Dear Editor,
We read with interest the article by Wong et al. [
1] reporting that HTD1801 “demonstrates promising potential for histologic improvements” in metabolic dysfunction-associated steatohepatitis (MASH). While preclinical results showed histologic benefits in a hamster model, the Phase 2 trial in patients with presumed MASH relied on non-invasive biomarkers. Although the concept is intriguing, we have several concerns regarding the study’s design and interpretation.
First, the Phase 2 trial enrolled patients with “
presumed MASH” based on imaging and laboratory criteria (MRI-proton density fat fraction [MRI-PDFF] liver fat ≥10%, iron-corrected T1 [cT1] ≥830 ms, aspartate aminotransferase ≥20 U/L, and type 2 diabetes), without confirmatory liver biopsies [
1]. This lack of histological diagnosis introduces significant uncertainty. MASH (formerly NASH) is fundamentally a histopathologic diagnosis characterized by steatosis with inflammation and hepatocyte ballooning; current guidelines still consider liver biopsy the gold standard for diagnosing and staging MASH [
2]. By including patients without histologic confirmation, the trial may have enrolled individuals who did not actually have active steatohepatitis (or who had varying degrees of severity), potentially confounding the results. This diagnostic ambiguity undermines interpretability— improvements in surrogate markers cannot be definitively attributed to resolving steatohepatitis if baseline histology was not documented.
Second, reliance on non-invasive surrogates to infer “histologic improvement” is problematic without biopsy data. The authors used MRI-PDFF, cT1, alanine aminotransferase (ALT), fibrosis-4 index (FIB-4), and a composite “MASH resolution index” as proxies [
1]. While reductions in liver fat content or ALT and improvements in cT1 or FIB-4 have been associated with histological improvements in some studies [
3], these remain surrogate markers. No non-invasive test or biomarker panel is fully validated to replace liver biopsy for assessing treatment efficacy in NASH/MASH trials [
2]. Regulatory agencies continue to require actual histologic endpoints (e.g., resolution of steatohepatitis or fibrosis regression on biopsy) as surrogates for clinical benefit in NASH [
2]. Inferring that HTD1801 achieved “histologic improvement” based solely on imaging and lab surrogates is speculative.
Third, the small sample size and short duration limit the robustness of the study. Only ~33 patients per arm (100 total) were treated for 18 weeks,1 providing limited statistical power. It is well documented that meaningful histological changes in NASH (especially fibrosis regression) often require 12–18 months or longer to manifest in clinical trials [
4]. For comparison, resmetirom showed biopsy-proven benefit only after 52 weeks, with 26–30% achieving resolution [
5]. Thus, an 18-week trial with such a modest cohort is unlikely to capture durable histologic effects, and variability may yield false-negative or false-positive results [
2].
Fourth, the findings stem from a post hoc secondary analysis, raising concerns about multiple comparisons and selective reporting. The original Phase 2 trial by Harrison et al. did not include biopsy endpoints [
6], and this exploratory analysis evaluated numerous biomarkers not pre-specified as primary outcomes. Post hoc analyses are prone to bias, as unplanned comparisons increase false-positive risk, especially without adjustment for multiplicity [
7]. Emphasizing favorable results (e.g., MRI-PDFF or cT1) while others may not have improved can mislead readers. A data-dredging bias is most concerning in exactly this scenario—when a study is small, with multiple endpoints and exploratory analyses, the risk of spurious significant results is high [
8].
Fifth, there are notable statistical concerns. In the animal model, Grubbs’ test was used to remove outliers despite very small group sizes (n≈8) [
1]. Such exclusions risk eliminating valid variability and exaggerating efficacy; Grubbs’ test is generally recommended for larger samples (n>30), with alternatives like Dixon’s Q test more suitable for small n [
9]. In the clinical study, multiple biomarker endpoints were analyzed without correction for multiplicity. Testing many outcomes at α=0.05 greatly increases false-positive risk [
7]. For example, if one tests 10 independent outcomes, the probability of at least one nominally “significant” finding at P<0.05 exceeds 40%. The authors did not report using a Bonferroni or false discovery rate correction. This lack of adjustment, combined with the post hoc nature of the analysis, means the results could be overstated.
