A phase I drug–drug interaction study to assess the effect of futibatinib on P‐gp and BCRP substrates and of P‐gp inhibition on the pharmacokinetics of futibatinib

Abstract

Futibatinib, an inhibitor of fibroblast growth factor receptor 1–4, is approved for the treatment of patients with advanced cholangiocarcinoma with FGFR2 fusions/rearrangements. In this phase I drug–drug interaction study, the effects of futibatinib on P‐glycoprotein (P‐gp) and breast cancer resistance protein (BCRP) substrates, and of P‐gp inhibition on futibatinib pharmacokinetics (PK) were investigated in healthy adults aged 18–55 years. In part 1, 20 participants received digoxin (P‐gp substrate) and rosuvastatin (BCRP substrate). Following a ≥10‐day washout, futibatinib was administered for 7 days, with digoxin and rosuvastatin coadministered on the third day. In part 2, 24 participants received futibatinib. Following a ≥3‐day washout, quinidine (P‐gp inhibitor) was administered for 4 days, with futibatinib coadministered on day 4. Blood samples were collected predose and for 24 (futibatinib), 72 (rosuvastatin), and 120 h (digoxin) postdose. Urine samples (digoxin) were collected predose and for 120 h postdose. PK parameters were compared between treatments using analysis of variance. Coadministration with futibatinib had no effect on the PK of digoxin and rosuvastatin, and coadministration with quinidine had minimal effects on the PK of futibatinib. Differences in C _max and AUC with and without futibatinib and quinidine, respectively, were <20%. The most common treatment‐emergent adverse events were diarrhea (80%) and increased blood phosphorous (75%) in part 1 and prolonged electrocardiogram QT interval (38%) in part 2. The data show that futibatinib has no clinically meaningful effects on the PK of P‐gp or BCRP substrates and that the effect of P‐gp inhibition on futibatinib PK is not clinically relevant.

Study Highlights.

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

Futibatinib is a highly selective, covalently binding, irreversible inhibitor of fibroblast growth factor receptor 1–4 that is approved for the treatment of patients with previously treated, unresectable, locally advanced, or metastatic intrahepatic cholangiocarcinoma harboring FGFR2 gene fusions or other rearrangements. Futibatinib is an inhibitor of P‐glycoprotein (P‐gp) and BCRP and a substrate of P‐gp in vitro. Previous clinical drug–drug interaction (DDI) studies showed that futibatinib exposure increased by 41%–51% when coadministered with a dual P‐gp and cytochrome P450 (CYP) 3A4 inhibitor (itraconazole), likely due to the combined effects of CYP3A and P‐gp inhibition.

• WHAT QUESTION DID THIS STUDY ADDRESS?

This study assessed the effect of futibatinib coadministration on the PK of P‐gp and BCRP substrates and the effect of P‐gp inhibition on futibatinib PK.

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

Coadministration with futibatinib had minimal effects on digoxin (P‐gp substrate) and rosuvastatin (BCRP substrate) exposure, and coadministration with quinidine (P‐gp inhibitor) resulted in an increase in futibatinib exposure that was not considered clinically relevant.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

Together with previous DDI studies, these results indicate that although concomitant use of dual perpetrators of CYP3A and P‐gp should be avoided, medications that inhibit and/or induce only P‐gp can be used concomitantly with futibatinib.

INTRODUCTION

Dysregulated fibroblast growth factor/fibroblast growth factor receptor (FGFR) signaling, driven by genomic alterations such as amplifications, mutations, and fusions/rearrangements, is implicated in the development and progression of various tumor types. ¹ , ² Consequently, FGFR has been evaluated as a therapeutic target across multiple cancer types, and reversible FGFR inhibitors have been approved for the treatment of FGFR‐aberrant cholangiocarcinoma (pemigatinib) and bladder cancer (erdafitinib). ¹ , ² , ³ , ⁴ More recently, futibatinib, a potent, covalently binding, irreversible inhibitor of FGFR1–4, was approved for patients with previously treated, advanced, intrahepatic cholangiocarcinoma harboring FGFR2 fusions or other rearrangements. ⁵

