Evaluation of Safety and Clinically Relevant Drug–Drug Interactions with Tucatinib in Healthy Volunteers

The study protocol and amendments were reviewed and approved by a central institutional review board. All participants provided written informed consent for participation and publication. This study was conducted in compliance with the principles of the Declaration of Helsinki and in accordance with the International Conference on Harmonisation Guidelines for Good Clinical Practice.

Healthy adults aged 18–65 years, with a body mass index of 18.0–32.0 kg/m², were included. Exclusion criteria included use of tobacco or other nicotine-containing products within 3 months prior to study initiation; consumption of alcohol within 48 h prior to study initiation and throughout the study; routine or chronic use of acetaminophen at a dose of > 3 g/day; and use of prescribed or over-the-counter medication, health supplements, or herbal remedies within 28 days prior to study initiation through to follow-up.

This phase I, multicenter, open-label, fixed-sequence study (ClinicalTrials.gov identifier: NCT03723395) was conducted in five parts (A–E; Fig. 1). Participants were confined to the clinical research center on the day before initiating treatment until the day of clinic discharge.

Part A evaluated the effect of the strong CYP3A4 inhibitor, itraconazole, on the single-dose pharmacokinetics of tucatinib. Participants received a single 300 mg dose of tucatinib administered orally 2 h after breakfast on days 1 and 6. A 200 mg dose of itraconazole was administered orally twice daily immediately after a meal on day 3 and once daily immediately after breakfast on days 4–7 (Fig. 1A). Participants were discharged from the clinic on day 8 and were assessed at a follow-up visit on day 12 (±1 day).

Part B evaluated the effect of rifampin, a strong inducer of CYP3A4 and CYP2C8, on the single-dose pharmacokinetics of tucatinib. Participants received a single, 300 mg dose of tucatinib administered orally on days 1 and 10, following an overnight fast of at least 8 h. A 600 mg dose of rifampin was administered orally once daily on days 3–11, following an overnight fast of at least 8 h (Fig. 1A). Participants were discharged from the clinic on day 12 and were assessed at a follow-up visit on day 16 (± 1 day).

Part C evaluated the effect of the strong CYP2C8 inhibitor gemfibrozil on the single-dose pharmacokinetics of tucatinib. Participants received a single 300 mg dose of tucatinib administered orally on days 1 and 7, following an overnight fast of at least 8 h. A 600 mg dose of gemfibrozil was administered orally twice daily on days 3–8, following an overnight fast of at least 8 h in the morning and approximately 30 minutes prior to the evening meal (Fig. 1A). Participants were discharged from the clinic on day 9 and were assessed at a follow-up visit on day 13 (± 1 day).

Part D evaluated the effects of steady-state tucatinib on the pharmacokinetics of substrate probes of CYP2C8 (repaglinide), CYP2C9 (tolbutamide), and CYP3A4 (midazolam). Participants received a single 0.5 mg dose of repaglinide administered orally on days 1 and 11, following an overnight fast. A 500 mg dose of tolbutamide and a 2 mg dose of midazolam were co-administered orally on days 2 and 12, following an overnight fast. A 300 mg dose of tucatinib was administered orally twice daily, at least 12 h apart, on days 4–13. On days 4, 10, 11, and 12, the morning dose was given after an overnight fast of at least 8 h. On days 11 and 12, the morning dose was given immediately after administration of repaglinide and tolbutamide/midazolam, respectively (Fig. 1B). Participants were discharged from the clinic on day 14 and were assessed at a follow-up visit on day 21 (±1 day).

Part E evaluated the effect of steady-state tucatinib on the pharmacokinetics of a substrate probe of P-gp (digoxin). Participants received a single, 0.5 mg dose of digoxin administered orally on days 1 and 15, following an overnight fast of at least 8 h. Tucatinib 300 mg was administered orally twice daily, at least 12 h apart, on days 8–21. On days 8, 14, and 15, the morning dose was given after an overnight fast of at least 8 h. On day 15, the morning dose was administered immediately after digoxin (Fig. 1B). Participants were discharged from the clinic on day 22 and were assessed at a follow-up visit on day 29 (± 1 day).

Foods and beverages containing poppy seeds, grapefruit, or Seville oranges were not allowed from 7 days prior to day −1 until the follow-up visit or early study termination. While confined to the study site, participants received a standardized diet at scheduled times that did not conflict with other study-related activities. All oral study drugs were administered with 240 mL of room-temperature water. When drugs were administered concurrently, only 240 mL of water was administered in total for all drugs. Participants were dosed upright in a seated position.

