Effects of GLPG3970 on Sulfasalazine and Methotrexate Pharmacokinetics in Healthy Adults: Two Open‐Label, Phase I, Drug–Drug Interaction Studies

Abstract

GLPG3970 is a selective salt‐inducible kinase 2/3 inhibitor intended for the treatment of inflammatory diseases. In vitro studies suggest GLPG3970 strongly inhibits breast cancer resistance protein (BCRP), indicating a possible interaction with BCRP substrates such as sulfasalazine (SSZ; a probe substrate for intestinal BCRP inhibition) and methotrexate (MTX), both inflammatory disease medications. Two open‐label, nonrandomized, phase I, drug–drug interaction (DDI) studies assessed the pharmacokinetics of SSZ 1,000 mg (NCT04720183) and MTX 7.5 mg (EudraCT: 2020‐000391‐37) with and without GLPG3970 350 mg. Healthy participants aged 18–55 years with wild‐type homozygous BCRP genotype (c421C/C) received: SSZ on day (D)1, GLPG3970 + SSZ on D5, and GLPG3970 2 hours after SSZ on D9 (N = 8; SSZ/GLPG3970 DDI study); MTX on D1, GLPG3970 + MTX on D5, and GLPG3970 on D6–8 (N = 15; MTX/GLPG3970 DDI study). Primary end points were AUC and C _max (“exposure”) of SSZ and its metabolite (sulfapyridine [SPD]), SPD:SSZ AUC ratio (SSZ/GLPG3970 study), and AUC and C _max (“exposure”) of MTX (MTX/GLPG3970 study). DDIs were evaluated using the geometric mean ratio of each end point; a > 2‐fold increase in SSZ or MTX exposure was deemed clinically relevant. GLPG3970 demonstrated mild inhibition of intestinal BCRP in vivo: GLPG3970 + SSZ increased SSZ exposure ~1.7–1.8‐fold and decreased SPD:SSZ ratio ~ 2‐fold vs. SSZ alone. GLPG3970 administered 2 hours after SSZ did not change the magnitude of the interaction. GLPG3970 + MTX had no relevant effect on MTX pharmacokinetics vs. MTX alone. Therefore, the strong in vitro BCRP inhibition was not confirmed in vivo. No safety concerns were observed when GLPG3970 was coadministered with SSZ or MTX.

Study Highlights.

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

FDA clinical guidelines indicate that sponsors should assess the DDI potential of new chemical entities with concomitant medications before their administration to patients if there is an interaction risk. Sulfasalazine is a proposed probe substrate to investigate intestinal breast cancer resistance protein (BCRP) inhibition in vivo.

• WHAT QUESTION DID THIS STUDY ADDRESS?

We investigated whether in vitro results indicating strong BCRP inhibition by GLPG3970, a small‐molecule SIK2/3 inhibitor, are also observed in vivo. GLPG3970 could be used in trials in patients with inflammatory disease who concomitantly use sulfasalazine and/or methotrexate, both of which are BCRP substrates.

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

GLPG3970 showed a mild inhibition of BCRP in vivo (sulfasalazine exposure increased less than twofold upon GLPG3970 coadministration) despite the strong in vitro effect. Methotrexate exposure was unchanged upon GLPG3970 coadministration. Conducting a DDI study with SSZ presents practical challenges that are highlighted herein.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

In vitro BCRP studies can be more sensitive than in vivo assessments; further refinement is required to improve DDI risk assessment.

GLPG3970 is an oral, small‐molecule, selective inhibitor of salt‐inducible kinase (SIK) 2/3. ¹ Preclinical data have shown that inhibition of SIK2/3 by GLPG3970 blocks the production of proinflammatory cytokines, such as tumor necrosis factor‐α and interleukin (IL)‐12, while simultaneously increasing the production of IL‐10, which has an immune‐regulatory role. ¹ , ² , ³ This mode of action gives SIK2/3 inhibitors therapeutic potential in various autoimmune inflammatory diseases, such as rheumatoid arthritis, inflammatory bowel disease, and psoriasis.