Unlike reversible adenosine triphosphate–competitive FGFR inhibitors, futibatinib forms a covalent adduct with a conserved cysteine residue in the P‐loop of the FGFR kinase domain. ⁶ Its irreversible binding nature and distinct binding site render futibatinib less susceptible to on‐target resistance mutations compared with infigratinib and pemigatinib. ⁷ , ⁸ Indeed, futibatinib has shown broad preclinical activity in FGFR‐deregulated cell lines and xenograft models, including against cell lines harboring distinct FGFR mutations resistant to reversible adenosine triphosphate–competitive FGFR inhibitors. ⁷ , ⁹ , ¹⁰ Clinically, futibatinib has demonstrated antitumor activity in patients with a wide spectrum of FGFR‐aberrant tumors in phase I trials. ¹⁰ , ¹¹ , ¹² In the pivotal phase II trial in patients with previously treated, advanced, intrahepatic cholangiocarcinoma harboring FGFR2 fusions or other rearrangements, futibatinib 20 mg once daily showed durable clinical activity (objective response rate: 42%; median duration of response: 9.7 months), ⁸ leading to its approval in the United States, Europe, and Japan.

Several studies have been conducted to investigate the pharmacologic properties of futibatinib. ¹⁰ , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ In a phase I, first‐in‐human study in participants with advanced solid tumors, futibatinib showed dose‐proportional exposure increases at doses between 4 and 24 mg once daily, with no accumulation after repeat doses. ¹⁰ Futibatinib was rapidly absorbed after administration: the time to reach maximum plasma concentration (T _max) was approximately 2 h, and the terminal elimination half‐life (t _½) was approximately 3 h. ¹⁰ A subsequent phase I food effect study in healthy participants found that futibatinib bioavailability was 11.2% lower and the median T _max was significantly delayed (4.0 vs. 1.5 h) under fed versus fasted conditions; however, the reduction in bioavailability was not expected to have a clinically meaningful impact on futibatinib safety or efficacy. ¹³ In addition, there were no significant differences in futibatinib exposure when administered with or without the acid‐reducing agent lansoprazole. ¹³

Although futibatinib is eliminated primarily through hepatic metabolism, no clinically relevant effects of hepatic impairment on futibatinib exposure were observed in participants with mild, moderate, or severe hepatic impairment compared with matched controls, indicating that futibatinib dose adjustments due to hepatic impairment are not necessary in patients receiving futibatinib 20 mg daily. ¹⁴ Moreover, previous studies have shown no clinically meaningful differences in futibatinib exposure in participants with mild to moderate renal impairment. ⁵ , ¹⁸

Futibatinib is metabolized primarily by cytochrome P450 (CYP) 3A4 and is a substrate of P‐glycoprotein (P‐gp) in vitro. ¹⁵ Therefore, concurrent administration of futibatinib with a P‐gp and CYP3A perpetrator may impact futibatinib pharmacokinetics (PK). When coadministered with rifampin, a dual P‐gp and CYP3A inducer, futibatinib exposure decreased by 53%–64%. Although coadministration with itraconazole, a dual P‐gp and CYP3A inhibitor increased futibatinib exposure by 41%–51%, the impact on futibatinib t_½ was limited ¹⁵ suggesting that itraconazole mainly affects drug–drug interactions (DDI) in the intestine. The limited impact on futibatinib t_½ suggests that the observed DDI more likely resulted from P‐gp inhibition than CYP3A activity, although the impact of CYP3A on intestinal first‐pass metabolism cannot be excluded. Futibatinib also inhibits CYP3A in vitro but does not affect the metabolism of concomitant medications metabolized by CYP3A (e.g., midazolam). ¹⁵

The magnitude of increases in area under the plasma concentration–time curve (AUC) and maximum plasma drug concentration (C _max) with coadministration of strong and moderate CYP3A inhibitors varies among FGFR inhibitors. Data from in vitro studies, for example, indicate that infigratinib is a substrate of P‐gp but not of CYP3A4, with coadministration of infigratinib with itraconazole shown to increase infigratinib plasma concentrations. ¹⁹ In an in vivo DDI study of pemigatinib, there was an 88% and 17% increase in pemigatinib AUC and C _max, respectively, with itraconazole, ²⁰ which appeared to be mediated predominantly by CYP3A inhibition. ²⁰ , ²¹ While the US prescribing information for pemigatinib advises against the concomitant use of strong or moderate CYP3A inhibitors, ⁴ futibatinib prescribing information states that coadministration with strong dual CYP3A4/P‐gp inhibitors should be avoided. ⁵

Additional in vitro studies have shown that futibatinib inhibits P‐gp and breast cancer resistance protein (BCRP) but does not inhibit other transporters (organic anion transporter [OAT]1, OAT3, organic cation transporter 2, organic anion transporting polypeptides [OATP]1B1, OATP1B3, multidrug and toxic compound extrusion [MATE]1, MATE2K) at clinically relevant concentrations. ⁵ , ¹⁸ Therefore, coadministration of futibatinib with P‐gp or BCRP substrates has the potential to increase their exposure.