Blood samples for tucatinib and probe drug pharmacokinetics were collected pre-dose and at intervals as indicated in Supplementary Fig. S1 for parts A–E. In parts A–C, blood samples for determination of plasma concentrations of tucatinib and ONT-993 were collected at 0.5, 1, 2, 3, 4, 6, 8, 10, 12, 16, 24, 36, and 48 h post-dose on days 1 and 6 (Part A) or day 10 (Part B) or day 7 (Part C). In Parts B and C, the 48-h post-dose timepoint relevant to day 1 dosing of tucatinib was collected prior to dosing on day 3 of rifampin (Part B) or gemfibrozil (Part C). A single blood sample was collected prior to dosing for determination of trough plasma concentrations on days 4–6 for itraconazole dosing in Part A, on days 8–10 for rifampin dosing in Part B, and on days 5–7 for gemfibrozil in Part C.

In Part D, blood samples for determination of plasma concentrations of repaglinide were collected at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, and 12 h post-dose on days 1 and 11. Blood samples for determination of plasma concentrations of midazolam and 1-hydroxymidazolam were collected at 0.5, 1, 2, 3, 4, 6, 8, 10, 12, 16, and 24 h post-dose on days 2 and 12. Blood samples for tolbutamide and 4-hydroxytolbutamide were collected at the same timepoints and also at 36 and 48 h post-dose. The 48-h post-dose timepoint relative to day 2 dosing was collected prior to the morning dose of tucatinib on day 4. In Part E, blood samples for determination of plasma concentrations of digoxin were collected at 0.5, 1, 2, 3, 4, 6, 8, 10, 12, 16, 24, 48, 72, 96, 120, 144, and 168 h post-dose on days 1 and 15. In Parts D and E, blood samples for determination of plasma concentrations of tucatinib and ONT-993 were collected at 0.5, 1, 2, 3, 4, 6, 8, 10, and 12 h after the morning dose of tucatinib on days 4, 10, 11, and 12 (Part D) or days 8, 14, and 15 (Part E). The 12-h post-dose samples were collected prior to administration of the second daily dose of tucatinib.

Plasma concentrations of tucatinib, ONT-993, itraconazole, rifampin, gemfibrozil, repaglinide, tolbutamide, 4­hydroxytolbutamide, midazolam, 1-hydroxymidazolam, and digoxin were determined using validated bioanalytical procedures, performed by Covance (Madison, WI, USA).

Protein precipitation was used for extraction of tucatinib, ONT-993, itraconazole, gemfibrozil, rifampin, tolbutamide, and 4-hydroxytolbutamide; liquid-liquid extraction for midazolam, 1-hydroxymidazolam, and digoxin; and supported-liquid extraction for repaglinide. Analyte detection and quantification was carried out by liquid chromatography with tandem mass spectrometry (LC-MS/MS) as described in the Data Supplement. Details of the reference analytes and internal standards are given in Supplementary Table S1 (see electronic supplementary material [ESM]).

Blood and urine samples were collected, after at least an 8-h fast, for clinical laboratory evaluations at specific times during the study for Parts A–E. Clinical chemistry, hematology, and urinalysis evaluations were performed at screening, check-in, and follow-up, and regularly at protocol-specified timepoints in all parts of the study. Assays for cystatin C and urine albumin-to-creatinine ratio were performed at all these timepoints except screening in Part D, and for all but the first six patients enrolled in Part E, following a protocol amendment.

Safety assessments included recording of all treatment-emergent adverse events (TEAEs), clinical laboratory parameters, vital signs, 12-lead electrocardiogram, and physical examination.

Pharmacokinetic analysis was conducted by Covance Early Clinical Biometrics using WinNonlin Version 8.1 (Certara L.P., Princeton, NJ, USA). Pharmacokinetic parameters were determined, as appropriate, from the plasma concentrations of tucatinib, concomitant drugs, and metabolites using non-compartmental methods.