The first‐in‐human study of GLPG3970 (NCT04106297) evaluated the safety and pharmacokinetics (PK) of single and multiple ascending doses of GLPG3970 in healthy volunteers. ⁴ GLPG3970 was well tolerated, with no deaths or serious treatment‐emergent adverse events (TEAEs) reported. Following both single (up to 600 mg) and repeated (up to 600 mg once daily) dosing as an oral solution, GLPG3970 was quickly absorbed (time of occurrence of maximum concentration [t _max]: 0.5–3.0 hours post‐dosing). GLPG3970 exposure (area under the concentration–time curve [AUC]) generally increased more than the dose proportionally across the evaluated dose range, and the GLPG3970 elimination half‐life (t _1/2) was 9.5–15.0 hours.

Breast cancer resistance protein (BCRP) is an efflux transporter that restricts the distribution of its substrates into organs by limiting their intestinal absorption. ⁵ It is included in the list of important drug transporters that both the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) consider necessary to investigate regarding interactions with new chemical entities. ⁶ , ⁷ In vitro assays suggest that GLPG3970 is an inhibitor of BCRP (unpublished data). Based on the observed in vitro half maximal inhibitory concentration of GLPG3970 for BCRP, and the anticipated in vivo maximum observed free plasma concentration (C _max) at steady state at the highest anticipated therapeutic dose of GLPG3970, there is a potential risk of interaction between GLPG3970 and BCRP substrates.

Sulfasalazine (SSZ) and methotrexate (MTX) are common background therapies for rheumatoid arthritis and inflammatory bowel disease, ⁸ , ⁹ and both are substrates of BCRP. ⁷ SSZ has been proposed as a clinical probe for intestinal BCRP inhibition given that SSZ is restricted largely to the gastrointestinal tract owing to its low gastrointestinal permeability and low solubility, and efficient intestinal BCRP efflux. ¹⁰ SSZ is poorly absorbed following oral administration (< 15%) and systemic concentrations are detectable at ~ 1.5 hours post‐dosing, with the t _max reached at 3–12 hours post‐dosing.

SSZ is metabolized extensively in the intestine by bacteria to sulfapyridine (SPD), which reaches its t _max at ~ 10 hours post‐dosing. The rate of SPD metabolism is dependent on the N‐acetyltransferase 2 (NAT2) phenotype; in individuals with fast and slow acetylation phenotypes, the mean plasma t _1/2 is 10.4 and 14.8 hours, respectively. Inhibition of intestinal BCRP reduces SPD formation by increasing SSZ absorption, thus reducing the total amount of SSZ in the intestinal lumen available for metabolism. ¹¹ As a result, the metabolite‐to‐parent (SPD:SSZ) ratio is a more sensitive indicator of intestinal BCRP modulation in humans than SSZ AUC alone, ¹⁰ especially in individuals with rapid and intermediate acetylation phenotypes. ¹²

BCRP inhibition may also decrease MTX elimination in vivo. ¹³ , ¹⁴ In general, MTX is quickly absorbed, reaching the t _max at 1–2 hours post‐dosing, and has ~ 60% bioavailability following oral dosing. MTX has a mean t _1/2 of 3–10 hours. MTX excretion takes place mainly via the kidneys, with 80–90% excreted unchanged in urine within 24 hours. Approximately 10% of an MTX dose is metabolized in the liver, and its main metabolite is 7‐hydroxy‐methotrexate (7‐OH‐MTX). ¹⁵ MTX toxicity may be related to the inhibition of renal excretion via various mechanisms including the inhibition of BCRP. ¹³ Additionally, the 7‐OH‐MTX concentrations resulting from MTX metabolism, which can exceed those of MTX, can further enhance MTX‐induced toxicity. ¹⁶

Given that SSZ and MTX are common background therapies for inflammatory diseases, these therapies could be taken by participants in clinical studies of GLPG3970. Therefore, these two drug–drug interaction (DDI) studies described herein aimed to investigate the effects of GLPG3970 on the PK of SSZ and its metabolite SPD (SSZ/GLPG3970 DDI study), and of MTX and its metabolite 7‐OH‐MTX (MTX/GLPG3970 DDI study). The safety profile of GLPG3970 when coadministered with SSZ or MTX was also assessed.

