Abstract
Background
Tumour necrosis is linked to worse outcomes in hepatocellular carcinoma (HCC). However, the relationship between the extent of necrosis in HCC and survival outcomes remains unclear, partly due to the lack of a standardised assessment method. This study evaluated the prognostic significance of both the presence and extent of tumour necrosis.
Methods
This retrospective study included 96 HCC patients from two Finnish centres, treated with surgical resection from 1986 to 2022. The presence of necrosis (yes/no) was evaluated. The extent of necrosis was also assessed using three methods: (1) The average percentage method (proportion of necrosis relative to total tumour area), (2) the hotspot method (proportion of necrosis within a 2 mm hotspot), and (3) the linear method (diameter of the largest necrotic focus).
Results
Tumour necrosis was associated with worse 5-year overall (HR 2.80, 95% CI 1.26–6.22, P = 0.011) and disease-specific survival (HR 3.89, 95% CI 1.45–10.47, P = 0.007). Tumours with low (but nonzero) necrosis level, as measured by the average percentage, hotspot, and linear methods, were linked to poorer survival outcomes compared to tumours without necrosis.
Conclusion
Tumour necrosis is associated with poorer overall and disease-specific survival in resected HCC. The extent of necrosis did not correlate linearly with survival, highlighting the need for further research to refine its prognostic value.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12885-026-16040-z.
Keywords: Hepatocellular carcinoma, Surgical resection, Prognostic biomarker, Tumour necrosis
Background
Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related mortality worldwide [1]. Although overall HCC management has improved, treatment strategies for surgically resectable HCC have remained largely unchanged for decades, with survival rates of 40–50% even after treatment with curative intent [2]. There is currently no adjuvant treatment protocol for HCC, although the recurrence rate after curative surgery is as high as 70% [3]. Furthermore, there is no proven benefit from neoadjuvant therapies, even though nearly 70% of newly diagnosed cases present at an advanced stage [4]. Prognostic biomarkers are urgently needed that can reliably predict tumour behaviour, recurrence risk and patient outcomes. Such biomarkers could inform decisions regarding closer postoperative surveillance or consideration of adjuvant therapy.
In HCC, rapid tumour growth and impaired perfusion can result in tumour necrosis, reflecting aggressive tumour biology associated with adverse outcomes [5, 6]. Damage-associated molecular patterns (DAMPs) released from dying tumour cells, may trigger inflammatory responses that promote tumour progression [7, 8]. Tumour necrosis has been associated with poor prognosis across several solid cancers, including colorectal [9], renal [10], breast [11], lung [12] and pancreatic cancer [13], as well as HCC [5, 14–16].
Because tumour necrosis can be assessed on standard haematoxylin- and eosin (H&E)-stained sections, it is an attractive candidate prognostic biomarker. However, the optimal approach for quantifying tumour necrosis in HCC remains inadequately defined [5, 14–16]. While comparative studies have been conducted in some other malignancies, such as colorectal cancer [9], there is a lack of consensus on standardised assessment methods specific to HCC [5].
Our study evaluated the prognostic role of tumour necrosis in a bi-institutional cohort of HCC patients. Additionally, we compared three methods for quantifying tumour necrosis in HCC with respect to feasibility, agreement between methods, and prognostic stratification.
Methods
Patient cohort
All consecutive patients undergoing surgical resection for hepatocellular carcinoma were identified from the pathology archives at Oulu University Hospital (1986–2022) and Central Finland Central Hospital (1997–2022). Eligible patients were required to have histologically confirmed hepatocellular carcinoma. Patients who received neoadjuvant therapy prior to resection were excluded (n = 3). Of the 99 patients initially identified, 96 fulfilled the inclusion criteria and constituted the final study cohort (Fig S1). The study received approval from the Finnish Medicines Agency Fimea (Dnro FIMEA/2021/004928) and the Ethical Committee of the North Ostrobothnia Hospital District (EETTMK 81/2008). The study was conducted in accordance with the Declaration of Helsinki.
Data collection
We identified patients from the pathology archives using the ICD-10 code C22.0&. Relevant clinical information was collected from the patient charts and pathology reports. The 8th edition of the TNM classification was used for staging. Comorbidities were assessed using the Charlson Comorbidity Index (CCI), and postoperative complications were classified according to the Clavien-Dindo classification system. Cause of death and date of death were obtained from Statistics Finland. Follow-up data was gathered up to the end of 2022.
