Pharmacokinetic Interaction between Sorafenib and Atorvastatin, and Sorafenib and Metformin in Rats

The tyrosine kinase inhibitor sorafenib is the first-line treatment for patients with hepatocellular carcinoma (HCC), in which hyperlipidemia and type 2 diabetes mellitus (T2DM) may often coexist. Protein transporters like organic cation (OCT) and multidrug and toxin extrusion (MATE) are involved in the response to sorafenib, as well as in that to the anti-diabetic drug metformin or atorvastatin, used in hyperlipidemia. Changes in the activity of these transporters may lead to pharmacokinetic interactions, which are of clinical significance. The study aimed to assess the sorafenib−metformin and sorafenib−atorvastatin interactions in rats. The rats were divided into five groups (eight animals in each) that received sorafenib and atorvastatin (ISOR+AT), sorafenib and metformin (IISOR+MET), sorafenib (IIISOR), atorvastatin (IVAT), and metformin (VMET). Atorvastatin significantly increased the maximum plasma concentration (Cmax) and the area under the plasma concentration–time curve (AUC) of sorafenib by 134.4% (p < 0.0001) and 66.6% (p < 0.0001), respectively. Sorafenib, in turn, caused a significant increase in the AUC of atorvastatin by 94.0% (p = 0.0038) and its metabolites 2−hydroxy atorvastatin (p = 0.0239) and 4−hydroxy atorvastatin (p = 0.0002) by 55.3% and 209.4%, respectively. Metformin significantly decreased the AUC of sorafenib (p = 0.0065). The AUC ratio (IISOR+MET group/IIISOR group) for sorafenib was equal to 0.6. Sorafenib did not statistically significantly influence the exposure to metformin. The pharmacokinetic interactions observed in this study may be of clinical relevance in HCC patients with coexistent hyperlipidemia or T2DM.


Introduction
Hepatocellular carcinoma (HCC) is one of the most common primary malignant liver tumors whose morbidity is on the rise [1]. Both abnormal lipid metabolism and type 2 diabetes mellitus in increased or decreased absorption of the drug, the aim of the study was to assess the influence of metformin and atorvastatin on the pharmacokinetics of sorafenib (and its active metabolite) and vice versa.

Materials and Methods
All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. Animals were given a standard diet and water ad libitum, and the experimental protocol for this study was approved by the Local Ethics Committee (Number 02/2019, from 1 March 2019), Poznań University of Life Sciences, Department of Animal Physiology and Biochemistry, Wołyńska 35 Str., 60-637 Poznań, Poland.

Reagents
All the reagents used for HPLC-UV and the UPLC-MS/MS assays are listed in Table 1. Table 1. List of all reagents for HPLC-UV and UPLC-MS/MS assays.
Separation was achieved by the gradient elution of the mobile phase, from ammonium acetate (0.1 M, pH = 3.4 (adjusted with glacial acetic acid)) as eluent A to an acetonitrile eluent B, at a flow rate of 1.0 mL/min through a reversed-phase C8 column (Symmetry ® C8, 250 × 4.6 mm, 5.0 µm particle size) (Waters Corporation ® , Milford, Massachusetts, USA). The linear gradient ran from 60% eluent A and 40% eluent B to 29% eluent A and 71% eluent B. The column temperature was maintained at 25 • C, the UV detection wavelength was set at 265 nm, and the injection volume was 20 µL; lapatinib was used as the internal standard (IS).
The metformin concentrations in rats' plasma were determined according to the method described by Gabr et al. [30]. The conditions were as follows: the UV detection wavelength was 236 nm; a Symmetry ® C8, 250 × 4.6 mm, 5.0 µm-particle-size column from Waters was used; the column temperature was maintained at 25 • C; the mobile phase consisted of ammonium formate/acetonitrile (95:5, v/v); the pH of the mobile phase was 6.0; the flow rate was 1.0 mL/min; the volume of each injection was 20 µL; and the retention times for metformin and the internal standard (acetaminophen) were 3.24 and 9.27 min, respectively.