METHODS

Participants

Healthy male and female participants of non‐childbearing potential aged 18–55 years with a body mass index of ≥ 18.0 to ≤ 30.0 kg/m² were eligible for these DDI studies. Furthermore, participants were required to be homozygous for the wild‐type BCRP genotype (c421C/C; “reference genotype”). This requirement was owing to previous study findings suggesting that several single‐nucleotide polymorphisms affect the functionality of BCRP, which can decrease the impact of BCRP inhibition by exogenous components and consequently affect the assessment of the worst‐case scenario. ¹⁷ Participants were excluded if they had a known hypersensitivity to any of the study drugs (i.e., GLPG3970, SSZ, sulfonamide drugs, or MTX) in the respective studies. For the SSZ/GLPG3970 DDI study, further key exclusion criteria included a family history of Stevens–Johnson syndrome, which has previously been reported in patients with ulcerative colitis shortly after starting daily SSZ treatment, ¹⁸ or a NAT2 slow acetylation genotype, to reduce inter‐participant variability and to provide greater sensitivity for determining a clinical interaction with BCRP.

Study designs

Both phase I DDI studies (SSZ/GLPG3970 DDI, NCT04720183; MTX/GLPG3970 DDI, EudraCT: 2020–000391–37) employed an open‐label, nonrandomized, fixed‐sequence design. SSZ/GLPG3970 DDI was conducted at a single site in the USA and MTX/GLPG3970 DDI at a single site in the Netherlands; participants were confined to the site for the entire duration of each study. The respective study designs, outlining the screening, treatment, and follow‐up periods are shown in Figure 1 . The duration of the treatment periods in the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies (12 and 9 days, respectively) were designed to take into account the t _1/2 of SSZ and MTX and their respective metabolites, as well as for the perpetrator GLPG3970. Owing to the hepatic metabolism of MTX, GLPG3970 was repeatedly administered over several days to ensure its presence for the full duration of MTX exposure.

Figure 1.

Study designs of (a) SSZ/GLPG3970 and (b) MTX/GLPG3970 DDI studies. AE, adverse event; BCRP, breast cancer resistance protein, DDI, drug–drug interaction; ECG, electrocardiogram; MTX, methotrexate; NAT2, N‐acetyltransferase 2; PK, pharmacokinetic; SSZ, sulfasalazine.

Participants received SSZ 1,000 mg and MTX 7.5 mg as oral tablets in the respective studies. The chosen doses were low and within the therapeutic ranges used in the treatment of inflammatory diseases (SSZ 500–4,000 mg and MTX 7.5–30 mg), to provide an acceptable safety margin in the event of increased exposures in the presence of GLPG3970. The dose regimen selected for the DDI assessments in these studies was GLPG3970 350 mg once daily. This dose was selected based on the results of the first‐in‐human study, ⁴ in which the safety profile and tolerability were established for a range of doses (doses of 10–500 mg), and was anticipated to be within the therapeutic dose range. GLPG3970 350 mg was administered once daily as an oral solution in both studies.

In the SSZ/GLPG3970 DDI study, participants received a single dose of SSZ in the morning on day 1 (treatment period 1). On the morning of day 5, participants received a single dose of GLPG3970 immediately before SSZ, and on the morning of day 9, participants received a single dose of GLPG3970 2 hours post‐SSZ. As preliminary work demonstrated that food consumption reduced the C _max of GLPG3970, all doses were administered after an overnight fast of at least 10 hours. In addition, participants had to remain fasted for at least 4 hours post‐SSZ dosing on day 1, day 5, and day 9. In the MTX/GLPG3970 DDI study, participants received a single dose of MTX in the morning on day 1 (treatment period 1) and day 5 (treatment period 2), after an overnight fast of at least 10 hours to maximize GLPG3970 exposure. On day 1, participants had to remain fasted for at least 4 hours post‐MTX dosing. On day 5, participants received a single dose of GLPG3970 immediately before MTX and then remained fasted for at least 4 hours. On days 6–8, participants had to be fasted for at least 2 hours in the morning before receiving GLPG3970 and were then allowed breakfast 1 hour post‐GLPG3970 dosing.

In the SSZ/GLPG3970 DDI study, blood samples were collected for PK analyses of SSZ and SPD before each SSZ dosing and at regular intervals up to 72 hours after each SSZ dosing. For PK analyses of GLPG3970, blood samples were collected before GLPG3970 dosing and at intervals up to 24 hours post‐dosing on days 5 and 9 when GLPG3970 was coadministered with SSZ. In the MTX/GLPG3970 DDI study, blood samples were collected for PK analyses of MTX and 7‐OH‐MTX before MTX dosing and at regular intervals up to 96 hours post‐MTX dosing on days 1 and 5. For PK analyses of GLPG3970, blood samples were collected before GLPG3970 administration on all GLPG3970 dosing days, and at regular intervals up to 24 hours post‐dosing on day 5.

