This phase 1 (NCT05051553), open-label, 2-part crossover study in healthy adult participants was conducted in compliance with the International Council on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use/Good Clinical Practice and applicable regulatory requirements. The protocol complied with the Declaration of Helsinki as well as applicable guidelines of the USA, the country where the study was conducted. The protocol was submitted to an independent review board (Advarra, Columbia, MD, USA) for review and written approval. This study was run by Syneos Health (Miami, FL, USA). Written informed consent was obtained from all participants at screening, prior to the conduct of any study-related procedures.

Healthy adults aged 18–65 years, with a body mass index 18–33 kg/m2, were eligible to be enrolled in either part of the study. Participants were confirmed to be healthy based on normal/clinically acceptable vital signs, laboratory results, and electrocardiograms at screening. People of childbearing potential were neither pregnant nor lactating, and all participants were required to use highly effective contraceptive measures until 30 days after the last dose of the study drug. Participants with a history of Wernicke’s encephalopathy, thiamine deficiency, hypersensitivity to ondansetron or any components of study drug, or contraindications for insertion of a nasogastric tube were excluded from the study.

This study comprised 2 parts in a randomized, multiple-sequence design. The study parts were planned to run in any order, or in parallel. Participants could only join 1 part of the study. Each study part was composed of a screening, a treatment phase, and a follow-up phone call (approximately 4 days [± 2 days] after discharge).

Study part 1 was a 2-period, 2-sequence, open-label, crossover design (Fig. 1A). Treatment 1A (reference) consisted of fedratinib 400 mg (4 × 100-mg capsules) administered orally along with approximately 180 mL of a commercially available nutritional supplement (ie, Ensure Plus). Treatment 1B (test) consisted of fedratinib 400 mg dispersed in a nutritional supplement and administered orally. Participants were randomized to receive treatment in one of the following sequences: 1A–1B or 1B–1A. All participants underwent a supervised overnight fast of ≥ 10 h the night before fedratinib dosing. No food or beverages (except water and any nutritional supplement administered with the treatment) were allowed for at least 4 h after dosing.

Study design for study part 1 (A) and part 2 (B). BID twice daily; ET early termination

Study part 2 was a 3-period, 6-sequence, open-label, crossover design (Fig. 1B). Treatment 2A (reference) consisted of fedratinib 400 mg (4 × 100-mg capsules) administered orally alongside a nutritional supplement. Treatment 2B (test) consisted of fedratinib 400 mg dispersed in the nutritional supplement (180 mL, Ensure Plus), administered via nasogastric tube, and flushed with 60 mL of sterile water. Treatment 2C (test) consisted of fedratinib 400 mg as a divided dose (2 × 100-mg capsules BID) administered alongside the nutritional supplement. Participants were randomized to receive treatment in one of the following sequences: 2A–2B–2C, 2A–2C–2B, 2B–2A–2C, 2B–2C–2A, 2C–2A–2B, or 2C–2B–2A. As in study part 1, all participants underwent a supervised overnight fast of ≥ 10 h the night before dosing. In the case of an evening (second) dose in the divided dosing plan of treatment 2C, participants were directed to pre-fast at least 2 h before fedratinib dose, and 4 h after fedratinib dose, with water restricted from approximately 2 h before and 1 h after dosing, excluding water given with the treatment.

Approximately 1 h before each fedratinib administration, ondansetron oral tablet (film coated) 8 mg was given orally to reduce the potential for fedratinib-related nausea and vomiting [16]. There was a washout period of at least 12 days between each dose of fedratinib. The duration from screening through the follow-up phone call was approximately 10 weeks. Participants resided at the clinical site from days -1 to 23 (study part 1), or days -1 to 35 (study part 2). If a participant discontinued from the study for any reason, an early termination visit was performed, with only the safety assessments scheduled for the day of discharge taking place. Additionally, each participant who discontinued also received a follow-up phone call 4 (± 2) days after completion of the early termination visit.

For study part 1, and part 2 treatments 2A and 2B, blood samples were collected pre-dose (0 h) and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 48, 72, 120, 168, 192, and 240 h post-dose. For treatment 2C, blood samples were collected pre-dose (0 h) and at 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 12.5, 13, 13.5, 14, 15, 16, 18, 24, 48, 72, 120, 168, 192, and 240 h post-dose of the first dose. The plasma concentration of fedratinib was measured using a validated liquid chromatography-tandem mass spectrometry assay with a lower limit of quantification of 1.00 ng/mL [12, 20].

Plasma PK parameters were calculated using noncompartmental methods by a validated PK analysis program (Phoenix® WinNonlin®, v8.3.4 [Certara USA, Inc., Princeton, NJ, USA]) using actual times. The PK parameters determined for fedratinib Cmax, Tmax, AUC from time 0 to the time of the last quantifiable concentration (AUC0–t), AUC from time 0 to infinity (AUC0–∞), and t½. Additionally for treatment 2C, Cmax and Tmax were estimated after both the first dose (Cmax and Tmax) and the second dose (Cmax2 and Tmax2). AUC0–t and AUC0–∞ were estimated from time 0 (the first dose) through 240 h after the first dose. AUC parameters were calculated using the linear-log trapezoidal method.

The safety population included all participants who received at least 1 dose of fedratinib. Participants were monitored for adverse events (AEs) during the study, including assessments for clinical symptoms, biochemical, hematologic, and urinalysis laboratory results, physical examinations, 12-lead electrocardiograms, and vital signs. AEs were assessed for seriousness, severity, and relationship to the drug. AEs were assigned to the last study treatment administered at the time of onset, starting during or after the administration of the first dose. AEs that occurred after discharge were assigned to the last study treatment received for up to 30 days after the last dose of fedratinib.

As an exploratory endpoint, participants in part 1 were asked to answer a questionnaire regarding taste and palatability on days 1 and 13 of the study. The questionnaire assessed participant perceptions of bitterness, astringency, and sandiness/grittiness at 2, 5, and 15 min after administration, other taste attributes, and impulse to chew or swallow.

The sample size for this study was based on consideration of the precision in the comparison of PK parameters, represented by the 90% confidence interval (CI) of the geometric mean ratios of Cmax and AUC of fedratinib. The evaluable PK population, used for PK summary and statistical analyses, was defined as all participants who received at least 1 dose of fedratinib and had at least 1 evaluable PK parameter. Only participants with valid PK data were included in the summary statistics and statistical analysis.

To compare the PK parameters and estimated relative bioavailability of fedratinib following different administration methods in each study part (treatment 1B to 1A, and treatment 2B and 2C to 2A), a linear mixed-effect model with treatment, period, and sequence as fixed effects and participant nested within sequence as a random effect was fitted to the natural log-transformed PK parameters (Cmax, AUC0–t, and AUC0–∞) for use in estimation of effects and construction of CIs.

In study parts 1 and 2, point estimates and 90% CI for the difference between test and reference treatments was exponentiated, and the ratio of geometric means (test/reference) and associated 90% CI was presented on the original scale. For comparison of treatment 2C to 2A in study part 2, the relative bioavailability assessment was based on only AUC. For Tmax, the difference between test and reference was assessed for significance using Wilcoxon signed-rank test. Kenward–Rogers degrees of freedom were specified in the linear mixed-effect model. The Hodges–Lehmann estimate and its 90% CI were calculated for median difference between treatments.

All statistical analyses were conducted using SAS v9.4 (SAS Institute Inc., Cary, NC, USA).