Diagnostic H&E-stained tumour slides were retrieved from pathology archives and digitised using Aperio AT2 (Leica Biosystems, Wetzlar, Germany) and viewed using Aperio ImageScope. Tumour necrosis was assessed by a single investigator (N.S.) following training and calibration with an experienced pathologist (J.P.V.). Ambiguous cases were reviewed jointly with J.P.V. The tumour grade was evaluated by J.P.V. when it was not documented in the original pathology report.
Histopathological examination
Tumour necrosis in H&E-stained sections was specified as an area exhibiting nuclear shrinkage, fragmentation and disappearance, and the presence of ghost-like tumour cell shadows, frequently accompanied by neutrophilic inflammatory cell infiltrate [17]. For presence/absence analyses, tumour necrosis was recorded as present if any unequivocal necrotic focus as defined above was identified in any reviewed slide, after excluding artefacts. Tumour necrosis was recorded as absent if no such necrotic areas were seen in any of the available tumour slides.
For samples exhibiting tumour necrosis, quantitative assessment was performed using the method described by Kastinen et al. [9] in colorectal cancer (Fig. 1). In the first method, “the average percentage method”, the average percentage of tumour necrosis relative to the tumour epithelial area was evaluated. In the second method, the “hotspot method”, a circle with a diameter of 2 mm was placed over the most necrotic hotspot, and the percentage of necrosis relative to the area of the circle was determined. Vessel structures and artefacts were excluded when assessing the proportions of necrosis. In the third method, the “linear method”, we measured the length of the single largest necrotic area. All available tumour slides were included in the assessment. The median number of slides reviewed was 3 (range 1–5).
Fig. 1.
Evaluation methods for quantifying tumour necrosis in hepatocellular carcinoma. A Overview of the whole sample. B Assessment of necrosis (black areas) using the average percentage method. C Assessment of necrosis (black areas) using a 2 mm hotspot. Blood vessels and artefacts (red areas) were excluded when calculating necrosis in both the average percentage and hotspot methods. D Assessment of necrosis using the linear method, measuring the diameter of the largest continuous necrotic area
Outcomes
Both the presence and extent of tumour necrosis were evaluated for their association with 5-year overall survival and disease-specific survival. Overall survival was defined as the time from surgical resection to death from any cause or censoring at the last known follow-up date. Disease-specific survival was defined as the time from surgical resection to death attributed to HCC, with deaths from other causes censored at the date of death.
Statistical analysis
Tumour necrosis was analysed (1) as a binary variable (present vs. absent) and (2) using three quantification methods. For method-specific analyses, patients were categorised into three groups (none, low, high) using pre-defined cut-off. For the average percentage method, the cut-off was 5% (low < 5%, high ≥ 5%). For the linear method, the cut-off was 6000 μm (low < 6000 μm, high ≥ 6000 μm). For the hotspot method, where the median was 100%, patients were categorised as < 100% (low) versus 100% (high).
Baseline variables are presented as proportions, means with standard deviations (SD), or medians with interquartile ranges (IQR), as appropriate. Categorical variables were compared using the chi-square test. Continuous variables were compared with Student’s t-test or the Mann–Whitney U test, and by one-way analysis of variance or the Kruskal–Wallis test when comparing three or more groups.
Overall and disease-specific survival were estimated using the Kaplan-Meier method and compared using the log-rank test. Univariable and multivariable associations with survival were evaluated using Cox proportional hazards regression with time from surgical resection as the time scale. Separate models were fitted for overall survival and disease-specific survival. Because the three necrosis measures are correlated, they were evaluated in separate Cox models (one necrosis variable per model). Covariates were selected a priori based on established prognostic factors and clinical plausibility, and all covariates were entered simultaneously in each multivariable model: age (continuous), gender (male vs. female), Charlson comorbidity index (0–1 vs. ≥ 2), BCLC stage (0-A vs. B-D), TNM stage (1 vs. ≥ 2), cirrhosis (no vs. yes), year of surgery (1986–2005 vs. 2006–2022), Child–Pugh class (A vs. B/C), and tumour grade (1–2 vs. 3). Results are reported as hazard ratios (HRs) with 95% confidence intervals (CIs).
To compare the three necrosis quantification methods (second aim), we assessed (i) feasibility and score distribution (including the proportion classified as low and high by each method), (ii) agreement between continuous necrosis measures among necrosis-positive tumours using Spearman rank correlation (rₛ), and (iii) prognostic stratification by contrasting effect sizes (HRs) and precision (95% CIs) from the method-specific Cox models.