The plasma concentrations of GLPG3970, SSZ, SPD, MTX, and 7‐OH‐MTX were determined using validated liquid chromatography–tandem mass spectrometry (LC–MS); the assay parameters are shown in Table S1 . The lower limit of quantification (LLOQ) was 1 ng/mL for GLPG3970, 10 ng/mL for both SSZ and SPD, and 5 ng/mL for both MTX and 7‐OH‐MTX. The accuracy and precision statistics for the plasma concentration analyses in both studies are shown in Table S2 . BCRP and NAT2 genotyping were performed by contract research laboratories during screening (SSZ/GLPG3970 DDI study: DNA Vision, Charleroi, Belgium; MTX/GLPG3970 DDI study [BCRP genotyping only]: Erasmus University Medical Center, Rotterdam, The Netherlands).

Study outcomes

The primary PK end points in the SSZ/GLPG3970 DDI study were the AUC_0–∞ and C _max of SSZ and SPD, and the SPD:SSZ AUC_0–t and AUC_0–∞ ratios. In the MTX/GLPG3970 DDI study, the primary PK end points were the AUC_0–∞ and C _max of MTX and 7‐OH‐MTX. Per EMA guidelines, ¹⁹ a clinically significant DDI was defined as a more than twofold increase in SSZ or MTX exposure, in terms of the primary PK end points. As low doses of SSZ and MTX were used, small increases in exposure of greater than twofold would likely still result in concentrations within the therapeutic window. The secondary PK end points were the C _max and AUC_0–t of GLPG3970 in the SSZ/GLPG3970 DDI study, and the C _max, t _max, and AUC_0–t of GLPG3970 in the MTX/GLPG3970 DDI study. Additional PK end points included the AUC_0–t and t _1/2 of SSZ, SPD, MTX, and 7‐OH‐MTX. Furthermore, both studies evaluated the safety of coadministration of GLPG3970 with SSZ or MTX, as assessed by the incidence and severity of TEAEs. TEAEs were defined as any adverse event (AE) with onset after the start of SSZ, MTX, SSZ/GLPG3970, or MTX/GLPG3970 intake. TEAEs and other safety‐related assessments (laboratory assessments, physical examination, 12‐lead electrocardiogram, and vital signs) were assessed during the screening, treatment, and follow‐up periods.

Statistical analysis

No formal sample size calculation was performed for either study. For the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies, 10 and 12 participants, respectively, were expected to be sufficient to detect a clinically significant DDI. To account for dropouts, up to 12 and 15 participants were planned to be enrolled in the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies, respectively. The safety analysis sets included all participants who received SSZ, MTX, SSZ with GLPG3970, or MTX with GLPG3970 at least once. The PK analysis sets included all participants from the safety analysis sets who had available and evaluable PK data, excluding data following protocol deviations or AEs that could have an impact on the PK analysis.

Descriptive statistics included the arithmetic mean, standard deviation, median, minimum, and maximum for continuous variables, and the number and percentage per category for categorical variables. For PK data, the geometric mean, coefficient of variation (CV), and geometric CV were also derived. Demographic data for participants in both studies were analyzed descriptively, using the safety analysis set.

DDI assessments were performed in the PK analysis set using natural logarithm‐transformed SSZ, SPD, MTX, and 7‐OH‐MTX PK parameters in a mixed‐effects model with the treatment as a fixed effect and participant as a random effect. Point estimates were calculated as the geometric least‐squares mean of the individual ratios of each parameter for the test/reference treatments and expressed as a percentage. The 90% confidence interval (CI) of the point estimates was calculated using the mean square error of the analysis of variance. Point estimates and 90% CIs were calculated for the following comparisons: SSZ and GLPG3970 (coadministration and deferred administration) vs. SSZ alone (reference); and MTX and GLPG3970 (coadministration) vs. MTX alone (reference).

