was not absorbed into the portal vein and that approximately 46.1%·[100% − (50.7% + 3.16%)] of orally administered tofacitinib was metabolized in the intestine before entering the portal vein. The AUC after intraportal administration of 10 mg/kg tofacitinib was 58.0% of that after intravenous administration, suggesting that the hepatic first-pass metabolism of tofacitinib after absorption into the portal vein was approximately 42.0%. Moreover, approximately 21.3% of the oral dose (42% of 50.7% of orally administered tofacitinib) was metabolized in the rat liver and 29.4% (50.7–21.3%) of the oral dose was absorbed into the systemic circulation. The latter percentage (29.4%) was close to the *F* value of 29.1%. Even though there is a species difference of bioavailability between human and rats, we could presume that 26%·(100–74%) of oral tofacitinib in humans was first-pass metabolized in the intestine and the liver with a similar ratio as rats. If a drug was not first-pass metabolized in the liver of the rat model, no hepatic first-pass metabolism was expected in humans. Considerable hepatic and intestinal first-pass metabolism has also been reported for other drugs, including ipriflavone [24], oltipraz [25], and sildenafil [26] in rats, and midazolam [27] in humans.

The AUCs were similar after intragastric and intraduodenal instillation of 10 mg/kg tofacitinib, suggesting that gastric first-pass effects on tofacitinib were negligible. However, the AUC after intraduodenal instillation of 10 mg/kg tofacitinib was 50.7% of that after intraportal administration, suggesting that approximately 49.3% of orally administered drug was not absorbed into the portal vein and that approximately 46.1%·[100% − (50.7% + 3.16%)] of orally administered tofacitinib was metabolized in the intestine before entering the portal vein. The AUC after intraportal administration of 10 mg/kg tofacitinib was 58.0% of that after intravenous administration, suggesting that the hepatic first-pass metabolism of tofacitinib after absorption into the portal vein was approximately 42.0%. Moreover, approximately 21.3% of the oral dose (42% of 50.7% of orally administered tofacitinib) was metabolized in the rat liver and 29.4% (50.7–21.3%) of the oral dose was absorbed into the systemic

circulation. The latter percentage (29.4%) was close to the *F* value of 29.1%. Even though there is a species difference of bioavailability between human and rats, we could presume that 26%·(100–74%) of oral tofacitinib in humans was first-pass metabolized in the intestine and the liver with a similar ratio as rats. If a drug was not first-pass metabolized in the liver of the rat model, no hepatic first-pass metabolism was expected in humans. Considerable hepatic and intestinal first-pass metabolism has also been reported for other drugs, including ipriflavone [24], oltipraz [25], and sildenafil [26] in rats, and midazolam [27] in humans.

In humans, hepatic microsomal CYP3A4 and to a lesser extent CYP2C19 are involved in the metabolism of tofacitinib, oxidizing the pyrrolopyrimidine moiety and producing a carbonyl moiety, the major metabolite of tofacitinib [3]. CYP3A1(23)/2 and CYP2C11 are the main enzymes involved in drug metabolism in rats and are highly expressed in the rat liver and small intestine [28,29]. Human liver and intestinal CYP2C19 and rat CYP2C11 are highly homologous and human liver and gastrointestinal CYP3A4 and rat CYP3A1(23) share 73% homology [28,30]. We recently observed [31] that CYP3A1(23)/2 and CYP2C11 are the main CYPs responsible for the metabolism of tofacitinib in rats, as evidenced by a 46% greater AUC in rats pretreated with ketoconazole, an inhibitor of CYP3A1/2 [32], and a 39% greater AUC in rats pretreated with fluconazole, an inhibitor of CYP2C11 [33]. In contrast, the AUC of tofacitinib reduced by 56% in rats pretreated with dexamethasone, an inducer of CYP3A1/2 [34], and 26% in rats pretreated with rifampin, an inducer of CYP2C11 [35]. The AUCs of tofacitinib in rats pretreated with specific inhibitors or inducers of different CYP isoforms did not differ significantly [31]. Therefore, the dose dependent increases in AUCs of tofacitinib after intravenous and oral administration to rats suggested that hepatic first-pass metabolism of tofacitinib (42%) was saturated after intravenous administration, whereas its intestinal (46.1%) and/or hepatic (23.1%) first-pass metabolism was saturated after oral administration. This saturation may have been due to the saturable metabolism of tofacitinib by CYP3A1/2 and/or CYP2C11 in rat liver and intestine.