UPLC-MS/MS Assay
The concentrations of atorvastatin, 2-hydroxy atorvastatin (2-OH AT), and 4-hydroxy atorvastatin (4-OH AT) were assayed using a UPLC-MS/MS method. The concentration of each of the working solutions was 10 µg/mL. The sample preparation involved liquid-liquid extraction. The dry residue was dissolved in 50 µL of the mobile phase, and 20 µL was injected into the UPLC Nexera system coupled to an LCMS−8030 Triple Quadrupole tandem mass spectrometer (Shimadzu Corp., Kioto, Japan). The analytes were separated in a Zorbax Plus C18 column (100 × 2.1 mm; 3.5 µm) (Agilent Technologies, Santa Clara, California, USA) at a column temperature of 40 • C. The mobile phase was a mixture of de-ionized water (A) and acetonitrile (B), both containing 0.1% (v/v) formic acid. The gradient was as follows: 0-2 min, linear from 50 to 70% B; 2-4 min, 70% B; 4-6 min, a return from 70 to 50% B; and a post-time of 4 min with 50% B for column equilibration. The mobile phase flow was set to 0.3 mL/min. The eluent from the HPLC column was introduced directly to the MS interface, using electrospray ionization in the positive ion mode. The MS parameters were as follows: interface temperature, 350 • C; DL temperature, 250 • C; heat block temperature, 400 • C; nebulizing gas flow, 2 L/min; and drying gas flow, 15 L/min. The electrospray needle voltage was 3.5 kV. The specific transitions for the analytes were monitored using the multiple reaction monitoring (MRM) mode. The most sensitive mass transition was monitored from m/z 558.8 to 440.1 (for atorvastatin), 547.9 to 440.2 (for 2-OH AT and 4-OH AT), and 481.7 to 258.0 (for rosuvastatin).
The UPLC−MS/MS method for atorvastatin, 2-OH AT, and 4-OH AT showed linearity in the concentration range of 0.4-200 ng/mL in plasma. The intra-and inter-assay precision values, expressed as relative standard deviations, were <18.7% for atorvastatin and 4-OH AT, and <14.0% for 2-OH AT. The intra-and inter-day accuracy of the method, expressed as the relative error, was <14.0, <6.1, and <10.2% for atorvastatin, 4-OH AT, and 2-OH AT, respectively.

Pharmacokinetic Evaluation
The Phoenix ® WinNonlin version 8.0 software (Certara, Princeton, New Jersey, USA) was used for the calculation of the following pharmacokinetic parameters:

•
The elimination rate constant (k e ); • The absorption rate constant (k a ); • The elimination half-life (t 1/2 ); • The maximum plasma concentration (C max ); • The time to reach the C max (t max ); • The total area under the concentration-time curve (AUC 0-t and AUC 0-∞ ); • The apparent plasma drug clearance (Cl/F); • The apparent volume of distribution (V d /F).
The above-mentioned parameters underwent statistical analysis.

Statistical Analysis
The statistical analyses were performed using the Statistica software (Statistica, Tulsa, OK, USA) version 13.3 (for atorvastatin, 2-OH AT, 4-OH AT, and metformin) and SAS software (SAS Institute Inc., Cary, NC 27513, USA) version 9.4 (for sorafenib and SR_NO). Normality was estimated with the Shapiro-Wilk test. For sorafenib and SR_NO, analysis of variance (PROC GLM) was used to test the significance of the differences between the groups for variables without significant deviations from normality. Welch's test was used when the assumption of homogeneity of variance was not met, and ANOVA was followed by post-hoc Tukey tests. For metformin, atorvastatin, and its metabolites, the differences between the normally distributed variables were determined with the Student's t-test. For all the remaining variables, the Mann-Whitney U test was applied, and a p-value < 0.05 was considered significant.

