It is not surprising that new drug development and conducting clinical trials have been continuous challenges for primary sclerosing cholangitis (PSC) for many reasons. Lower incidence, slower progression of the disease, and poorly understood pathogenesis, including genetic predisposition, are among the top reasons. The AASLD in 2023 recommended including all patients with PSC in clinical trials if possible due to these challenges [
1].
Ursodeoxycholic acid (UDCA), in low to moderate doses, has shown beneficial effects in treating PSC, while its derivative, Nor-UDCA, effectively reduces alkaline phosphatase (ALP). Farnesoid X receptor (FXR) agonists target liver biochemical pathways and can improve fibrosis markers, making them promising for PSC treatment. Peroxisome proliferator-activated receptor (PPAR) agonists, known for regulating fatty acid metabolism and reducing inflammation, also lower ALP levels, particularly when combined with UDCA. Other drugs, including antibiotics, steroids, and immunomodulators, are still in early-stage exploration but show potential for PSC management. Fibroblast growth factor-19 (FGF-19), cenicriviroc (CVC), and HTD1801 are currently being studied, though their efficacy and safety require further validation [
2].
In the study by Paik et al. [
3], published in this issue of
Clinical and Molecular Hepatology, the primary and secondary outcomes were assessed in a double-blind, placebo-controlled phase 2 trial. Patients were treated with either 100 mg of HK-660S or a placebo twice daily with concomitant UDCA for 12 weeks. They concluded that HK-660S is well-tolerated among patients with PSC and may improve bile duct strictures, decrease serum ALP levels, and reduce liver fibrosis [
3].
What I particularly like in this study is the bravery of the authors to include the Anali score as one primary endpoint and the improvement of bile duct strictures, enhanced liver fibrosis (ELF) score, and vibration-controlled transient elastography (VCTE) as secondary endpoints, in addition to conventional biochemical markers. Many studies designed to evaluate the efficacy of PSC have chosen biochemical parameters as endpoints. While those are important, they might not fully reflect the effect on disease progression and survival of PSC patients [
4].
Favourable improvements in PSC severity based on the magnetic resonance cholagio-pancreatography (MRCP) Anali score, biliary strictures, ELF score, liver fibrosis, and ALP compared to the placebo group are encouraging. Still, some issues are seen within the study. The smaller sample size has impacted significance calculations. Readers may appreciate the famous quote: “Not everything that counts can be counted, and not everything that can be counted counts,” elaborated by Austin Bradford Hill in reference to causality [
5,
6]. Higher age, baseline Anali score, and liver stiffness (with a mean in the advanced fibrosis stages) in the placebo group might have positively affected the favourable outcomes. The seven IBD patients in the treatment group also need more description, such as the type and extent of bowel involvement and whether the patients were on steroids or biologics.
Another important aspect is the assessment of the drug’s safety in the target population. It was shown that adverse drug reactions (ADRs) were mostly gastrointestinal-related and occurred only in the treatment group. Paik et al. [
3] mentioned that most of them were mild, with one moderate case and one case of bacteremia. This is a positive finding that will encourage further research into HK-660S for PSC.
As with any study, this one also has its limitations, which the authors have nicely identified and explained, such as the need for a longer treatment duration. Still, some aspects need further re-evaluation. For instance, the use of the NQO1 genotype as a stratification factor lacks an explanation of its relevance to PSC or the study drug’s mechanism. A clearer rationale for its inclusion would strengthen the study’s scientific foundation. Additionally, the handling of blinding is noted as problematic, as potential unblinding due to adverse drug reactions (e.g., abdominal pain, nausea) is not sufficiently addressed, which could bias the study. The primary endpoint, based on the MRCP Anali score to assess PSC severity, is appropriate but subject to inter-observer variability. Although two independent radiologists were employed, there is little information about agreement rates or how variability was managed. Furthermore, the decision to use a 10% significance level instead of the standard 5% undermines the robustness of the results, suggesting a less rigorous approach to statistical significance. Exploratory biomarkers are mentioned, but their relevance to PSC or the drug HK-660S is not well justified, leaving a gap in understanding their inclusion. While some AEs are listed, there is insufficient detail on how they were monitored or adjudicated, particularly in the case of a patient with bacteremia, where the cause and relationship to the drug remain unclear. More information on whether additional safety measures were implemented or if study protocols were adjusted after an SAE would improve the overall transparency of safety management in the study. This study will inform the planning of future research.
I would like to highlight another key aspect for future studies. We know that PSC has been classified into small-duct PSC (sdPSC), large-duct PSC (ldPSC), and PSC-AIH variant. Although rates vary, 10–30% of sdPSC cases can convert into ldPSC over 5–10 years [
7]. Histopathological changes differ between sdPSC and ldPSC (neutrophilic infiltrate is seen in ldPSC compared to lymphocytic and plasma cell infiltrates in sdPSC). Some genetic studies have shown that sdPSC with IBD and ldPSC differ from sdPSC without IBD [
8]. Therefore, it is important to involve different sets of patients at the recruitment phase and plan data analysis accordingly. Testing a drug may be easy, but outcome measurements might be more challenging in sdPSC compared to ldPSC.
Nonetheless, I would like to congratulate Paik et al. [
3] for planning and reporting this study, which will help PSC patients, who face a difficult and symptomatic natural history with higher cancer risk and lower survival.
FOOTNOTES
-
Conflicts of Interest
The author has no conflicts to disclose.
Abbreviations
primary sclerosing cholangitis
REFERENCES
- 1. Bowlus CL, Arrivé L, Bergquist A, Deneau M, Forman L, Ilyas SI, et al. AASLD practice guidance on primary sclerosing cholangitis and cholangiocarcinoma. Hepatology 2023;77:659-702.
- 2. Sheng X, Ji Q, Li X, Wang L, Niu J. Advances in drug therapy for primary sclerosing cholangitis. J Clin Hepatol 2024;40:1032-1038.
- 3. Paik WH, Park JK, Chung MJ, Huh G, Park CH, Lee SH, et al. Safety and efficacy of HK-660S in patients with primary sclerosing cholangitis: A randomized double-blind phase 2a trial. Clin Mol Hepatol 2025;31:119-130.
- 4. Shah A, Jones MP, Callaghan G, Fairlie T, Ma X, Culver EL, et al. Efficacy and safety of biologics in primary sclerosing cholangitis with inflammatory bowel disease: a systematic review and meta-analysis. Hepatol Commun 2024;8:e0347.
- 5. Cameron WB. Informal Sociology, a casual introduction to sociological thinking. New York: Random House; 1967. p. 13.
- 6. Hill AB. The environment and disease: association or causation? Proc R Soc Med 1965;58:295-300.
- 7. Nguyen CM, Kline KT, Stevenson HL, Khan K, Parupudi S. Small duct primary sclerosing cholangitis: A discrete variant or a bridge to large duct disease, a practical review. World J Hepatol 2022;14:495-503.
- 8. Naess S, Björnsson E, Anmarkrud JA, Al Mamari S, Juran BD, Lazaridis KN, et al. Small duct primary sclerosing cholangitis without inflammatory bowel disease is genetically different from large duct disease. Liver Int 2014;34:1488-1495.
Citations
Citations to this article as recorded by
- Correspondence to editorial on “Safety and efficacy of HK-660S in patients with primary sclerosing cholangitis: A randomized double-blind phase 2a trial”
Woo Hyun Paik, Do Hyun Park
Clinical and Molecular Hepatology.2025; 31(2): e158. CrossRef