Missing data were not imputed in either study. For the PK analyses, values below the LLOQ were imputed with 0 for the calculation of descriptive statistics and graphical presentation, except for the geometric mean and the geometric CV, for which data were imputed as the LLOQ/2. The PK parameters were calculated using Phoenix WinNonLin version 8 (Certara, Princeton, NJ, USA). All statistical analyses were performed using Statistical Analysis System version 9.4 (SAS Institute, Cary, NC, USA).

Ethics statement

The studies were carried out in accordance with the principles of the Declaration of Helsinki, the guidelines of the International Conference on Harmonisation for Good Clinical Practice, and local ethical and legal requirements. The SSZ/GLPG3970 and MTX/GLPG3970 DDI study protocols were approved by the institutional review board and the independent ethics committee, respectively, according to local regulations before initiation. All participants in both studies provided written informed consent before any clinical study procedures commenced.

RESULTS

Participant disposition and demographics

In the SSZ/GLPG3970 DDI study, 8 out of 12 participants planned for enrollment were screened and enrolled. This was fewer than the planned number of participants expected to be required to detect a clinically significant DDI (n = 12), and was largely due to constraints in participant recruitment. However, based on analyses of available data from the 8 participants, it was concluded that there were sufficient data to detect a more than twofold increase in exposure and consequently draw a conclusion from the study. In the MTX/GLPG3970 DDI study, all 15 participants planned for enrollment were screened and enrolled. Study completion rates were high in both the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies: 87.5% (n = 7/8) and 93.3% (n = 14/15), respectively. The respective reasons for the discontinuations were withdrawal of consent and premature termination of the study owing to an AE (upper respiratory tract infection) (Table S3 ). The median (min, max) ages of participants in the SSZ/GLPG3970 (N = 8) and MTX/GLPG3970 (N = 15) DDI studies were 28.0 (21, 55) years and 24.0 (18, 49) years, respectively. All participants in the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies were Black or African American men and White men, respectively. The demographic data in each study are summarized in Table 1 .

Table 1.

Demographic data (safety analysis sets)

SSZ/GLPG3970 DDI study
Demographic, median (min, max)	All participantsa (N = 8)
Age,b years	28.0 (21, 55)
Height, cm	177.0 (167, 184)
Weight, kg	80.5 (67.9, 98.5)
BMI, kg/m2	26.9 (22.2–29.4)

MTX/GLPG3970 DDI study
Demographic, median (min, max)	All participantsc (N = 15)
Age,b years	24.0 (18, 49)
Height, cm	181.0 (174, 192)
Weight, kg	78.5 (65.7, 100.5)
BMI, kg/m2	25.4 (20.3, 29.7)

BMI, body mass index; DDI, drug–drug interaction; MTX, methotrexate; SSZ, sulfasalazine.

^a All participants in the SSZ/GLPG3970 DDI study were Black or African American men.

^b Age at signing the informed consent form.

^c All participants in the MTX/GLPG3970 DDI study were White men.

PK analyses

The primary PK end points in the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies are summarized in Table 2 ; absolute PK parameters are provided in Table S4 . SSZ exposure (AUC_0–t, and AUC_0–∞, and C _max) increased by ~ 1.7–1.8‐fold when SSZ was administered simultaneously with GLPG3970 on day 5 and when SSZ was administered 2 hours before GLPG3970 on day 9, compared with administration of SSZ alone (day 1). SPD exposure was similar when SSZ was administered with GLPG3970 (both simultaneously and 2 hours before GLPG3970) or alone. The SPD:SSZ ratios (AUC_0–t and AUC_0–∞) decreased by approximately twofold following the administration of SSZ, either simultaneously with GLPG3970 or 2 hours before GLPG3970, compared with the administration of SSZ alone. The intra‐participant variability for the SPD:SSZ ratio was high (32%). The plasma concentration–time profiles of SSZ and SPD in the 72 hours following each SSZ administration are shown in Figure 2 . The mean t _1/2 was similar for SSZ and SPD and was independent of whether SSZ was administered alone, simultaneously with GLPG3970, or 2 hours before GLPG3970.

Table 2.