Results
All the data are expressed as the mean values ± standard deviations (SD).

The Influence of Atorvastatin on the Pharmacokinetics of Sorafenib and SR_NO
After the administration of sorafenib with atorvastatin (I SOR+AT group), the C max , AUC 0-t , and AUC 0-∞ of sorafenib increased by 134.6, 66.6, and 58.7%, respectively, compared with those in the III SOR group (Table 2, Figure 1). The C max of sorafenib occurred later (t max increased by 116.9% and k a decreased by 73.0%) in the I SOR+AT group. The value of the t 1/2 of sorafenib decreased in the presence of atorvastatin (the t 1/2 was lower by 0.3-fold). However, there were no statistically significant differences for the k el and Cl/F of sorafenib. V d /F was also comparable between the III SOR and I SOR+AT groups.

The Influence of Sorafenib on the Pharmacokinetics of Atorvastatin, 2-OH AT, and 4-OH AT
When atorvastatin and sorafenib were co-administered, the AUC0-t and AUC0--∞ of atorvastatin increased by 94.0 and 93.2%, respectively (Table 3, Figure 3). In the group of rats receiving both drugs, atorvastatin tmax was longer than that in the IVAT group. No statistically significant differences were revealed for Cmax, ka, kel, and t1/2.
The Cmax of 2-OH AT was similar in both the atorvastatin and atorvastatin+sorafenib groups (Table 3, Figure 4). The exposure to 2-OH AT was significantly higher in the presence of sorafenib, which was reflected by increased values of AUC0-t and AUC0-∞, but for AUC0-∞, there was no statistical significance. The ratios of AUC0-t and AUC0-∞ (ISOR+AT group/IVAT group) for atorvastatin were increased by 0.5-and 0.4-fold, respectively. In the ISOR+AT group, the tmax of 2-OH AT was longer Atorvastatin increased the SR_NO C max , AUC 0-t , and AUC 0-∞ by 245.5, 242.4, and 89.5%, respectively (Table 2, Figure 2); the t 0.5 of SR_NO in the I SOR+AT group decreased by 0.4-fold. However, no statistically significant differences were revealed for k el and t max .

The Influence of Sorafenib on the Pharmacokinetics of Atorvastatin, 2-OH AT, and 4-OH AT
When atorvastatin and sorafenib were co-administered, the AUC0-t and AUC0--∞ of atorvastatin increased by 94.0 and 93.2%, respectively (Table 3, Figure 3). In the group of rats receiving both drugs, atorvastatin tmax was longer than that in the IVAT group. No statistically significant differences were revealed for Cmax, ka, kel, and t1/2.
The Cmax of 2-OH AT was similar in both the atorvastatin and atorvastatin+sorafenib groups (Table 3, Figure 4). The exposure to 2-OH AT was significantly higher in the presence of sorafenib, which was reflected by increased values of AUC0-t and AUC0-∞, but for AUC0-∞, there was no statistical significance. The ratios of AUC0-t and AUC0-∞ (ISOR+AT group/IVAT group) for atorvastatin were increased by 0.5-and 0.4-fold, respectively. In the ISOR+AT group, the tmax of 2-OH AT was longer

