K.C. van Son^{1, 2}, S. Driessen¹, G.L. Haverkamp^1,3, M.J. Denters^1,3, S.J. Pinto-Sietsma¹, N.M.J. Hanssen¹, M. Castro Cabezas^4,5, R.B. Takkenberg¹, H. Schers^2,6, M. Nieuwdorp¹, J.P. Drenth¹, M.E. Tushuizen⁷, A.G. Holleboom¹

Affiliations:

1. Amsterdam UMC, 2. Radboudumc, 3. Zaans Medisch Centrum, 4. Julius Clinical, 

5. Franciscus Gasthuis, 6. Huisartsenpraktijk Thermion, 7. LUMC

Background and Aims:

Non-invasive liver tests (NITs) can detect advanced fibrotic MASLD and reduce unnecessary referrals. Yet data comparing numbers-needed-to-screen (NNS) in multiple lines of care are scarce. Therefore we introduced NLA2, the first Dutch MASLD care path, encompassing primary, secondary and tertiary clinics.

Method:

Patients at cardiometabolic risk for MASLD were recruited from GPs, regional clinics and a UMC, whilst excluding other liver diseases. Simultaneous FIB4, MAF5, Enhanced Liver Fibrosis (ELF)-test and vibration-controlled transient elastography (VCTE/FibroScan®) allowed testing of NIT combinations. FIB4 ≥3.25 and/or Liver Stiffness Measurement (LSM) ≥8.0 kPa indicated potential advanced fibrosis, prompting referral to hepatology. Referral patterns were compared to regular care between 2016-2024 using a predefined evaluation standard.

Results:

655 participants entered NLA2, 270 from primary, 156 from secondary and 229 from tertiary care. After excluding patients with MetALD or other liver diseases, 597 were analyzed. Median age was 60 years (51-68), 45.9% were women, 51.4% had obesity and 45.2% had T2DM. 73.9% of participants had CAP ≥248 dB/m, suggesting ≥S1 steatosis. 15.5% had LSM ≥8.0 kPa, yielding a NNS of 6.7. NNS was even lower in T2DM compared to non-diabetics; 4.3 vs 11.2. NNS decreased across lines of care: 12.1, 5.1 and 5.2 for primary, secondary and tertiary care respectively, inversely related to cardiometabolic comorbidities (mean number 2.1, 2.3, 2.9). ELF was available for 535 participants. 20.0% had ELF ≥9.8, yielding NNS of 5.0. FIB4 stratifies 64% as low risk for advanced fibrosis and 36% as intermediate/high risk. MAF5 stratifies 40% as low risk for advanced fibrosis and 60% as high risk.

In a 2-tiered algorithm in case of MAF5 ≥1, VCTE as second test would refer 6.4%; using ELF would refer 11.7%. In a 2-tiered algorithm in case of intermediate FIB4 (1.3-3.25), VCTE as second test would refer 13.1%; using ELF would refer 13.5%. Of 96 patients at risk for advanced fibrosis, 72 were referred to hepatology. 465 patients were referred to hepatology via regular care between 2016 and 2024. Compared to regular care, NLA2 improved correct referrals 4-fold (RR 4.03; 95% CI 2.53-6.41). Unnecessary referrals were reduced 2-fold (RR 0.44; 95% CI 0.27-0.69).

Conclusion:

The first Dutch MASLD care path study shows incrementally lower NNS for potential advanced fibrosis detection across multiple lines of care and a significantly improved referral pattern for fibrotic MASLD. NNS in T2DM compared to non-diabetics were remarkably lower.We found a marked difference in referral-rate between different NIT combinations.