Post-Translational Changes in Serum Albumin in Patients with Alcohol-Associated Hepatitis

Bibliographic info

• Authors: Jonathan Montomoli, Maurizio Baldassarre, Thomas Damgaard Sandahl, Marina Naldi, Emilie Glavind, Enrico Pompili, Peter Jepsen, Francesco Palmese, Paolo Caraceni, Hendrik Vilstrup, Marco Domenicali
• Journal: International Journal of Molecular Sciences (IJMS), 2026, vol. 27, art. 1503; Open Access (CC BY); Received 27 Nov 2025, Accepted 30 Jan 2026, Published 3 Feb 2026
• Institutions: Aarhus University Hospital, Denmark; IRCCS AOU Bologna; University of Bologna, Italy; Infermi Hospital, Rimini, Italy

Key question

Do HSA post-translational changes — specifically the HMA/HNA1/HNA2 fractions and the full isoform profile including native, cysteinylated, glycated, and truncated forms — differ between survivors and non-survivors in patients with alcohol-associated hepatitis (AAH), and can they predict 90-day outcome?

Methods

• Sample type: Human serum — baseline (admission) and day 14
• Cohort: 49 patients with alcohol-associated hepatitis (AAH); 38 survivors at 90 days; 11 non-survivors at 90 days; 20 healthy subjects (comparison)
• Clinical scoring: MELD (median = not shown), GAHS (Glasgow Alcohol-associated Hepatitis Score)
• Technique: LC-MS (top-down; HPLC-ESI-MS) for HSA isoform quantification
• Isoforms measured: Native HSA; HMA (total mercaptalbumin); HNA1; HNA2; HSA-DA (N-term truncated); HSA-L (C-term truncated); HSA+Cys-DA; HSA-DHA (dehydroalanine at Cys34)
• Timepoints: Admission (D0) and day 14
• Statistical: Wilcoxon rank-sum test; chi-squared; comparison AAH vs healthy subjects

Main findings

Cohort characteristics (at admission)

Characteristic	AH Population (N=49)	90-Day Survivors (N=38)	90-Day Non-Survivors (N=11)	p-value
Age (median IQR)	53 (48–58)	52 (47–57)	54 (48–58)	0.611
Male sex	34 (69%)	27 (71%)	7 (64%)	0.638
Albumin (g/L)	26 (22–30)	26 (22–30)	26 (23–31)	0.622
CRP (mg/dL)	32 (23–…)	—	—	—

HSA isoform fractions at admission (% median IQR)

Isoform	AH (N=49)	90-Day Survivors	90-Day Non-Survivors	p-value
Native HSA	34.2 (30.2–44.1)	36.2 (30.4–46.4)	32.6 (29.7–37.4)	0.523
HMA	62.3 (56.1–67.9)	63.5 (57.3–68.3)	59.6 (53.3–64.2)	0.337
HNA1	32.6 (27.2–39.5)	31.7 (26.8–39.1)	34.79 (30.9–43.3)	0.337
HNA2	4.4 (3.1–5.9)	4.1 (3.2–5.9)	4.8 (2.8–6.2)	0.915

HSA isoform fractions at day 14 (% median IQR)

Isoform	AH Population	90-Day Survivors	90-Day Non-Survivors	p-value
Native HSA	39.3 (31.2–43.7)	39.8 (32.0–43.7)	38.9 (29.7–44.0)	0.775
HMA	62.2 (56.9–68.7)	62.1 (57.0–67.6)	68.7 (58.5–71.0)	0.145
HNA1	32.4 (25.2–37.6)	32.9 (26.4–38.2)	26.2 (23.8–35.0)	0.171
HNA2	4.9 (4.3–6.1)	4.8 (4.2–6.1)	5.0 (4.4–5.9)	0.881

Key finding

• No significant difference in HSA isoform fractions between 90-day survivors and non-survivors at admission or day 14 (all p > 0.05)
• Trend: non-survivors had slightly lower native HSA and HMA, higher HNA1 at admission — but underpowered (N=11 non-survivors)
• AAH patients showed different HSA isoform distribution vs healthy subjects (Figure 1 in paper)
• ⚠️ HSA isoforms in AAH do not significantly predict 90-day survival in this cohort

PTMs characterized

Isoform	Modification	Detection	Notes
HMA	Reduced Cys34	LC-MS	Dominant healthy/AAH form
HNA1	Reversible oxidation Cys34 (+119 Da cysteinylation, or disulfide)	LC-MS	Elevated in AAH vs healthy
HNA2	Irreversible oxidation	LC-MS	Low but elevated in AAH
HSA-DHA	Dehydroalanine conversion at Cys34	LC-MS	Novel isoform mentioned
HSA-DA	N-term truncation	LC-MS	Measured
HSA+Cys-DA	CYS + N-term truncation	LC-MS	Measured

Clinical context

• Disease: Alcohol-associated hepatitis (AAH) — acute-on-chronic alcohol-related liver disease; 49 patients
• Cohort sites: Aarhus University Hospital (Denmark); Bologna (Italy) — Scandinavian + Italian cohort
• Implication: HSA PTM profiling in AAH shows altered isoform distribution vs healthy but does not predict survival at 90 days in this small cohort. Underpowered for the 11 non-survivor group.

Limitations

• Small sample (49 total; only 11 non-survivors) — major limitation for survival analysis
• Relative quantification only (%)
• No multivariate survival analysis reported in available pages — MELD and GAHS likely dominate
• Day 14 timepoint provides additional information but no significant discrimination

Connections

• HSA — isoform profiling in AAH; first data with day 14 timepoint
• naldi-2016-ah — preceded this paper; smaller characterization study of HSA in AH
• das-2017-sah — SAH albumin modulates neutrophils; provides mechanistic context
• domenicali-2014 — established isoform profiling in stable cirrhosis; AAH is a different clinical context
• el-balkhi-2025 — ALBOM study (stable CLD fibrosis); native HSA predicts fibrosis stage; AAH here is acute setting where this may not apply similarly

Take home notes

• The null result (no survival prediction) is important to capture — it does not mean HSA isoforms are useless in AAH, but the n=11 non-survivors is severely underpowered. A larger study would be needed.
• Montomoli is the first author — a clinician at Aarhus + Rimini. He is co-author on naldi-2016-ah too. This study is the clinical expansion of that earlier work.
• The HSA-DHA (dehydroalanine) isoform is mentioned — this is a chemical conversion of Cys34 that can occur in alkaline conditions or as a PTM. Not typically quantified in the ALBOM panel — worth investigating.
• The comparison of AAH vs healthy subjects confirms that AAH does alter the isoform distribution (HNA1 elevated, native HSA decreased relative to healthy) — the question is just about survival prediction, which requires larger cohorts.