The Influence of Sorafenib on the Pharmacokinetics of Atorvastatin, 2-OH AT, and 4-OH AT
When atorvastatin and sorafenib were co-administered, the AUC 0-t and AUC 0-∞ of atorvastatin increased by 94.0 and 93.2%, respectively (Table 3, Figure 3). In the group of rats receiving both drugs, atorvastatin t max was longer than that in the IV AT group. No statistically significant differences were revealed for C max , k a , k el , and t 1/2 . curve from zero to the time of last measurable concentration; AUC0-∞, area under the plasma concentration-time curve from zero to infinity; tmax, time to the first occurrence of Cmax; ka, absorption rate constant; kel, elimination rate constant; t1/2, half-life in the elimination phase; Cl/F, apparent plasma drug clearance; Vd/F, apparent volume of distribution; b.w., body weight. Arithmetic means ± standard deviations (SD) are shown with coefficients of variation (CV) (%) in brackets; 1 t-test; 2 Mann-Whitney U test.  The C max of 2-OH AT was similar in both the atorvastatin and atorvastatin+sorafenib groups (Table 3, Figure 4). The exposure to 2-OH AT was significantly higher in the presence of sorafenib, which was reflected by increased values of AUC 0-t and AUC 0-∞ , but for AUC 0-∞ , there was no statistical significance. The ratios of AUC 0-t and AUC 0-∞ (I SOR+AT group/IV AT group) for atorvastatin were increased by 0.5-and 0.4-fold, respectively. In the I SOR+AT group, the t max of 2-OH AT was longer than that in the IV AT group, but there was no statistical significance. No statistically significant differences were revealed for k el and t 1/2 . Sorafenib elevated the 4-OH AT C max by 127.0% (Table 3, Figure 5). When atorvastatin and sorafenib were co-administered, the AUC 0-t and AUC 0-∞ of 4-OH AT increased by 320.9 and 220.5%, respectively. There were no significant differences among the groups for the following pharmacokinetic parameters of 4-OH AT: k el , t 1/2 , and t max .

The Influence of Metformin on the Pharmacokinetics of Sorafenib and SR_NO
In the presence of metformin (IISOR+MET group,) the sorafenib plasma AUC0-t and AUC0-∞, were 44.0 and 45.1% lower than those in the IIISOR group (Table 2, Figure 1). Metformin decreased the ka by 0.5-fold. However, the Cmax and tmax of sorafenib were comparable. The clearance of sorafenib in the IISOR+MET group was higher by 76.3% than that in the IIISOR group, but there were no significant differences for kel and t0.5. Likewise, the Vd/F of sorafenib did not differ significantly between the groups.
All the pharmacokinetic parameters of SR_NO in the IISOR+MET and IIISOR groups were comparable (Table 2, Figure 2).

The Influence of Sorafenib on the Pharmacokinetics of Metformin
Comparing both groups of animals (VMET vs. IISOR+MET), statistically significant differences were observed only for tmax, which was 71.39% higher in the IISOR+MET group (Table 4, Figure 6). The values of the other pharmacokinetic parameters were comparable. High values of CV for ka and kel in the VMET and IISOR+MET groups, respectively, were observed. It may indicate significant inter-individual variability in animals in terms of metformin absorption and elimination.

The Influence of Metformin on the Pharmacokinetics of Sorafenib and SR_NO
In the presence of metformin (II SOR+MET group,) the sorafenib plasma AUC 0-t and AUC 0-∞ , were 44.0 and 45.1% lower than those in the III SOR group (Table 2, Figure 1). Metformin decreased the k a by 0.5-fold. However, the C max and t max of sorafenib were comparable. The clearance of sorafenib in the II SOR+MET group was higher by 76.3% than that in the III SOR group, but there were no significant differences for k el and t 0.5 . Likewise, the V d /F of sorafenib did not differ significantly between the groups.
All the pharmacokinetic parameters of SR_NO in the II SOR+MET and III SOR groups were comparable (Table 2, Figure 2).

The Influence of Sorafenib on the Pharmacokinetics of Metformin
Comparing both groups of animals (V MET vs. II SOR+MET ), statistically significant differences were observed only for t max , which was 71.39% higher in the IISOR+MET group (Table 4, Figure 6). The values of the other pharmacokinetic parameters were comparable. High values of CV for k a and k el in the V MET and II SOR+MET groups, respectively, were observed. It may indicate significant inter-individual variability in animals in terms of metformin absorption and elimination.  Cmax, maximum observed plasma concentration; AUC0-t, area under the plasma concentration-time curve from zero to the time of last measurable concentration; AUC0-∞, area under the plasma concentration-time curve from zero to infinity; tmax, time to the first occurrence of Cmax; ka, absorption rate constant; kel, elimination rate constant; t1/2, half-life in the elimination phase; Cl/F, apparent plasma drug clearance; Vd/F, apparent volume of distribution; b.w., body weight. Arithmetic means ± standard deviations (SD) are shown with coefficients of variation (CV) (%) in brackets; 1 t-test; 2 Mann-Whitney test.

