In the article “Tumor Mutational Burden as a Prognostic Biomarker for Relapse-Free Survival in Hepatocellular Carcinoma: Insights from Long-Term Follow-Up” [Liver Cancer. 2025; https://doi.org/10.1159/000547996] by Miura et al., the authors noted there was an error in Figure 2 and Table 2.
Fig. 2.
Landscape of genomic alterations in patients with hepatocellular carcinoma stratified by tumor mutation burden (TMB). The frequency of somatic mutations in specific genes, including CTNNB1, TTN, and MUC16, was significantly higher in the TMB-high group than in the TMB-low group. An inverse relationship was observed between CTNNB1 mutations and TP53 mutations.
Table 2.
Clinicopathological factors according to TMB in 115 cases with no recurrence within 2 years and with more than 2 years of follow-up
| Variables | TMB-high (≥4.8) | TMB-low (<4.8) | p value |
|---|---|---|---|
| n = 19 | n = 96 | ||
| Patient characteristics | |||
| Age, yearsa | 72 (68–79) | 70 (66–76) | 0.103 |
| Sex (male) | 18 (94.7) | 71 (74.0) | 0.069 |
| AFP, ng/mLa | 5.2 (3.3–35.2) | 7.1 (3.2–67.6) | 0.892 |
| PIVKA-II, mAU/mLa | 381 (37–2,390) | 75 (24–776) | 0.130 |
| Hepatitis virus | 10 (52.6) | 48 (50.0) | 1.000 |
| HBV | 2 (10.5) | 20 (20.8) | 0.522 |
| HCV | 8 (42.1) | 28 (29.2) | 0.287 |
| Habitual alcohol drinking | 4 (22.1) | 13 (13.5) | 0.478 |
| ICG K | 0.165 (0.129–0.197) | 0.167 (0.146–0.195) | 0.543 |
| ICG R15, % | 10.2 (6.9–15.8) | 9.1 (6.1–12.9) | 0.248 |
| Pathological findings | |||
| Background | | | |
| Cirrhosisb | 2 (10.5) | 26 (27.1) | 0.153 |
| Fibrosisc | | | 0.033 |
| F0 | 6 (31.6) | 52 (54.2) | |
| F1–2 | 3 (15.8) | 23 (24.0) | |
| F3–4 | 10 (52.6) | 21 (21.9) | |
| Tumor size, mma | 50 (34–78) | 32 (25–51) | 0.010 |
| Tumor number (single lesion) | 18 (94.7) | 85 (88.5) | 0.687 |
| Differentiation | | | 0.291 |
| Well | 6 (31.6) | 22 (22.9) | |
| Mod | 12 (63.2) | 72 (75.0) | |
| Por | 1 (5.3) | 2 (2.1) | |
| Microscopic vascular invasion (+) | 6 (31.6) | 21 (21.9) | 0.382 |
| Portal vein | 4 (21.1) | 17 (17.7) | 0.748 |
| Hepatic vein | 3 (15.8) | 5 (5.2) | 0.125 |
| Microsatellite lesion (+) | 1 (5.3) | 5 (5.2) | 1.000 |
| UICC stage (II or III) | 1 (5.3) | 14 (14.6) | 0.459 |
| Molecular correlates | |||
| Estimated tumor content (tumor tissue samples)a | 0.57 (0.44–0.73) | 0.51 (0.32–0.63) | 0.106 |
| MSI-high | 1 (5.3) | 2 (2.1) | 0.421 |
| CNV size: gain (MB) | 319.5 (199.1–432.7) | 244.3 (132.5–395.9) | 0.101 |
| CNV size: loss (MB) | 210.4 (138.7–323.1) | 189.6 (94.6–281.0) | 0.207 |
| CNV size: LOH (MB) | 30.7 (11.7–59.0) | 8.3 (0.1–36.0) | 0.018 |
| TP53 mutation | 17 (10.5) | 80 (16.7) | 0.733 |
| CTNNB1 mutation | 11 (42.1) | 65 (32.3) | 0.435 |
| ARID1A mutation | 2 (10.5) | 10 (10.4) | 1.000 |
| Recurrence | 12 (63.2) | 33 (34.4) | 0.023 |
| Initial recurrence sites | |||
| Liver | 10 (52.6) | 26 (27.1) | 0.055 |
| Distantd | 2 (10.5) | 7 (7.3) | 0.642 |
| Lung | 1 (8.3) | 1 (3.0) | |
| Bone | 1 (8.3) | 6 (18.2) | |
| Lymph node | 0 (0.0) | 1 (3.0) | |
| Peritoneum | 0 (0.0) | 1 (3.0) | |
AFP, alpha-fetoprotein; PIVKA-II, protein induced by vitamin K absence or antagonist-II; UICC, Union for International Cancer Control; CNV, copy number variation.
Values in parentheses are percentages unless otherwise indicated.
aMedian (interquartile range).
bAccording to the New Inuyama Classification.
cFor transparency, the “cirrhosis” row is retained as originally recorded in pathology reports, although the fibrosis staging (F0–F4) was re-evaluated for this study and defines F4 as cirrhosis.
dThere are some duplicates in the details of distant recurrence.
During figure preparation, the incorrect cutoff of tumor mutation burden (TMB) >5 mutations/Mb was included; the correct cutoff should be TMB ≥4.8 mutations/Mb. The TTN and MUC16 genes were omitted. The authors checked the original analysis, and it was performed using the cutoff of TMB ≥4.8 mutations/Mb. The error in Figure 2 does not affect the results. The corrected Figure 2 is shown below.
In Table 2, the row “cirrhosis” was derived from pathology report descriptions and was not part of the re-evaluated fibrosis staging (F0–F4) performed specifically for this study. Because F4 corresponds to cirrhosis, the cirrhosis row conceptually overlaps with the fibrosis staging and contains incomplete information. A note has been added at the end of Table 2 to reflect this.