The pharmacokinetic parameters were area under the plasma concentration–time curve from time 0 to the last available measurement (AUC_0–last), area under the plasma concentration–time curve from time 0 to infinity (AUC_0–inf), and maximum observed plasma concentration (C_max) for tucatinib, repaglinide, tolbutamide, midazolam, and digoxin. A linear mixed-model analysis was applied to analyze the natural log-transformed pharmacokinetic parameters, which included treatment as a fixed effect and subject as a random effect. The impact of drug co-administration on the pharmacokinetic parameters for each part of the study was assessed by deriving estimates of geometric mean ratios, together with the corresponding 90% confidence intervals (CIs), for comparison between reference and test timepoints (days 1 and 6 in Part A, days 1 and 10 in Part B, days 1 and 7 in Part C, days 1 and 11 for repaglinide and days 2 and 12 for tolbutamide and midazolam in Part D, and days 1 and 15 for digoxin in Part E). Exploratory analyses of the geometric mean ratios and 90% CIs of AUC_0–inf, AUC_0–last, and C_max for metabolites ONT-993, 4-hydroxytolbutamide, and 1-hydroxymidazolam were evaluated as required.

Other pharmacokinetic parameters estimated by noncompartmental analysis included time to C_max, terminal elimination half-life, apparent total clearance (tucatinib, repaglinide, tolbutamide, midazolam, and digoxin), apparent total clearance at steady state (tucatinib in Parts D and E only), metabolic ratio based on AUC_0–inf, and metabolic ratio based on C_max.

Parts A, B, and C each enrolled 28 participants, all of whom were evaluable for safety and pharmacokinetics. In Part D, all 17 participants enrolled were evaluable for safety and pharmacokinetics. In Part E, of the 15 participants enrolled, 2 were withdrawn from the study on day 8 due to non-treatment-related upper respiratory infections and did not receive tucatinib, leaving 13 participants evaluable for safety and pharmacokinetics. Participant demographics are summarized in Supplementary Table S2 (see ESM).

The CYP3A4 inhibitor itraconazole, in combination with tucatinib, increased the plasma exposure of tucatinib approximately 1.3-fold based on AUC_0–last, AUC_0–inf, and C_max compared with tucatinib alone (Fig. 2, Table 1). This effect was statistically significant based on 90% CI values. Consistent with this, mean clearance was lower and mean terminal elimination half-life was longer for tucatinib in combination with itraconazole than for tucatinib alone (Supplementary Table S3, see ESM). Plasma exposure of ONT-993, indicated by the geometric least squares mean ratios for AUC_0–inf and C_max, also was increased by coadministration of itraconazole (1.6 and 2.0, respectively). However, this is most likely due to the increased plasma exposure of tucatinib in the presence of itraconazole. The metabolic ratios for ONT-993 were similar or slightly higher in the presence of itraconazole as compared with tucatinib alone (Supplementary Table S3, see ESM).

The strong CYP3A4 inducer and moderate CYP2C8 inducer rifampin, in combination with tucatinib, decreased the plasma exposure of tucatinib approximately 48% based on AUC_0–last and AUC_0–inf, and 37% based on C_max, compared with tucatinib alone (Fig. 2, Table 1). This effect was statistically significant, based on 90% CI values. Consistent with this, mean clearance for tucatinib was increased in combination with rifampin as compared with tucatinib alone, while terminal half-life was similar (Supplementary Table S4, see ESM). Co-administration of rifampin decreased the plasma exposure and increased the C_max of ONT-993 (Supplementary Table S4, see ESM). The metabolic ratios for AUC_0–inf and C_max of ONT-993 were 1.4-fold and 3.3-fold higher, respectively, in the presence of rifampin than with tucatinib alone (Supplementary Table S4, see ESM).

The strong CYP2C8 inhibitor gemfibrozil, in combination with tucatinib, increased plasma exposure of tucatinib approximately 3.0-fold based on AUC_0–last and AUC_0–inf, and 1.6-fold based on C_max, compared with administration of tucatinib alone (Fig. 2, Table 1). This effect was statistically significant, based on 90% CI values. Coadministration of gemfibrozil decreased the plasma exposure of ONT-993, with a statistically significant effect on C_max (Supplementary Table S5, see ESM). The metabolic ratios for AUC_0–inf and C_max of ONT-993 were decreased by approximately 71% and 81%, respectively, following administration of gemfibrozil, compared with tucatinib alone (Supplementary Table S5, see ESM).