Discussion
To properly evaluated our findings, it is important to note the limitations of the conducted studies. First of all, no animal model used in the experiment experienced HCC or T2DM, and as we know from our previous in vivo study with streptozotocin-induced diabetic Wistar rats receiving only sorafenib, T2DM has its impact upon exposure to sorafenib [31]. Additionally, the pharmacokinetic profile of sorafenib, metformin, and atorvastatin may differ between Wistar rats and humans, and the concentrations of sorafenib glucuronide were not measured.
On the other hand, since we used non-diabetic rats, this condition did not enhance the exposure to sorafenib and SR_NO [31], and therefore, we were able to examine potential pharmacokinetic DDIs in this animal model.
The doses of metformin (100 mg/kg) and atorvastatin (20 mg/kg) used in our study were selected based on available literature [27,28,32] and did not cause any adverse effects in healthy rats. A dose of atorvastatin of 20 mg/kg may cause liver injury in diabetic rats [27] but is safe for healthy rats [32].
Drug-drug interactions are clinically relevant in humans when changes in the AUC, Cmax, or clearance are greater than 25%, or there is a statistically significant influence on pharmacodynamics

Discussion
To properly evaluated our findings, it is important to note the limitations of the conducted studies. First of all, no animal model used in the experiment experienced HCC or T2DM, and as we know from our previous in vivo study with streptozotocin-induced diabetic Wistar rats receiving only sorafenib, T2DM has its impact upon exposure to sorafenib [31]. Additionally, the pharmacokinetic profile of sorafenib, metformin, and atorvastatin may differ between Wistar rats and humans, and the concentrations of sorafenib glucuronide were not measured.
On the other hand, since we used non-diabetic rats, this condition did not enhance the exposure to sorafenib and SR_NO [31], and therefore, we were able to examine potential pharmacokinetic DDIs in this animal model.
The doses of metformin (100 mg/kg) and atorvastatin (20 mg/kg) used in our study were selected based on available literature [27,28,32] and did not cause any adverse effects in healthy rats. A dose of atorvastatin of 20 mg/kg may cause liver injury in diabetic rats [27] but is safe for healthy rats [32].
Drug-drug interactions are clinically relevant in humans when changes in the AUC, C max , or clearance are greater than 25%, or there is a statistically significant influence on pharmacodynamics endpoints. For example, for metformin, glycosylated hemoglobin A1c (HbA1c) levels or oral glucose tolerance tests (OGTTs) may be relevant [16,33].

The Influence of Atorvastatin on the Pharmacokinetics of Sorafenib and SR_NO
We found that the co-administration of sorafenib and atorvastatin could lead to a higher risk of adverse reactions to sorafenib, such as hand-foot syndrome, alopecia, gastrointestinal disorders, a hypertensive crisis, or cardiotoxicity [34]. The almost doubling of exposure to sorafenib (p < 0.0001 for AUC 0-t , and p = 0.0002 for AUC 0-∞ ), along with the enhanced C max (+134.6%, p < 0.0001) and k el (+42.9%, p = 0.0637), most closely corresponds to Kalliokoski A. et al.'s results [35]. They concluded, based on a randomized crossover study with healthy volunteers (n = 24), that increased mean values of C max and AUC for repaglinide (among those who received repaglinide and atorvastatin, 0.25 + 40 mg) were most likely the effect of OATP1B1-mediated hepatic uptake inhibition by atorvastatin [34]. Since sorafenib is highly protein-bound [33], there is the potential for it to be released from protein binding by atorvastatin, which would improve the unbound fraction concentration of the chemotherapeutic agent as well. Additionally, an in vitro study on a murine monocytic leukemia cell line overexpressing P-gp showed that the co-use of atorvastatin may result in the increased absorption of P-gp substrates (digoxin and verapamil) [36].
Additional sorafenib N-oxide exposure ( Table 2) is probably a consequence of increased exposure to sorafenib. Considering the fact that both sorafenib and SR_NO exhibit inhibitory effects on CYP3A4 [37], and since atorvastatin itself is its substrate, the induction of CYP3A4 is highly unlikely. Since SR_NO exhibits pharmacological activity, the risk of toxicity is further enhanced.