Primary PK parameters in the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies (PK analysis sets)

SSZ/GLPG3970 DDI study
	Geometric LS mean (95% CI)a			Geometric LS mean ratio, % (90% CI)b
	SSZ 1,000 mg  Day 1 (n = 8)	GLPG3970 350 mg + SSZ 1,000 mg Day 5 (n = 7)	GLPG3970 350 mg + SSZ 1,000 mg (−2 h)c Day 9 (n = 7)	Day 5 vs. day 1	Day 9 vs. day 1
SSZ PK parameters
AUC0–t (μg.h/mL)	59.3 (39.0, 90.2)	108 (70.2, 167)	109 (70.7, 167)	183 (143, 233)	184 (144, 234)
AUC0–∞ (μg.h/mL)	59.8 (39.4, 90.9)	109 (71.0, 168)	110 (71.5, 169)	182 (143, 232)	184 (144, 234)
C max (μg/mL)	7.27 (4.74, 11.2)	12.0 (7.65, 18.9)	13.1 (8.36, 20.6)	165 (114, 239)	181 (125, 261)
SPD PK parameters
AUC0–t (μg.h/mL)	119 (77.0, 184)	107 (69.1, 166)	106 (68.2, 164)	90 (81, 100)	89 (80, 99)
AUC0–∞ (μg.h/mL)	125 (78.5, 200)	113 (70.6, 181)	112 (69.9, 179)	90 (81, 101)	89 (80, 100)
C max (μg/mL)	5.68 (4.48, 7.20)	5.09 (3.99, 6.49)	5.31 (4.17, 6.78)	90 (78, 102)	94 (82, 107)
SPD:SSZ ratio
AUC0–t	3.21 (1.65, 6.27)	1.59 (0.81, 3.14)	1.56 (0.79, 3.09)	50 (37, 66)	49 (36, 65)
AUC0–∞	3.35 (1.67, 6.72)	1.66 (0.82, 3.37)	1.64 (0.81, 3.32)	50 (37, 67)	49 (36, 66)

MTX/GLPG3970 DDI study
	Geometric LS mean (95% CI)a		Geometric LS mean ratio, % (90% CI)b
	MTX 7.5 mg Day 1 (n = 15)	GLPG3970 350 mg + MTX 7.5 mg Day 5 (n = 14)	Day 5 vs. day 1
MTX parameters
AUC0–t (ng.h/mL)	627 (565, 697)	583 (523, 649)	93 (86, 101)
AUC0–∞ (ng.h/mL)d	677 (601, 762)	614 (543, 693)	91 (83, 99)
C max (ng/mL)	180 (159, 204)	193 (170, 220)	108 (96, 121)
7‐OH‐MTX parameters
AUC0–t (ng.h/mL)	490 (418, 576)	392 (332, 463)	80 (69, 92)
AUC0–∞ (ng.h/mL)e	NC	NC	NC
C max (ng/mL)	35.0 (31.1, 39.4)	27.0 (23.9, 30.4)	77 (72, 82)

7‐OH‐MTX, 7‐hydroxy‐methotrexate; AUC_0–∞, area under the plasma concentration–time curve from time zero until infinity; AUC_0–t, area under the plasma concentration–time curve from time zero until the last observed quantifiable concentration; AUC_ext, area under the plasma concentration–time curve extrapolated from last observed quantifiable concentration until infinity; CI, confidence interval; C _max, maximum observed plasma concentration; DDI, drug–drug interaction; LS, least‐squares; MTX, methotrexate; NC, not calculated; PK, pharmacokinetics; SPD, sulfapyridine; SSZ, sulfasalazine.

^a Point estimates and 95% CI from a mixed‐effects model on natural logarithm‐transformed values, with the treatment as a fixed effect and the participant as a random effect.

^b Point estimate and 90% CI of the geometric LS mean ratio.

^c GLPG3970 was administered 2 hours after SSZ on day 9.

^d n = 13 for MTX 7.5 mg and n = 12 for GLPG3970 350 mg + MTX 7.5 mg.

^e AUC_0–∞ could not be calculated for 7‐OH‐MTX owing to AUC_ext > 20%.

Figure 2.

Mean (± SD) plasma concentrations (log‐linear scale) of (a) SSZ and (b) SPD in the 72 hours following SSZ administration alone on day 1, simultaneous administration of SSZ with GLPG3970 on day 5, and SSZ administration 2 hours before GLPG3970 on day 9 (PK analysis set). ^aGLPG3970 was administered 2 hours after SSZ on day 9. PK, pharmacokinetic; SD, standard deviation; SPD, sulfapyridine; SSZ, sulfasalazine.