Finally, we feel the conclusions in the title and abstract are overstated given the data. The article proclaims “promising potential for histologic improvements” and “proof of concept of liver histologic improvement” with HTD1801, yet no histological endpoints were assessed. The authors themselves acknowledge that “histologic data are not yet available” [
1]. Evidence in humans is limited to surrogate markers; claiming histologic improvement is therefore premature. At best, the study shows improvements in biomarkers associated with histology, not direct resolution or fibrosis regression. Such overstatements risk misleading readers. We note positively that the authors have an ongoing Phase 2b trial specifically evaluating histological endpoints with HTD1801 [
1], which will be crucial to confirm whether these signals translate into real tissue benefit. Until then, assertions of “histologic improvement” should be viewed with caution.
In summary, Wong et al. present intriguing preliminary data on HTD1801, but the absence of biopsy-confirmed diagnosis, reliance on surrogate endpoints, small sample size, short follow-up, post hoc analysis, and statistical issues limit confidence in their conclusions. Future trials must include histological confirmation and adequately powered designs to establish efficacy. Until then, claims of proof-of-concept histologic improvement remain premature.
FOOTNOTES
-
Conflicts of Interest
The author has no conflicts to disclose.
Abbreviations
metabolic dysfunction-associated steatohepatitis
MRI-proton density fat fraction
REFERENCES
- 1. Wong VW, Neff GW, Di Bisceglie AM, Bai R, Cheng J, Yu M, et al. HTD1801 demonstrates promising potential for histologic improvements in metabolic dysfunction-associated steatohepatitis in both a preclinical and phase 2 study. Clin Mol Hepatol 2025;31:1071-1083.
- 2. Pulaski H, Harrison SA, Mehta SS, Sanyal AJ, Vitali MC, Manigat LC, et al. Clinical validation of an AI-based pathology tool for scoring of metabolic dysfunction-associated steatohepatitis. Nat Med 2025;31:315-322.
- 3. Alkhouri N, Beyer C, Shumbayawonda E, Andersson A, Yale K, Rolph T, et al. Decreases in cT1 and liver fat content reflect treatment-induced histological improvements in MASH. J Hepatol 2025;82:438-445.
- 4. Vuppalanchi R, Noureddin M, Alkhouri N, Sanyal AJ. Therapeutic pipeline in nonalcoholic steatohepatitis. Nat Rev Gastroenterol Hepatol 2021;18:373-392.
- 5. Brennan PN, Kopka CJ, Agirre-Garrido L, Hansen CD, Alkhouri N, Schattenberg JM, et al. Reviewing MAESTRO-NASH and the implications for hepatology and health systems in implementation/accessibility of Resmetirom. npj Gut Liver 2025;2:3.
- 6. Harrison SA, Gunn N, Neff GW, Kohli A, Liu L, Flyer A, et al. A phase 2, proof of concept, randomised controlled trial of berberine ursodeoxycholate in patients with presumed non-alcoholic steatohepatitis and type 2 diabetes. Nat Commun 2021;12:5503.
- 7. Chen SY, Feng Z, Yi X. A general introduction to adjustment for multiple comparisons. J Thorac Dis 2017;9:1725-1729.
- 8. Simmons JP, Nelson LD, Simonsohn U. False-positive psychology: undisclosed flexibility in data collection and analysis allows presenting anything as significant. Psychol Sci 2011;22:1359-1366.
- 9. Natural History Museum, University of Oslo. Outlier tests. Natural History Museum, University of Oslo web site, <https://www.nhm.uio.no/english/research/resources/past/help/outliers.html>. Accessed 1 Aug 2025.
Citations
Citations to this article as recorded by
- Safety and metabolic effects of HTD1801 in type 2 diabetes and MASLD: a phase Ib randomized trial
Jiajia Mai, Hong Zhang, Min Wu, Meng Yu, Kui Liu, Guixia Wang, Yanhua Ding
Diabetology & Metabolic Syndrome.2026;[Epub] CrossRef