The Influence of Sorafenib on the Pharmacokinetics of Atorvastatin, 2-OH AT, and 4-OH AT
Sorafenib+atorvastatin treatment may also lead to excessive side effects from atorvastatin, such as hyperglycemia, hepatitis, or musculoskeletal and connective tissue disorders [37], due to an almost doubling of the exposure to atorvastatin (Table 4). Since bioavailability (F) depends on the AUC 0-∞ for extravascular administration and is directly proportional, a significant (p < 0.005) decrease in CL/F and V d /F is consistent with the reported data. Even though t max was prolonged by 2.2-fold (p = 0.0419), there were no significant differences for the absorption (C max , k a ) or elimination (k el or t 1/2 ) of atorvastatin. The lack of reduction of exposure to 2-OH AT and 4-OH AT among the I SOR+AT group (Figures 4 and 5) suggests that the CYP3A4-mediated metabolism of atorvastatin was not suppressed by sorafenib. In the light of those findings, interaction occurs most likely during atorvastatin distribution at the protein transporter level. The overlap between sorafenib and atorvastatin transporter-related activity suggests that the induction of efflux transporters (P-glycoprotein) and/or inhibition of atorvastatin hepatic uptake transporters (Oatp1b1/3) by sorafenib could play a significant role in this DDI [5,38].
Furthermore, as a consequence of the highly enhanced exposure to atorvastatin, the concentration of its metabolites (2-OH AT and 4-OH AT) should increase as well. However, decreases in the 2-OH AT/atorvastatin ratios for the AUC 0-t (−19.4%) and AUC 0-∞ (−26.0%, p = 0.0312) ( Table 3) may suggest a smaller increase upon exposure to 2-OH AT than expected. However, the differences in the AUC 0−∞ for 2-OH AT and the AUC 0-t for 2-OH AT/atorvastatin ratio were statistically insignificant, and the C max for 2-OH AT remains comparable. At the same time, the I SOR+AT group was heavily exposed to 4-OH AT (p = 0.0002 for AUC 0-t , p = 0.0001 for AUC 0-∞ , and p = 0.0019 for C max ), which resulted in 2.5-, 1.6-, and 1.7-fold increases in the C max (p = 0.0084), AUC 0-t (p = 0.0239), and AUC 0-∞ (p = 0.0239) for the 4-OH AT/atorvastatin ratio, respectively, as well. The reported data suggest that the presence of sorafenib altered the proportions of 2-OH AT and 4-OH AT, in favor of 4-OH AT.