The distribution process of tofacitinib did not contribute to its dose-dependent profiles, as *V*ss values did not differ significantly among the four intravenous doses. However, tofacitinib was widely distributed in rat tissues, especially in the small and large intestines, with the T/P ratios being higher for the intestines than for other tissues at both 30 min and 2 h. These findings suggest a mechanism for the effectiveness of tofacitinib in the treatment of ulcerative colitis, resulting in its approval in 2018 as the first oral drug for the treatment of chronic ulcerative colitis [6]. Tofacitinib is undergoing evaluation in clinical trials for the treatment of various diseases, including psoriasis [36,37], alopecia [38], atopic dermatitis [39], and ankylosing spondylitis [40].

Recently, several studies on tofacitinib pharmacokinetics were reported in patients with various diseases, including hepatic injury [41], renal failure [42], psoriasis [43], as well as inflammatory bowel disease [44], and most of them focused on the relationship between the drug concentration and the therapeutic efficacy. It was not well explained that the changes of plasma concentration according to diseases was related to the pharmacokinetic basis. In addition, pharmacokinetic drug interaction of tofacitinib is also expected since tofacitinib is mainly metabolized by CYP3A and is a substrate of P-glycoprotein [45]. Some pharmacokinetic drug interactions with tofacitinib were reported [46–48]. However, it is difficult to get the information from the clinical settings in order to evaluate the pharmacokinetic mechanism of the drug–disease or drug-drug interaction. Therefore, we need to further investigate the pharmacokinetic mechanism of the drug-disease or drug-drug interaction of tofacitinib in the rat model based on our pharmacokinetic characteristics of the drug in rats.

#### **5. Conclusions**

In conclusion, the low *F* of 10 mg/kg tofacitinib (29.1%) after oral administration to rats was mainly due to significant intestinal (46.1%) and hepatic (23.1%) first-pass metabolism. Our observation that the dose-normalized AUCs of tofacitinib in rats increased with increasing intravenous and oral doses, suggests that the hepatic and intestinal first-pass metabolism of tofacitinib was saturated by increasing its intravenous and oral doses.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1999-4923/11/7/318/s1, **Figure S1:** Mean dose (mg/kg) versus dose-normalized AUC (µg·min/mL) of tofacitinib in Sprague-Dawley rats after (**A**) 1-min intravenous infusion of 5 (*n* = 9), 10 (*n* = 8), 20 (*n* = 7), and 50 (*n* = 7) mg/kg tofacitinib and (**B**) oral administration of 10 (*n* = 7), 20 (*n* = 8), 50 (*n* = 9), and 100 (*n* = 7) mg/kg tofacitinib. Bars represent standard deviations (SD). (**A**) 20 mg/kg was significantly different (*p* < 0.05) from5 mg/kg. 50 mg/kg was significantly different (*p* < 0.001) from 5, 10 and 20 mg/kg. (**B**) 10 mg/kg was significantly different (*p* < 0.05) from 50mg/kg. 100 mg/kg was significantly different from 10 (*p* < 0.001), 20 (*p* < 0.001) and 50 (*p* < 0.01) mg/kg, respectively, **Figure S2:** Plots of dose versus AUC of tofacitinib in Sprague-Dawley rats after (**A**) 1-min intravenous infusion of 5, 10, 20, and 50 mg/kg tofacitinib and (**B**) oral administration of 10, 20, 50, and 100 mg/kg tofacitinib. Dose and AUC ratios were calculated based on 5 and 10 mg/kg dose and respective AUC for intravenous and oral administration, respectively.

**Author Contributions:** J.S.L. performed all of the animal experiments and the HPLC analysis of tofacitinib in the biological samples and estimated the pharmacokinetic parameters. S.H.K. designed the experiments, performed the statistical analysis and graphic works, and drafted the manuscript. All authors have read and approved the final manuscript.

**Funding:** This work was supported by the Korea Health Technology R&D Project (HI16C0992) through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health and Welfare, Korea.

**Conflicts of Interest:** The authors declare no competing financial interests.
