TARGET-EU: Tolvaptan and risk associated to hepatotoxicity in autosomal dominant polycystic kidney disease

Study type

• Study topic:
  Disease /health condition

  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Safety study (incl. comparative)
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

  Other
• Non-interventional study design, other:
  This study will emulate a hypothetical target trial using a retrospective, prevalent new-user cohort design based on routinely collected electronic health records.

Study drug and medical condition

• Medicinal product name:
  TOLVAPTAN ACCORD
• Study drug International non-proprietary name (INN) or common name:
  TOLVAPTAN
• Anatomical Therapeutic Chemical (ATC) code:
  (C03XA01) tolvaptan

  tolvaptan
• Medical condition to be studied:
  Hepatotoxicity
• Additional medical condition(s):
  Autosomal dominant polycystic kidney disease

Population studied

• Short description of the study population:
  The study population will consist of adults with ADPKD, without use of medication with hepatotoxic effects, history of hepatotoxicity, liver enzyme elevation or use of diuretics. ; ; For the intervention group, time 0 is defined as the first prescription of tolvaptan. This is when participants enter the study, which mimics initiation of tolvaptan therapy in the target randomised controlled trial. For participants in the control group, time 0 is the date of a physician visit without a tolvaptan prescription, matched on time since ADPKD diagnosis to the date of a treatment receipt in the intervention group. Additionally, the control and intervention time 0 are also matched on the calendar time scale. The matching will not be on the exact value but on the quintile. In the first stage of matching, on calendar time, matching will be done 5 : 1 for control group: intervention group. This is necessary to end up with a 1:1 control: intervention group allocation for the second matching step on propensity score. Individuals can only be eligible as matched control once and can only be assigned to either the control group or the intervention group once (at time 0).
• Age groups:
  • Adults (18 to < 65 years)

    • Adults (18 to < 46 years)
    • Adults (46 to < 65 years)
  • Elderly (≥ 65 years)

    • Adults (65 to < 75 years)
    • Adults (75 to < 85 years)
    • Adults (85 years and over)

Study design details

• Study design:
  A pragmatic design with real-world data captures care as it is delivered in practice. Although the observational emulation cannot perfectly mimic randomisation in the target trial, several measures are taken to minimise bias and confounding. The prevalent new-user design avoids immortal time bias an
• Main study objective:
  To investigate the safety of tolvaptan use through assessing the risk of hepatotoxicity associated with tolvaptan treatment in ADPKD patients compared to the risk in patients who are unexposed to tolvaptan.
• Setting:
  This study is conducted using routinely collected electronic health records from 2015 to 2024, reflecting the period of tolvaptan use in routine clinical practice. The study is set primarily in primary care, drawing on longitudinal data from general practices with linkage to hospital data. Data are sourced from the United Kingdom (Clinical Practice Research Datalink [CPRD]), providing population-based and representative coverage of real-world clinical care.
• Comparators:
  There is no alternative treatment to tolvaptan to which the risk of hepatotoxicity could be compared. Clinically it is most relevant to compare tolvaptan use to ADPKD patients unexposed to tolvaptan, who may receive symptomatic treatment, as per clinical practice. Any dose or regimen of tolvaptan is eligible for inclusion. Comparators will thus be adults with ADPKD who do not use tolvaptan, without use of medication with hepatotoxic effects, history of hepatotoxicity, liver enzyme elevation or use of diuretics.
• Outcomes:
  Clinical trials have suggested that there is an association between tolvaptan treatment and elevations in liver enzyme levels (alanine aminotransferase (ALT) and aspartate aminotransferase (AST)) [2,6]. In these studies, hepatotoxicity is defined as increases in ALT or AST levels to more than 3 times the upper limit of normal (ULN) [2,6]. It is expected that, at least in the unexposed group, liver enzyme levels are not registered regularly. In this emulation, hepatotoxicity will be defined as the registration of ICD-10, Read, or SNOMED codes related to hepatotoxicity or increases in the following liver enzymes if data are available: ; ; ALT > 3x ULN ; ; AST > 3x ULN ; ; Total bilirubin > 2x ULN ; ; Alkaline phosphatase > 2x ULN + elevated GGT and/or 5-NT
• Data analysis plan:
  The analyses are conducted within a target trial emulation framework to estimate the effect of tolvaptan exposure compared with non exposure on the risk of hepatotoxicity. ; ; For Estimand 1, the main estimand supporting decision making, the primary causal effect summary measure is the hazard ratio for time to first hepatotoxicity event, estimated using a stratified Cox proportional hazards model in the propensity score-matched sample. ; ; Sensitivity analyses will assess robustness of the primary findings to key assumptions, including inverse probability of censoring weighting (IPCW) (assuming censoring is independent of the outcome, with all common causes of both the outcome and censoring being accounted for conditional on the treatment, time-varying covariates and survival up to the time of censoring as well as indirectly baseline covariates used to estimate the treatment weights), best/worst case scenario analysis, and outcome misclassification (details in Section 7.6.5). ; ; Two supplemental estimands are also defined: Estimand 2, applying a while-on-treatment strategy for intercurrent events, and Estimand 3, estimating treatment effects using restricted mean survival time (RMST) derived from a Propensity score matched Weibull accelerated failure time (AFT) model. In addition, supplemental analyses (e.g., crude and PS-matching adjusted Kaplan–Meier curves, crude Cox models, event counts and incidence rates, propensity score and weight distributions, covariate balance before and after weighting, censoring and intercurrent event patterns, proportional hazards diagnostics, positivity checks, and multiple-imputation diagnostics) will be conducted to support interpretation of the main analysis. ; ; For all estimands, conditional propensity scores will be calculated, and patients will be matched accordingly. For estimand 1 and 2, the Cox model will be used to calculate the HR of hepatotoxicity of tolvaptan users vs. nonusers.
• Summary results:
  Not yet available.