The Influence of Metformin on the Pharmacokinetics of Sorafenib and SR_NO
The co-administration of metformin with sorafenib significantly reduced and nearly halved rats' exposure to the chemotherapeutic agent (Figure 1), and the increased CL/F value (p < 0.005) is consistent with the reported data (Table 2). Furthermore, when a 0.5-fold decrease in the absorption rate constant is compared to the inconsistent changes in elimination-related parameters, interaction at the absorption level seems to occur. Since the C max and t max values were comparable among the study and control groups, the possibility of interference within the protein-bound sorafenib fraction may be ruled out. Additionally, accelerated intestinal passage is one of the most common adverse reactions observed in patients during metformin treatment [39], and a clinically important interaction between a high-fat diet and sorafenib (resulting in a 30% decrease in sorafenib absorption) is known as well [34]. Although the defecation rate noted during the experiment was not unusual, it might interrupt the intestinal absorption and enterohepatic cycle, and eventually decrease sorafenib exposure.
Furthermore, despite the decreased exposure to sorafenib, all the pharmacokinetic parameters of SR_NO were still comparable in both the study and control groups ( Table 2), which resulted in a 2.3-fold increased value of the AUC 0-t for SR_NO/sorafenib ratios in the II SOR+MET group vs. the III SOR group (p = 0.0002). Since SR_G plays a crucial role in the enterohepatic circulation of sorafenib [5], it is more likely that the increased SR_NO/sorafenib exposure ratios are a result of rapid intestinal transit, which shortens the time for SR_G transformation to sorafenib and sorafenib re-absorption, than a result of CYP3A4-mediated DDIs. However, sorafenib itself has shown CYP3A4-inhibitory properties in an in vitro study [37,40], and therefore, metformin could promote the CTP3A4-mediated pathway by alleviating the obstacles (by sorafenib concentration reduction). However, this metabolization hypothesis seems to be less established, especially considering that the direct impact of metformin on CYP3A4 activity remains largely unknown [41]. Although the pharmacological activity of sorafenib N-oxide and sorafenib are comparable [34], a halved exposure to sorafenib carries the risk of reduced, unsatisfactory responses to oncological treatment.
However, in our study, sorafenib did not significantly affect the total exposure to metformin in rats ( Figure 6), and the C max values of metformin in the V MET and the II SOR+MET groups were comparable (Table 4). After the combined use of these two drugs, only an increase in the t max of metformin (by 71.39%, p = 0.0069) was observed. Those data suggest that DDIs during the absorption process may occur. Metformin is known for its saturable absorption and prolongs t max due to its co-administration with food [14,39]. Since both PMAT and OCT1 transporters participate in the intestinal uptake of metformin, and sorafenib is still considered as an OCT1 substrate, these proteins might be a direct cause of metformin−sorafenib DDI [5,14].
Studies have reported that metformin partitions into erythrocytes, which are most likely its second compartment [14]. However, since the C max values remain comparable between the study and control groups, interference within this fraction is insignificant.

Conclusions
Statins are commonly used by elderly patients that usually also struggle with other conditions, including kidney and liver disorders (these organs are important for proper xenobiotic elimination). Since that population is at a higher risk of presenting with cancer, further pharmacokinetic studies of sorafenib−atorvastatin interaction in patients are needed. Moreover, plasma sorafenib and SR_NO monitoring during sorafenib−atorvastatin therapy might be beneficial for preventing sorafenib overdose.
Since 2-OH AT and 4-OH AT are together responsible for approximately 70% of the therapeutic activity of atorvastatin [32], and its profiles were strongly affected by the presence of sorafenib, it is also important to evaluate the influence of sorafenib−atorvastatin therapy on its pharmacokinetics in future clinical studies as well.
It can be assumed that in clinical practice, the combined use of sorafenib and metformin in HCC therapy will probably not result in an increase in the plasma levels of metformin and, thus, the risk of its side effects. It should also be noted that genetic variation in OCTs, MATEs, and PMAT transporter genes may bias the metformin plasma and intracellular levels and, hence, the patient's response to metformin [16]. At the same time, there is a growing need to clarify the potential impact of metformin on sorafenib's and sorafenib N-oxide's pharmacokinetics, both in patients and possibly with a suitable time interval between the administration of those two drugs. However, it should be remembered that altered drug pharmacokinetics are not always translatable into significant pharmacodynamic changes. Further clinical studies should investigate pharmacokinetic relationships between drugs, in parallel with the pharmacological effect of combined therapies, as they may occur in patients as well.