Given the potential for such DDIs, we conducted a two‐part study to evaluate the effect of futibatinib on P‐gp and BCRP substrates as a perpetrator, and the effect of a P‐gp inhibitor on futibatinib as a victim.

METHODS

Study design and procedures

This was a two‐part, open‐label, fixed‐sequence, two‐period, phase I, DDI study that assessed the effect of futibatinib on the PK of P‐gp and BCRP substrates (part 1) and the effect of P‐gp inhibition on the PK of futibatinib (part 2; Figure 1).

FIGURE 1.

Design of the two‐part, two‐period, phase I drug–drug interaction study. q.d., once daily; q.i.d, four times daily.

In part 1, an oral dose of 0.25 mg digoxin and 10 mg rosuvastatin was coadministered on day 1 of period 1, followed by a washout period of ≥10 days. In period 2, an oral dose of 20 mg futibatinib was administered once daily for 7 consecutive days, and a single oral dose of digoxin and rosuvastatin was coadministered on the morning of day 3, 2 h after futibatinib dosing to ensure maximum inhibition (T _max: ~2 h ¹⁰ ). For determination of rosuvastatin and digoxin PK with and without futibatinib administration, blood (rosuvastatin and digoxin) and urine (digoxin) samples, were collected before dosing and for 72 and 120 h after dosing, respectively, in each period (Table S1).

In part 2, an oral dose of 20 mg futibatinib was administered on day 1 of period 1, followed by a ≥3‐day washout period. In period 2, an oral dose of 200 mg quinidine was administered four times daily (every 6 h ± 20 min) on days 1–4, and a single oral dose of futibatinib was coadministered on the morning of day 4, 2 h following quinidine dosing to ensure maximum inhibition. Blood samples for futibatinib plasma PK, with or without quinidine coadministration, were collected prior to and for 24 h following futibatinib dosing in each period (Table S1).

The study was conducted in accordance with the principles of the Declaration of Helsinki, US Code of Federal Regulations, and Good Clinical Practice guidelines of the International Council for Harmonisation. The protocol was reviewed and approved by the Advarra Institutional Review Board (Columbia, MD, USA), and all study participants provided written informed consent before enrollment.

Participants

Both parts of the study enrolled healthy adult male and female participants of nonchildbearing potential who were aged 18–55 years at screening, minimum body weight of 50 kg and a body mass index between 18 and 32 kg/m², inclusive. Exclusion criteria included the use of nicotine‐ and tobacco‐containing products for ≤3 months prior to the first dose of study treatment; a history or presence of a clinically significant medical or psychiatric condition currently requiring treatment, including seated blood pressure <90/40 mmHg or >155/95 mmHg, heart rate <40 or >99 bpm, and liver transaminases above the upper limit of normal; serum creatinine clearance <60 mL/min based on Cockcroft‐Gault formula; Fridericia‐corrected QT interval >450 milliseconds; a history of illness that, in the opinion of the investigator, might confound the study results or pose additional risk to the participant by their participation in the study; use of any drug, including prescription and nonprescription medications, herbal remedies, or vitamin supplements within 14 days prior to the first dose of study treatment; and use of drugs known to be significant inducers of CYP3A enzymes, BCRP, and/or P‐gp within 28 days prior to the first dose of study treatment.

Bioanalytical methods and PK parameters

Samples were analyzed for plasma and urine digoxin, plasma rosuvastatin, and plasma futibatinib using high‐performance liquid chromatography–tandem mass spectrometry methods validated with respect to the accuracy, precision, linearity, sensitivity, and specificity at Celerion, Lincoln, NE. Bioanalytic methods for futibatinib concentrations have been described previously. ¹³ The analytic range (lower limit of quantitation–upper limit of quantitation) for plasma futibatinib was 0.500–250 ng/mL.

Aliquots of human plasma and urine containing digoxin and internal standard (IS; d₃‐digoxin) were extracted by liquid–liquid extraction. The extracted samples were analyzed by high‐performance liquid chromatography equipped with an AB SCIEX QTRAP^® 5500 (for plasma) or 6500 (for urine) triple quadrupole mass spectrometer (SCIEX, Framingham, MA), using an electrospray ionization source. Positive ions were monitored in the multiple reaction monitoring mode, and quantitation was determined using a weighted linear regression analysis (L/concentration²) of peak area ratios of the analyte and IS. The analytic ranges for plasma were 10.0–3500 pg/mL and for urine were 0.100–100 ng/mL.