Administration of MTX with GLPG3970 did not have any relevant effect on MTX exposure (AUC_0–∞ and C _max) compared with administration of MTX alone (Table 2 ). After coadministration of MTX and GLPG3970, the exposure of 7‐OH‐MTX was slightly reduced compared with MTX administration alone (20% and 23% reductions in AUC_0–t and C _max, respectively; AUC_0–∞ could not be determined). The intra‐participant variability of MTX and 7‐OH‐MTX AUC_0–∞ and C _max was low, ranging from 10–22%. The plasma concentration–time profiles of MTX and 7‐OH‐MTX in the 24 hours following each MTX administration are shown in Figure 3 . The observed MTX t _1/2 (~ 3 hours) was at the shorter end of the expected range for MTX and was independent of coadministration with GLPG3970. MTX plasma concentrations fell below the LLOQ after 12 hours. The exposure of GLPG3970 in the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies was as expected based on previous studies involving GLPG3970 alone.

Figure 3.

Mean (± SD) plasma concentrations (log‐linear scale) of (a) MTX and (b) 7‐OH‐MTX in the 24 hours following MTX administration alone on day 1 and simultaneous administration of MTX and GLPG3970 on day 5 (PK analysis set). 7‐OH‐MTX, 7‐hydroxy‐methotrexate; MTX, methotrexate; PK, pharmacokinetic; SD, standard deviation.

Safety

A summary of TEAEs in the SSZ/GLPG3970 and MTX/GLPG3970 DDI studies is given in Table S3 . In the MTX/GLPG3970 DDI study, there was one TEAE of an upper respiratory tract infection that led to discontinuation of study treatment; the TEAE was determined by the investigator as not related to the study drug. Overall, GLPG3970 350 mg demonstrated an acceptable safety profile and was well tolerated both when administered alone and when coadministered with SSZ or MTX.

DISCUSSION

These SSZ/GLPG3970 and MTX/GLPG3970 DDI studies evaluated the effect of GLPG3970, a small‐molecule SIK2/3 inhibitor intended for the treatment of inflammatory diseases, on the PK of SSZ and SPD, and of MTX and 7‐OH‐MTX, respectively, in healthy adults. These are the first reported DDI evaluations of GLPG3970. The studies were carried out following the results of in vitro studies that indicated a strong inhibition of BCRP at clinically relevant concentrations of GLPG3970 (unpublished data). This is in turn suggestive of a potential interaction risk with the BCRP substrates SSZ and MTX, which are commonly used background medications for inflammatory diseases. ⁸ , ⁹

These DDI studies demonstrated that GLPG3970 has a mild inhibitory effect on intestinal BCRP in vivo, despite the strong in vitro signal that was previously observed. Coadministration of SSZ with GLPG3970 increased SSZ exposure by 1.7–1.8‐fold, whereas there was no relevant effect on SPD exposure. In turn, this decreased the SPD:SSZ ratio, a more sensitive indicator of intestinal BCRP modulation than SSZ exposure, by approximately twofold compared with SSZ administration alone. Delaying GLPG3970 administration by 2 hours did not alter the magnitude of the DDI effect.

In the MTX/GLPG3970 DDI study, there was no relevant effect of GLPG3970 on MTX exposure; the fold change in exposure was ≤ 1.25, which is indicative of a less than mild DDI effect, per guidelines by the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use. ²⁰ The exposure of 7‐OH‐MTX was slightly reduced (by 20–23%), but this was not considered clinically relevant because the changes were within the determined range of intra‐participant variability of exposure (10–22%). The mean t _1/2 was similar for MTX and 7‐OH‐MTX regardless of whether MTX was administered with or without GLPG3970, further supporting that the observed changes in 7‐OH‐MTX PK were not clinically relevant.

