Authors’ Reply to “Methodological and Outcome Interpretation Concerns for the Food-Effect Study of Bersiporocin” [Response to Letter]

Dear editor

We thank Shen and Xu for their interest in our recent publication regarding the food effect on the pharmacokinetics and gastrointestinal tolerability of bersiporocin. We appreciate the opportunity to address their concerns and clarify our interpretation of the clinical findings.

Interpretation of the C_max Reduction

We understand the correspondents’ concern that our statement that food did not have a clinically significant effect on systemic exposure could, without further clarification, be interpreted as applying to C_max as well as AUC. To clarify, this conclusion referred specifically to the preservation of overall systemic exposure, as assessed by AUC_last, not to the absence of any food-related pharmacokinetic change. The 33–48% reduction in C_max and the prolongation of T_max were explicitly reported and discussed in our manuscript as evidence that food altered the rate, rather than the extent, of absorption.1 Although both AUC and C_max are relevant to characterizing a food effect, their clinical significance should be interpreted in the context of the drug’s exposure–response relationship.2 In the prior multiple-ascending-dose study of bersiporocin, AUC was explicitly used to inform dose selection based on the predicted biologically active exposure in a bleomycin-induced pulmonary fibrosis model.3 Mechanistically, a cumulative, time-averaged parameter such as AUC would be expected to more closely reflect the sustained PRS inhibition required to suppress ongoing collagen synthesis than would an instantaneous peak concentration. In addition, based on published nonclinical data, the estimated unbound C_max of bersiporocin remained above its in vitro PRS IC50 of 78 nM across all three prandial conditions, providing additional pharmacological context for interpreting the observed reduction in C_max.4 We acknowledge that a formal exposure–response analysis directly comparing AUC- and C_max-based models has not been reported for this agent; nevertheless, the preserved AUC across prandial conditions and the maintained margin above the in vitro inhibitory concentration provide some reassurance that antifibrotic activity is unlikely to be meaningfully compromised. The altered absorption rate nonetheless remains clinically relevant to dosing and gastrointestinal tolerability, an issue our dosing recommendation directly addresses.

Potential Role of Local Gastric Exposure

The correspondents correctly note that the delayed T_max is most readily explained by delayed gastric emptying and prolonged gastric residence of the formulation. This does not, however, preclude the simultaneous occurrence of local gastric irritation. A high-fat meal can transiently increase intragastric pH and delay gastric emptying, thereby prolonging gastric residence time.5–7 Under these conditions, a limited local loss of enteric-coating integrity could hypothetically permit a small fraction of the dose to contact the gastric mucosa without materially affecting the overall systemic absorption profile. We acknowledge that our study did not directly assess intragastric pH, coating integrity, or the site of drug release; therefore, premature or partial disruption of the enteric coating should be regarded as a hypothesis rather than an established mechanism. The occurrence of vomiting before T_max is compatible with, but does not establish, a peripheral mechanism. Formulation-specific dissolution testing or intragastric imaging would be required to evaluate this hypothesis.

Peripheral versus Central Mechanisms of Nausea and Vomiting

The correspondents suggest that our interpretation involves circular reasoning, dismissing a central mechanism for nausea and vomiting without analyzing C_max in relation to gastrointestinal adverse events (GI AEs). We would like to clarify that our original manuscript did in fact compare both AUC_last and C_max between participants with and without vomiting (Figure S1) and found no significant difference in either parameter. In addition, we reported the mean onset times of nausea and vomiting under fasted, 30-minute postprandial, and 2-hour postprandial conditions (1.77, 4.60, and 5.15 hours, respectively), noting that all vomiting events occurred before the corresponding median T_max. As a complementary, treatment-level observation, we note that if GI AEs were primarily driven by high peak plasma concentrations, the highest incidence would be expected in the fasted condition (Treatment A), which had the highest C_max (358.57 μg/L); instead, the highest incidence of GI AEs (45.7%) occurred in the 30-minute postprandial condition (Treatment B), which had the lowest C_max (197.12 μg/L).1 Although we did not formally analyze the early rate of concentration rise, the lack of a detectable difference in C_max between participants with and without vomiting, together with the inverse treatment-level pattern, does not support a simple C_max-driven explanation for the observed GI AEs.

Open-Label Study Design

The open-label design is unlikely to compromise the primary pharmacokinetic comparison, although it may influence the reporting of subjective tolerability outcomes. Food conditions cannot readily be blinded, and current regulatory guidance recommends a randomized, balanced, single-dose crossover design for a definitive food-effect study.2 Our randomized within-participant crossover design minimizes between-subject variability in pharmacokinetic and tolerability assessments, and the principal pharmacokinetic endpoints were objective. We acknowledge that subjective symptoms such as nausea can be influenced by expectations. However, the observed difference was also apparent for vomiting, which occurred in 31.4% of participants 30 minutes after the high-fat meal, compared with 5.7% under fasting and 9.1% 2 hours after the meal.1 Future studies could strengthen symptom assessment through prespecified symptom scales and standardized electronic diaries.

Single-Dose Design and Repeated-Dose Tolerability

We had already identified the single-dose design as a limitation of our study. In the previous multiple-ascending-dose study, nausea was reported in 83.3% of participants receiving bersiporocin 200 mg twice daily for 14 days, and gastrointestinal disorders were the predominant adverse drug reactions across the active-treatment cohorts.3 Although cross-study comparisons should be interpreted cautiously, these findings suggest that single-dose data alone are insufficient to characterize repeated-dose gastrointestinal tolerability. We note, however, that the IQ analysis cited by the correspondents concerns whether food-effect studies must be repeated across formulation or development changes, not whether such studies should incorporate blinded, repeated-dose safety designs.8 Given the higher incidence of GI AEs with repeated dosing, our recommendation should be regarded as a pragmatic starting point for subsequent clinical studies, with optimal dosing conditions to be confirmed in dedicated multiple-dose studies.

We thank Shen and Xu for their thoughtful comments and hope that our response clarifies the interpretation of the pharmacokinetic and safety findings of our study.

Funding Statement

No additional funding was received for the preparation of this response. The original study was sponsored by Daewoong Pharmaceutical Co., Ltd., Seoul, Republic of Korea.

Disclosure

YeaRa Kwak is an employee of Daewoong Pharmaceutical Co., Ltd. The authors report no other conflicts of interest in this communication.