Rosuvastatin and the IS (¹³C,d₃‐rosuvastatin sodium) were extracted from human plasma samples by liquid–liquid extraction, and mass spectrometric detection and quantitation was performed as described for digoxin using an AB SCIEX Quad™ 6500 mass spectrometer. The analytic range for plasma rosuvastatin was 0.100–50.0 ng/mL.

All samples for a given participant were analyzed together in a single run that at a minimum included one control (no IS), one standard zero sample (IS only), one replicate of at least six (plasma) or 10 (urine) different calibration standards (non‐zero standards), and replicate low, medium, and high concentration quality control samples. Standards were rejected if they were greater than ±15.0% (all standards except the lower limit of quantitation) or ±20.0% (lower limit of quantitation only) of the nominal concentration. For the run to be accepted, ≥75% of the non‐zero standards must have been within the respective acceptance criterion, and at least two‐thirds of the low, medium, and high QC samples, including at least 50% at each concentration, must have been valid data points and within ±15.0% of the nominal concentration.

Plasma PK parameters assessed for digoxin, rosuvastatin, and futibatinib included C _max; T _max; t _½; AUC from time 0 to 24 (futibatinib; AUC_0–24), 72 (rosuvastatin; AUC_0–72), or 120 (digoxin; AUC_0–120) hours after dosing; AUC from time zero to infinity (AUC_inf); apparent total plasma clearance; and apparent volume of distribution. PK parameters for urine digoxin included total amount of drug excreted unchanged in urine over the entire period of sample collection (Ae); renal clearance, calculated as Ae[0–t″]/AUC[0–t″], where 0–t″ was the longest interval of time during which Ae and AUC were both obtained; and fraction of drug excreted unchanged in urine. Plasma PK parameters were estimated via noncompartmental analysis using Phoenix^® WinNonlin^® version 8.3.4, and urinary PK parameters were calculated using SAS^® version 9.4 or higher.

Safety evaluation

Safety was monitored throughout the study until 14 days after the last dose or resolution of adverse events (AEs). Safety assessments consisted of physical examination, 12‐lead electrocardiograms (ECGs), vital signs, and clinical laboratory testing. AEs were coded using the Medical Dictionary for Regulatory Activities version 26.0 and graded per the National Cancer Institute's Common Terminology Criteria for Adverse Events version 5.0.

Statistical analysis

The sample size (part 1, n = 20; part 2, n = 24) was selected without statistical considerations but was considered sufficient to evaluate the magnitude of any potential DDI considering the possibility of early discontinuation. A larger sample size was chosen for part 2 due to the known QT prolongation effects of quinidine, ²² , ²³ which were expected to result in a greater dropout rate.

The PK population included all participants who complied sufficiently with the protocol and had ample PK samples collected to display an evaluable PK profile. The safety population included all participants who received at least one dose of the study drug.

PK parameters and safety data were summarized using descriptive statistics. Log‐transformed PK parameters were compared between treatments using a one‐way analysis of variance, for which treatment was a fixed effect and participant was a random effect. The geometric mean ratios (GMRs) and associated 90% confidence intervals (CIs) were estimated based on the least square means from the analysis of variance.

RESULTS

Participants

A total of 20 participants were enrolled in part 1, 19 of whom completed the study. One participant was discontinued by the investigator on day 6 of period 2 due to a treatment‐emergent AE (TEAE) of grade 2 increased alanine aminotransferase. This participant had a plasma digoxin percentage of AUC_inf extrapolated of >30% in period 2 and was therefore not included in the reporting of AUC_inf for this period. All 20 participants were included in the safety analysis.

Of 24 participants enrolled in part 2, 15 completed the study, and nine participants discontinued on days 2–3 of period 2 due to 12‐lead ECG‐related TEAEs (grade 3) following quinidine administration. All 24 participants were included in the safety analysis and PK analysis for period 1, and 15 participants were included in the PK analysis for period 2.

Baseline demographics in the safety population are summarized in Table 1.

TABLE 1.

Participant demographics.