The threshold for statistical significance was P < 0.05. All analyses were performed using IBM SPSS Statistics 29.0 (IBM Corp., Armonk, NY).
Results
Patients
A total of 96 patients were included in this cohort. The median follow-up time was 2.5 years (IQR 1.4–5.7). The median age was 69.6 years (IQR 64.5–73.6), and most patients were men (66%). The median CCI was 1.3 and most patients (82.0%) had preserved liver function (Child-Pugh class A). The median tumour size was 50 mm (IQR 31–80). Most patients (72.9%) had an early-stage disease (BCLC 0-A) and well- or moderately-differentiated tumours (88.5%). No patients in this cohort received locoregional therapy prior to surgery.
Presence of tumour necrosis.
Histological assessment revealed tumour necrosis in 50 (52.1%) tumours (Table 1). When comparing patient and tumour characteristics, the presence of tumour necrosis was significantly associated with female sex (P = 0.038), larger tumour size (P = 0.009) and higher tumour grade (P < 0.001) (Table 1).
Table 1.
Baseline clinicopathological characteristics of patients according to tumour necrosis
| Tumour necrosis | ||||
|---|---|---|---|---|
| Absent (N = 46) | Present (N = 50) | P Value | ||
| Age. years. Median (IQR) | 68.7 (64.0-72.6) | 70.8 (65.4–74.6) | 0.798 | |
| Male. N (%) | 35 (76.1%) | 28 (56.0%) | 0.038 | |
| Charlson Comorbidity Index. Mean (SD) | 1.2 (0.8) | 1.5 (1.0) | 0.083 | |
| Cirrhosis. N (%) | 22 (47.8%) | 17 (34.0%) | 0.168 | |
| Child-Pugh Score. N (%) | A | 33 (76.7%) | 40 (87.0%) | 0.213 |
| B | 10 (23.3%) | 6 (13.0%) | ||
| C | 0 (0%) | 0 (0%) | ||
| Tumour size. mm. Median (IQR) | 44 (30–60) | 55 (43–100) | 0.009 | |
| Unifocal tumour. N (%) | 32 (69.6%) | 39 (78.0%) | 0.347 | |
| TNM stage. N (%) | Stage 1 | 27 (58.7%) | 32 (64.0%) | 0.691 |
| Stage 2 | 12 (26.1%) | 11 (22.0%) | ||
| Stage 3 | 7 (15.2%) | 4 (8.0%) | ||
| Stage 4 | 0 (0%) | 3 (6.0%) | ||
| BCLC stage. N (%) | Very early (0) | 1 (2.2%) | 1 (2.0%) | 0.575 |
| Early (A) | 31 (67.4%) | 37 (74.0%) | ||
| Intermediate (B) | 11 (23.9%) | 8 (16.0%) | ||
| Advanced (C) | 3 (6.5%) | 4 (8.0%) | ||
| Terminal (D) | 0 (0%) | 0 (0%) | ||
| Tumour grade. N (%) | Grade 1 | 24 (52.2%) | 12 (24.0%) | < 0.001 |
| Grade 2 | 21 (45.7%) | 28 (56.0%) | ||
| Grade 3 | 1 (2.2%) | 10 (20.0%) | ||
| Positive resection margin. N (%) | 3 (7.1%) | 3 (6.4%) | 0.887 | |
| Year of resection | 1986–2005 | 12 (26.1%) | 21 (42.0%) | 0.101 |
| 2006–2022 | 34 (73.9%) | 29 (58.0%) | ||
| AFP. Median (IQR) | 5 (3–8) | 10 (4-259) | 0.082 | |
Values are presented as median (IQR) or number (percentage). Baseline characteristics are compared between patients with absent and present tumour necrosis. P-values < 0.05 were considered statistically significant and are shown in bold
IQR Interquartile range, SD Standard deviation, TNM Tumour–node–metastasis, BCLC Barcelona Clinic Liver Cancer, AFP Alpha-fetoprotein
In patients with tumour necrosis, the 1-, 3- and 5-year overall survival rates were 83.6%, 58.2% and 40.1%, while the disease-specific survival rates were 85.4%, 66.8% and 56.9%, respectively (Fig 2). For patients without necrosis, the respective 1-, 3- and 5-year overall survival rates were 95.6%, 79.5% and 67.5% and disease-specific survival rates were 97.7%, 84.3% and 75.1% (Fig 2). The presence of tumour necrosis was associated with worse 5-year overall (HR 2.45, 95% CI 1.21–4.98, P = 0.013) and disease-specific survival (HR 2.41, 95% CI 1.05–5.55, P = 0.038) in univariable analysis (Table 2). After adjusting for confounding factors, tumour necrosis was associated with poorer overall (HR 2.80, 95% CI 1.26–6.22, P = 0.011) and disease-specific survival (HR 3.89, 95% CI 1.45–10.47, P = 0.007) (Table 2). A complete univariable and multivariable Cox regression results for all covariates included in the survival models are displayed in the supplementary material (Table S1). Along with tumour necrosis, BCLC stage and year of surgery were associated with survival outcomes in multivariable models (Table S1).