Administration of tucatinib in combination with tolbutamide/midazolam significantly increased the plasma exposure of the CYP3A4 substrate, midazolam, 5.3-, 5.7-, and 3.0-fold based on AUC_0–last, AUC_0–inf, and C_max, respectively, compared with administration of tolbutamide/midazolam alone (Fig. 3, Table 2). The lower mean clearance and longer mean terminal half-life of midazolam in the presence of tucatinib, as compared with tolbutamide/midazolam administered alone, are consistent with this effect (Supplementary Table S6, see ESM). Although tucatinib had no effect on the exposure of the metabolite 1-hydroxymidazolam, based on AUC_0–last and AUC_0–inf, C_max was reduced by approximately 40%. Tucatinib had a strong effect on the metabolism of midazolam, reducing the metabolic ratios for AUC_0–inf and C_max by approximately 83% and 80%, respectively (Supplementary Table S7, see ESM).

Administration of tucatinib in combination with the CYP2C8 substrate repaglinide increased the plasma exposure of repaglinide approximately 1.6- and 1.7-fold, based on AUC_0–inf and C_max, respectively, compared with administration of repaglinide alone (Fig. 3, Table 2). Consistent with this, mean clearance of repaglinide was lower in the presence of tucatinib than when administered alone (Supplementary Table S7, see ESM).

The plasma exposures of the CYP2C9 substrate tolbutamide were similar after administration of tolbutamide/midazolam in combination with tucatinib and after administration of tolbutamide/midazolam alone (Fig. 3, Table 2). Similar plasma exposures were also observed for the metabolite 4-hydroxytolbutamide before and after tucatinib coadministration (Supplementary Table S8, see ESM). The metabolic ratios for AUC_0–inf and C_max of 4­hydroxytolbutamide were similar before and after tucatinib coadministration (Supplementary Table S8, see ESM).

Administration of tucatinib in combination with the P-gp substrate digoxin increased plasma exposure of digoxin approximately 1.5-fold based on AUC_0–last and AUC_0–inf, and 2.4-fold based on C_max, compared with administration of digoxin alone (Fig. 3, Table 2). Consistent with this, the mean clearance of digoxin was lower in the presence of tucatinib than when administered alone (Supplementary Table S9, see ESM).

Tucatinib was well tolerated in Parts A–E of the study, either after single 300 mg doses alone or when administered in combination with other drugs, or after multiple days of twice-daily dosing either alone or in combination with other drugs. Overall, in Parts A through E, 37 of 114 (32.5%) subjects reported a total of 71 TEAEs. The majority of TEAEs were mild (65 events in total were considered Grade 1) and considered unrelated to tucatinib. No Grade 3 or higher events were reported. There were no deaths or serious adverse events, and all but two TEAEs resolved by end of study. The two TEAEs that had not resolved by end of study were considered unrelated to tucatinib.

TEAEs considered related to tucatinib (Supplementary Table S10, see ESM) were consistent with its known safety profile. Increases in serum creatinine are recognized to occur with tucatinib due to inhibition of renal tubular transport of creatinine without affecting glomerular function [11]. These were observed after dosing of tucatinib alone or in combination with probe drugs, both in the single-dose and the multiple-dose parts of the study. Serum creatinine increases were observed in 14 (50%) participants in Part A, none of which were reported as TEAEs; in one (4%) participant in Part B, which was recorded as a Grade 1 TEAE; in 19 (68%) participants in Part C, none of which were TEAEs; in 12 (71%) participants in Part D, one of which was a Grade 2 TEAE; and in seven (54%) participants in Part E, five of which were Grade 1 TEAEs (Supplementary Table S10, see ESM). Most of these elevations were transient and resolved after cessation of tucatinib dosing. Assessment of renal function by cystatin C levels in Parts D and E did not reveal any elevated values, and no out-of-range urine albumin or urine albumin-to-creatinine ratios were observed, suggesting no corresponding effect on glomerular filtration rate or overall renal function.

Elevations in hepatic transaminases (alanine aminotransferase [ALT] and/or aspartate aminotransferase [AST]), which have been reported previously in subjects receiving tucatinib [4, 12‐15], were noted, especially after multiple dosing of tucatinib in Parts D and E. Overall, five events of ALT elevation and two events of AST elevation were reported as Grade 1 TEAEs during the study in Parts D and E (Supplementary Table S10, see ESM). These were generally transient, reversible, and resolved after cessation of tucatinib dosing. Other tucatinib-related adverse events were limited in Parts A–E of the study and included one participant with diarrhea each in Part A and Part E, one participant with pruritus each in Part D and Part E, one participant with dry skin each in Part B and Part E, one participant with headache each in Part C and Part E, one event of blood bilirubin increase in Part B, one participant with dizziness in part E, and one participant with gout and arthralgia in Part C.