	Part 1 (n = 20)	Part 2 (n = 24)	Total (N = 44)
Median (range) age, years	37.5 (21–53)	49.0 (20–55)	39.5 (20–55)
Male, n (%)	13 (65)	16 (67)	29 (66)
Race, n (%)
Black or African American	1 (5)	1 (4)	2 (5)
White	19 (95)	23 (96)	42 (95)
Ethnicity, n (%)
Hispanic or Latino	14 (70)	19 (79)	33 (75)
Not Hispanic or Latino	6 (30)	5 (21)	11 (25)
Mean (SD) BMI, kg/m2	27.6 (2.42)	26.4 (3.61)	26.9 (3.15)
Mean (SD) height, cm	172.0 (10.88)	169.9 (9.58)	170.8 (10.13)
Mean (SD) body weight, kg	82.1 (14.80)	76.4 (13.03)	78.9 (13.98)

Abbreviations: BMI, body mass index; SD, standard deviation.

Effects of futibatinib on digoxin PK

Following oral administration of digoxin and rosuvastatin, digoxin plasma concentration–time profiles with and without futibatinib were similar throughout the sampling duration (Figure 2a; Table 2). Mean t _½ was similar regardless of futibatinib coadministration, indicating that the elimination rate of digoxin was unchanged: digoxin concentrations declined in parallel following both treatments. Median T _max was delayed by ~0.3 h following administration of digoxin, rosuvastatin, and futibatinib. The GMRs for digoxin C _max, AUC_0–120, and AUC_inf (with vs. without futibatinib) were 95.1%, 103.9%, and 100.2%, respectively, with all 90% CIs fully within the 80%–125% bioequivalence limits (Table 2). Mean renal clearance was 13% lower when digoxin was administered with futibatinib, with the 90% CIs of the GMR for renal clearance also within the lower limit of bioequivalence (87.1% [90% CI: 80.8–93.9]).

FIGURE 2.

Mean plasma (a) digoxin and (b) rosuvastatin concentration–time profiles when administered alone and in combination with futibatinib (n = 20). Linear scale (left); semi‐log scale (right).

TABLE 2.

Summary of plasma and urine digoxin PK parameters following single oral doses of 0.25 mg digoxin and 10 mg rosuvastatin with and without multiple oral doses of 20 mg futibatinib once daily (n = 20).

PK parameter	Digoxin + rosuvastatin	Digoxin + rosuvastatin + futibatinib	GMR, % (90% CI)
AUC0–120, pg*h/mL	15,640 (26.0)	16,240 (29.3)	103.9 (94.5–114.2)
AUCinf, pg*h/mL	17,670 (26.6)	17,950 (29.9) a	100.2 (90.8–110.5)
C max, pg/mL	1003 (38.0)	953.3 (45.1)	95.1 (81.1–111.5)
CLr, L/h	7.3 ± 1.3	6.5 ± 2.0	87.1 (80.8–93.9)
T max, h	1.0 (0.67, 2.0)	1.3 (0.67, 3.0)
t ½, h	39.2 ± 5.7	37.7 ± 11.1
CL/F, L/h	14.6 ± 4.1	14.6 ± 5.4 a
Vz/F, L	820.4 ± 223.8	746.4 ± 293.2 a

Note: AUCs and C _max values are presented as geometric mean (geometric CV%); T _max values are presented as median (min, max); other parameters are presented as arithmetic mean ± SD.

Abbreviations: AUC_0–120, area under the plasma concentration–time curve from time zero to 120 h after dosing; AUC_inf, area under the plasma concentration–time curve from time zero to infinity; AUC%extrap, percentage of AUC_inf extrapolated; CI, confidence interval; CL/F, apparent total plasma clearance; CLr, renal clearance; C _max, maximum plasma concentration; CV, coefficient of variation; GMR, geometric mean ratio; PK, pharmacokinetic; SD, standard deviation; t _½, apparent terminal elimination half‐life; T _max, time to maximum plasma; Vz/F, apparent volume of distribution.

^a Summary only includes 19 participants with AUC%extrap <30%.

Effects of futibatinib on rosuvastatin PK

Rosuvastatin plasma concentrations over time with and without futibatinib coadministration exhibited a similar profile, albeit with slightly higher mean concentrations after coadministration with futibatinib (Figure 2b). Median T _max was similar with and without futibatinib coadministration, whereas mean C _max and AUC were slightly higher, and mean t_½ was shorter when rosuvastatin was administered with futibatinib (Table 3). Geometric mean C _max and AUCs of rosuvastatin were <20% higher (C_max 10.2% and AUCs ≤18.8%) after coadministration with futibatinib compared with rosuvastatin plus digoxin. The observed ranges for C _max and AUC_inf with rosuvastatin plus digoxin versus coadministration with futibatinib were 1.06–10.2 versus 1.73–15.2 ng/mL and 12.3–129 versus 21.6–168 ng*h/mL, respectively. The 90% CIs of the GMR for C _max were within bioequivalence limits (80%–125%); the upper 90% CIs of the GMRs for AUC_0–72 (118.8% [90% CI: 107.9–130.7]) and AUC_inf (113.5% [90% CI: 102.6–125.5]) were slightly higher than the upper limit of bioequivalence.

