Twenty-one patients with lesions 2 cm or smaller. Ultrasound detected none.
In the smallest lesions in the trial, ultrasound sensitivity was 0%. The HelioLiver Dx test detected 6 of 21.
How well does abdominal ultrasound detect very early hepatocellular carcinoma lesions 2 cm or smaller?
In the CLiMB trial, abdominal ultrasound detected 0% of the 21 hepatocellular carcinoma lesions measuring 2 cm or smaller, with a 95% confidence interval of 0.0 to 16.1. The HelioLiver Dx test detected 6 of those 21, a sensitivity of 28.6%.
Cohort at a glance
| Cohort | 21 CLiMB participants whose largest HCC lesion was 2 cm or smaller |
|---|---|
| Reference standard | Multiphasic MRI |
| Ultrasound sensitivity | 0% (95% CI 0.0-16.1) |
| HelioLiver Dx sensitivity | 28.6% (95% CI 11.3-52.2) |
| Other modalities | 0% to 4.8% |
| Share of all HCC cases in CLiMB | 46% |
Low ultrasound sensitivity for small HCC nodules
Patients at high risk for HCC are recommended to undergo biannual abdominal ultrasound surveillance. The publication notes, however, that ultrasound has low sensitivity for small HCC nodules. The publication describes the survival benefit of early HCC detection as deriving from the opportunity for timely initiation of potentially curative treatments, including surgical resection, ablation therapies, and, for eligible patients, liver transplantation.
Cohort
Of the 46 participants confirmed to have HCC by MRI in the CLiMB validation cohort, 21 had a largest lesion of 2 cm or smaller. The publication reports that most participants with HCC were found to have small lesions: 80% were 4 cm or smaller in diameter and 46% were 2 cm or smaller.
What the data showed
Every HCC lesion in CLiMB, by size. The trial reported the largest lesion for each of the 46 participants with HCC. Nearly half were 2 cm or smaller, the band in which ultrasound detected nothing at all.
| Largest lesion 2 cm or smaller | 21 of 46 (45.7%) — HelioLiver Dx 28.6%, ultrasound 0% |
|---|---|
| Larger than 2 cm, up to 3 cm | 13 of 46 (28.3%) — HelioLiver Dx 53.8% |
| Larger than 3 cm, up to 4 cm | 3 of 46 (6.5%) |
| Larger than 4 cm | 9 of 46 (19.6%) |
Ultrasound detected none of them. Zero of twenty-one, with a 95% confidence interval of 0.0 to 16.1. The other comparators performed similarly, with sensitivities ranging from 0% to 4.8%.
The HelioLiver Dx test detected 6 of the 21, a sensitivity of 28.6% with a 95% confidence interval of 11.3 to 52.2. In the next size band up, lesions larger than 2 cm and up to 3 cm, its sensitivity was 53.8% (95% CI 25.1–80.8).
The trial's prespecified secondary endpoint was small lesions, defined as ≤4 cm. As reported in the publication, the HelioLiver Dx test had sensitivities that were 28.6% and 24.3% higher than ultrasound for HCC lesions ≤2 cm and ≤4 cm, respectively.
For context, across all HCC lesions of any size, the HelioLiver Dx test had a sensitivity of 47.8% and a specificity of 87.6%, compared with 28.3% sensitivity and 93.9% specificity for ultrasound. The trial met its prespecified co-primary endpoints for superior sensitivity and non-inferior specificity.
What it means at the point of care
In this trial, at that size, ultrasound found nothing. The publication attributes variability in ultrasound performance to patient characteristics, including BMI and lesion size, and to operator experience. It also notes the wide range of cumulative sensitivities, from 28% to 100%, reported for ultrasound in a recent meta-analysis, and distinguishes those cumulative figures from the single-visit sensitivity measured in CLiMB.
The publication highlights this result: ultrasound alone failed to detect any very early HCC lesions 2 cm or smaller, while the HelioLiver Dx test detected 6 of the 21.
Three things to carry forward
- Nearly half of all HCC cases detected in CLiMB had a largest lesion of 2 cm or smaller.
- Abdominal ultrasound detected 0% of those lesions when compared against concurrent multiphasic MRI.
- The HelioLiver Dx test detected 28.6% of them, identifying 6 patients in a size band where ultrasound alone detected none.
- Taggart DJ, Mahajan S, Gallant MA, et al. A multi-analyte cfDNA-based blood test for early detection of hepatocellular carcinoma. J Hepatol. 2026;1-9. doi:10.1016/j.jhep.2026.04.012
- Tzartzeva K, Obi J, Rich NE, et al. Surveillance imaging and alpha fetoprotein for early detection of hepatocellular carcinoma in patients with cirrhosis: a meta-analysis. Gastroenterology. 2018;154:1706-1718.e1.
About the CLiMB trial
CLiMB was a cross-sectional, prospective, blinded, multicenter validation study conducted across 42 clinical sites in the United States, evaluating the HelioLiver Dx test against abdominal ultrasound for the detection of hepatocellular carcinoma in adults with cirrhosis. Multiphasic MRI served as the reference standard. The HelioLiver Dx test met prespecified co-primary endpoints for superior sensitivity and non-inferior specificity compared to ultrasound.
Enrolled vs. evaluable. A total of 1,556 patients were enrolled in the prospective CLiMB study and assigned to the validation cohort, of whom 1,285 completed the initial visit. At the 6-month follow-up visit, 17 participants were lost to follow-up and were excluded from analysis. Only participants with valid results from the HelioLiver Dx test, ultrasound, and multiphasic MRI were considered evaluable. A total of 1,268 participants (81.5%) were evaluable, including 46 participants with HCC and 1,222 participants without HCC. Every performance figure in this series is calculated on that evaluable population. Published in the Journal of Hepatology, 2026. ClinicalTrials.gov identifier NCT03694600.
Have a case to contribute?
Clinicians using the HelioLiver Dx test can submit de-identified cases for consideration in this series.