The reasons for the lack of concordance between the in vitro and in vivo DDI studies are not clear. Although in vitro studies allow for controlled targeted investigations and provide important insight into the specific relationships between transporters and investigated drugs, they do not take into account the complexities of a living organism, including distribution of the drug in relation to the transporter, competitive binding with endogenous molecules or compensatory mechanisms. In general, in vitro DDI assessments and their threshold tend to be conservative to prevent false negatives. In contrast, in vivo studies offer maximal physiological relevance, and are crucial for contextualizing the information obtained from in vitro studies; because of this, they are considered the definitive assessment. For example, although inhibition of a transporter by a drug is demonstrated through an in vitro assay, the in vivo relevance for a substrate is difficult to predict, as a drug is often a substrate of multiple transporters. BCRP has been shown to contain multiple binding sites. Therefore, there may be competition of the drug with endogenous components in vivo that were not present in the in vitro scenario. Another explanation is that in vitro assessments do not take into account compensatory pathways that may exist in an organism. Further work should focus on the reasons for the lack of correlation between in vivo and in vitro assessment of GLP3970‐mediated inhibition of BCRP so that in vitro DDI models can be refined. For example, the use of organoid models would help capture the influence of tissue structure cell heterogeneity and three‐dimensional cell organization, which may enable more sophisticated in vitro modeling of DDIs. Furthermore, the cutoff values used to determine whether in vitro inhibition of transporter could translate into clinical DDI (based on the FDA and EMA guidance), are based on a limited dataset, and retrospective analysis with all in vivo data generated could further be considered to refine these cutoff values. In both studies, GLPG3970 exposure upon simultaneous administration with SSZ or MTX was as expected based on previous studies in which GLPG3970 was administered alone. No safety concerns were found in either study and coadministration of GLPG3970 with SSZ or MTX was well tolerated.

Even though SSZ is a recommended probe substrate for the assessment of BCRP inhibition, and the design of the SSZ/GLPG3970 DDI study followed published recommendations, conducting such a clinical study in practice is not straightforward. Indeed, the present study is one of only a few published studies in which SSZ has been assessed in this way. Therefore, we consider that the SSZ/GLPG3970 DDI study represents a substantially important undertaking. We recommend that future SSZ DDI studies also use the immediate‐release formulation of SSZ, which is more sensitive to the DDI effect than the enteric‐coated formulation. ²¹ However, the immediate‐release formulation is available in the USA but not in Europe, which restricts potential study locations. Additionally, participant recruitment for such studies can be challenging owing to the BCRP and NAT2 genotyping inclusion requirements, which may necessitate long screening periods because of the time required for a genotyping laboratory to perform the relevant analyses. During the screening period of the SSZ/GLPG3970 DDI study in particular, potential participants were reluctant to wait for genotyping results and frequently started participating in another study, leaving them ineligible for participation in the present study. As a result, a possible limitation of the SSZ/GLPG3970 DDI study is the small number of participants recruited (N = 8), with four fewer participants than planned. Nonetheless, the data were considered sufficiently reliable to be able to draw conclusions. With regards to the MTX/GLPG3970 DDI study, it is worth emphasizing that the role of BCRP in MTX disposition, unlike SSZ disposition, is not fully established. In these DDI studies, we demonstrated that mild inhibition of BCRP by GLPG3970 did not affect MTX exposure.

In summary, GLPG3970 has a mild inhibitory effect on intestinal BCRP in vivo, as shown by the less than twofold increase in SSZ exposure, whereas the effect of GLPG3970 administration on MTX exposure was not clinically relevant. Therefore, the strong signal for GLPG3970 inhibition of BCRP previously observed in vitro was not confirmed in vivo in these clinical DDI studies. GLPG3970 had an acceptable positive safety profile and was well tolerated when administered alone and when coadministered with SSZ or MTX.

FUNDING

This study was funded by Galapagos NV (Mechelen, Belgium).

CONFLICTS OF INTEREST

M.P., F.G., and S.B. were employees of Galapagos NV at the time of the study. J.D. and V.Z. were employees of Galapagos SASU at the time of the study.

AUTHOR CONTRIBUTIONS

M.P., J.D., V.Z., F.G., and S.B. wrote the manuscript; M.P. and S.B. designed the research; V.Z. performed the research; M.P., J.D., V.Z., F.G., and S.B. analyzed the data.

Supporting information

Table S1‐S4.

DATA AVAILABILITY STATEMENT

Anonymized individual patient data will be shared upon request for research purposes dependent upon the nature of the request, the merit of the proposed research, and the availability of the data and its intended use. The full data sharing policy for Galapagos NV can be found at https://www.clinicaltrials‐glpg.com/us/en/data‐transparency.html.