TABLE 3.

Summary of plasma rosuvastatin PK parameters following single oral doses of 0.25 mg digoxin and 10 mg rosuvastatin with and without multiple oral doses of 20 mg futibatinib once daily (n = 20).

PK parameter	Digoxin + rosuvastatin	Digoxin + rosuvastatin + futibatinib	GMR, a % (90% CI)
AUC0–72, ng*h/mL	45.9 (53.4) b	53.9 (54.8)	118.8 (107.9–130.7)
AUCinf, ng*h/mL	46.9 (56.7) c	54.7 (56.4)	113.5 (102.6–125.5)
C max, ng/mL	4.4 (61.4)	4.8 (65.5)	110.2 (97.3–124.8)
T max, h	4.0 (1.0, 6.1)	4.0 (1.5, 6.1)
t ½, h	16.7 ± 8.5 c	10.1 ± 5.8
CL/F, L/h	247.3 ± 167.2 c	207.2 ± 105.7
Vz/F, L	5083 ± 2263.5 c	2617 ± 1163.9

Note: AUCs and C _max values are presented as geometric mean (geometric CV%); T _max values are presented as median (min, max); other parameters are presented as arithmetic mean ± SD.

Abbreviations: AUC_0–72, area under the plasma concentration–time curve from time zero to 72 h after dosing; AUC_inf, area under the plasma concentration–time curve from time zero to infinity; CI, confidence interval; CL/F, apparent total plasma clearance; C_max, maximum plasma concentration; CV, coefficient of variation; GMR, geometric mean ratio; PK, pharmacokinetic; SD, standard deviation; t _½, apparent terminal elimination half‐life; T _max, time to maximum plasma; Vz/F, apparent volume of distribution.

^a Only participants for whom PK parameters were determined for both period 1 and period 2 were included in the comparison.

^b Data available for 19 participants.

^c Data available for 18 participants.

Effects of quinidine on futibatinib PK

Futibatinib plasma concentration–time profiles were similar with and without quinidine coadministration, but there were slightly higher peak futibatinib concentrations following quinidine coadministration (Figure 3). Median T _max was consistent regardless of quinidine coadministration, whereas mean AUC and C_max were slightly higher and mean t_½ was slightly longer with quinidine coadministration (Table 4). The upper 90% CIs for GMRs of C _max (108.0% [90% CI: 90.9–128.3]), AUC_0–24 (116.4% [90% CI: 95.4–142.1]), and AUC_inf (117.0% [90% CI: 96.0–142.7]) were slightly higher than the upper limit of bioequivalence, and the magnitudes of increases in C _max and AUC with quinidine coadministration compared with futibatinib alone were <20% (8%, 16%, and 17% for C _max, AUC_0–24, and AUC_inf, respectively). The observed ranges for C _max and AUC_inf with futibatinib alone versus coadministration with quinidine were 44.5–380 versus 77.4–345 ng/mL and 95–1300 versus 326–1214 ng*h/mL, respectively.

FIGURE 3.

Mean plasma futibatinib concentration–time profiles when administered alone and in combination with quinidine (n = 15^a). Linear scale (left); semi‐log scale (right). ^aA total of nine participants were discontinued prior to futibatinib administration in period 2 due to change in corrected QT interval using Fridericia's formula >60 ms while on quinidine alone.

TABLE 4.

Summary of PK parameters for futibatinib when administered alone and in combination with quinidine.

PK parameter	Futibatinib (n = 24)	Futibatinib + quinidine (n = 15)	GMR, a % (90% CI)
AUC0–24, ng*h/mL	579.5 (66.1)	622.8 (41.4)	116.4 (95.4–142.1)
AUCinf, ng*h/mL	581.5 (66.3)	627.0 (41.6)	117.0 (96.0–142.7)
C max, ng/mL	151.4 (53.5)	161.1 (48.4)	108.0 (90.9–128.3)
T max, h	1.5 (1.0, 5.0)	1.3 (0.76, 3.0)
t ½, h	2.5 ± 0.83	3.2 ± 1.1
CL/F, L/h	43.1 ± 40.8	34.3 ± 13.2
Vz/F, L	132.0 ± 72.2	149.3 ± 56.6

Note: AUCs and C _max values are presented as geometric mean (geometric CV%); T _max values are presented as median (min, max); other parameters are presented as arithmetic mean ± SD.

Abbreviations: AUC_0–24, area under the plasma concentration–time curve from time zero to 25 h after dosing; AUC_inf, area under the plasma concentration–time curve from time zero to infinity; CI, confidence interval; CL/F, apparent total plasma clearance; C _max, maximum plasma concentration; CV, coefficient of variation; GMR, geometric mean ratio; PK, pharmacokinetic; SD, standard deviation; t _½, apparent terminal elimination half‐life; T _max, time to maximum plasma; Vz/F, apparent volume of distribution.

^a Only participants for whom PK parameters were determined for both period 1 and period 2 were included in the comparison.

Safety

Most participants in part 1 (n = 19; 95%) experienced at least one TEAE, the most common of which were diarrhea (80%), increased blood phosphate (75%), abdominal pain (30%), dry mouth (25%), abdominal discomfort (20%), and constipation (20%; Table S2). Of a total of 114 TEAEs, 98 events were considered grade 1, 11 events grade 2, and five events grade 3 (four events of diarrhea and one event of hematochezia). The investigator considered 94 events to be related to futibatinib, 45 to be related to digoxin, 24 to be related to rosuvastatin, and 10 to be unrelated to any of the three treatments. With the exception of one increased blood phosphate event, all events resolved within 10 days. One participant discontinued treatment due to a grade 2 TEAE of increased alanine aminotransferase that was considered related to futibatinib and rosuvastatin. Three participants received concomitant medication (acetaminophen [headache and back pain], hydrocortisone cream [arthropod bite], and acidophilus probiotic [diarrhea]). All diarrhea events resolved without the need for dose modifications. None of the 15 participants with increased blood phosphate exhibited clinical symptoms associated with hyperphosphatemia or required any phosphate‐lowering therapy.

In part 2, 15 (63%) participants experienced at least one TEAE. The most common TEAEs were prolonged ECG QT (i.e., corrected QT interval using Fridericia's formula; 38%) and diarrhea (13%); all other TEAEs were each reported in two (8%) or fewer participants (Table S2). Of a total of 27 TEAEs, most (n = 17) were grade 1 in intensity, one event was grade 2, and nine events were grade 3. The investigator considered one event to be related to futibatinib and 22 events to be related to quinidine; the remaining four events were considered to be unrelated to either treatment. One participant received acetaminophen for the treatment of headache. Nine participants discontinued treatment due to events of grade 3 prolonged ECG QT following administration of quinidine alone; all events were considered related to quinidine treatment, and all resolved. There were no deaths or serious AEs in either part of the study.

DISCUSSION

Investigation of potential DDI is important for informing clinical decision‐making in cancer treatment. In vitro studies identified futibatinib as an inhibitor of P‐gp and BCRP with half‐maximal inhibitory concentration values of 0.296 ¹⁵ and 0.348 μmol/L and R‐values of 646 and 549, respectively (data on file). The R‐values were calculated by dividing the intestinal luminal concentration of the inhibitor by the half‐maximal inhibitory concentration. Here, an R‐value threshold of 10 was used to calculate interaction risk at the level of the intestine, whereby R‐values ≥10 would warrant further investigation of DDI potential. ²⁴ Accordingly, the potential for DDI could not be excluded at clinically relevant concentrations, and a clinical DDI study was conducted to further investigate the potential impact of futibatinib coadministration on P‐gp and BCRP substrates.

Digoxin and rosuvastatin were selected as clinical substrates for P‐gp and BCRP, respectively, as recommended by the US Food and Drug Administration (FDA). ²⁵ Rosuvastatin is also a known substrate of sodium taurocholate cotransporting polypeptide; however, because there is no regulatory requirement to assess for interactions, futibatinib was not evaluated as a potential inhibitor of this carrier protein. Digoxin and rosuvastatin were administered together as a well‐established transporter probe cocktail for which interaction among components has not been detected. ²⁶ , ²⁷ , ²⁸ Quinidine is recommended by the US FDA for use as an inhibitor of P‐gp in clinical DDI studies. ²⁵ Futibatinib is primarily metabolized by CYP3A4 and to a lesser extent by CYP2D6; therefore, despite quinidine being a strong inhibitor of CYP2D6, ²⁹ , ³⁰ , ³¹ it was selected over other commonly used clinical P‐gp inhibitors that inhibit CYP3A4, as inhibition of CYP2D6 was not expected to impact futibatinib PK.

In this phase I study, futibatinib did not affect the exposure of digoxin, a P‐gp substrate, with the GMRs for C _max, AUC_0–120, and AUC_inf all close to 100%. Although the study was not powered for bioequivalence evaluation, the 90% CIs were entirely within the bioequivalence limits of 80%–125%, indicating a lack of DDI effect with digoxin. For rosuvastatin (a BCRP substrate), coadministration with futibatinib had no clinically relevant effect on plasma exposure in healthy adult participants, as the rosuvastatin exposure increase was <20% for AUC and C _max compared with treatment without futibatinib, and the observed ranges showed large overlaps. Overall, therefore, these results suggest that futibatinib has no clinically meaningful effect on the exposures of P‐gp or BCRP substrates.

In the second part of this study, coadministration with quinidine (a P‐gp inhibitor) had no clinically relevant effect on plasma futibatinib exposure in healthy adult participants, as the futibatinib exposure increase was also <20% for AUC and C _max compared with futibatinib alone. Furthermore, the observed ranges for C _max and AUC_inf with futibatinib alone fully encompassed the ranges observed when futibatinib was coadministered with quinidine. Thus, the contribution of P‐gp to futibatinib absorption appears to be limited, and clinical perpetrators of P‐gp are unlikely to have a clinically meaningful effect on the bioavailability of futibatinib. The observed increase in futibatinib exposure (17%) with quinidine coadministration supports the hypothesis that the previously observed effects of itraconazole on futibatinib PK resulted from the combined effects of P‐gp and CYP3A inhibition. ¹⁵ Notably, there was a 2‐h interval between the administration of futibatinib and that of digoxin and rosuvastatin (part 1) or quinidine (part 2). While it cannot be excluded simultaneous administration of the drugs may have modified the magnitude of DDI, it is unlikely that it would have changed the observed findings.

Futibatinib was well tolerated in healthy adult participants when administered with and without digoxin and rosuvastatin, or quinidine, and the safety profile was consistent with prior evaluations. ³² Hyperphosphatemia is an on‐target effect of pan‐FGFR inhibitors and a known side effect of futibatinib. ³² , ³³ Consistent with previous studies in healthy volunteers, hyperphosphatemia was not associated with any clinical symptoms and was reversible upon discontinuation of futibatinib administration without any further intervention. Considering the dose‐dependent QT‐prolonging effects of quinidine, ²² the lowest daily dose proven to show renal P‐gp inhibition (200 mg four times a day) was selected for this study. Although the most common TEAE observed in part 2 of the study after administration of quinidine alone was QT prolongation, the effects of the corrected QT interval using Fridericia's formula were not associated with any clinical symptoms and were reversible upon discontinuation of quinidine administration.

In conclusion, the results of this analysis confirm that futibatinib has no clinically meaningful effects on the PK of substrates of P‐gp or BCRP and that the effect of quinidine, a P‐gp inhibitor, on futibatinib bioavailability is limited. Therefore, although concomitant use of dual perpetrators of CYP3A and P‐gp should be avoided, medications that inhibit and/or induce only P‐gp can be used concomitantly with futibatinib. Overall, futibatinib has been shown to exhibit a favorable clinical pharmacology profile, and the potential DDI risk is limited to dual P‐gp and strong CYP3A inhibitors/inducers.

AUTHOR CONTRIBUTIONS

All authors wrote the manuscript; all authors designed the research; all authors performed the research; all authors analyzed the data.

FUNDING INFORMATION

This study was funded by Taiho Oncology, Inc.

CONFLICT OF INTEREST STATEMENT

A.L., N.H., B.A., and V.W. are employees of Taiho Oncology, Inc. L.G. was an employee of Taiho Oncology, Inc. at the time of study and manuscript preparation. I.Y. is an employee of Taiho Pharmaceutical Co, Ltd. M.V. and Z.M. are employees of Celerion.

Supporting information

Table S1–Table S2.

Long A, Yamamiya I, Valentine M, et al. A phase I drug–drug interaction study to assess the effect of futibatinib on P‐gp and BCRP substrates and of P‐gp inhibition on the pharmacokinetics of futibatinib. Clin Transl Sci. 2024;17:e70012. doi: 10.1111/cts.70012