Fig. 2.
Survival of resected hepatocellular carcinoma by tumour necrosis status. A Overall survival. B Disease-specific survival
Table 2.
Hazard ratios (HR) with 95% confidence intervals (CI) of overall and disease-specific survival, comparing resected HCC patients according to tumour necrosis evaluation method
| Tumour necrosisa | Average percentage methodb | Hotspot methodc | Linear methodd | |||||
|---|---|---|---|---|---|---|---|---|
| Absent (N = 46) | Present (N = 50) | Low (N = 25) | ≥ 5% (N = 25) | Low (N = 20) | High (N = 30) | Low (N = 25) | High (N = 25) | |
| 5-year overall survival | ||||||||
| Crude | 1.00 (reference) | 2.45 (1.21–4.98; p = 0.013) | 3.09 (1.42–6.74; p = 0.005) | 1.37 (0.89–2.10; p = 0.152) | 3.15 (1.39–7.14; p = 0.006) | 1.43 (0.96–2.13; p = 0.083) | 2.63 (1.18–5.88; p = 0.018) | 1.51 (1.00-2.27; p = 0.049) |
| Adjustede | 1.00 (reference) | 2.80 (1.26–6.22; p = 0.011) | 6.59 (2.27–19.15; p < 0.001) | 1.71 (0.56–5.20; p = 0.346) | 8.37 (2.47–28.34;p < 0.001) | 1.76 (0.67–4.64; p = 0.255) | 5.66 (1.95–16.46; p = 0.001) | 2.11 (0.77–5.82; p = 0.148) |
| 5-year disease-specific survival | ||||||||
| Crude | 1.00 (reference) | 2.41 (1.05–5.55; p = 0.038) | 3.41 (1.39–8.35; p = 0.007) | 1.25 (0.73–2.12; p = 0.414) | 3.66 (1.44–9.28; p = 0.006) | 1.31 (0.80–2.14; p = 0.280) | 3.09 (1.24–7.69; p = 0.015) | 1.35 (0.81–2.24; p = 0.246) |
| Adjustede | 1.00 (reference) | 3.89 (1.45–10.47; p = 0.007) | 10.48 (2.90-37.95; p < 0.001) | 2.24 (0.44–11.31; p = 0.329) | 16.14 (3.43–75.98; p < 0.001) | 2.41 (0.68–8.56; p = 0.173) | 8.50 (2.48–29.18); p < 0.001) | 2.56 (0.66–9.95; p = 0.174) |
aTumour necrosis was classified as present or absent based on the identification of necrotic tumour areas showing nuclear shrinkage, fragmentation, cell loss, and ghost-like cell outlines
bThe average percentage method quantified necrosis as the proportion of necrotic tissue relative to the total tumour epithelial area. Low and high tumour necrosis percentage were defined using the median cut-off (5%)
cThe hotspot method assessed necrosis within a 2-mm circular region placed over the most necrotic area, excluding vessels and artefacts. Low and high tumour necrosis hotspot were defined using the median cut-off (100%)
dThe linear method measured the length of the single largest continuous necrotic area. Low and high tumour necrosis length were defined using the median cut-off (6000 μm)
eAdjusted for age (continuous), sex (female, male), Charlson Comorbidity Index (0–1, 2 or higher), cirrhosis (no, yes), Child-Pugh Score (A, B or C), TNM stage (1, 2 or higher), Barcelona Clinic Liver Cancer staging (0-A, B-D), tumour grade (1–2, 3) and year of surgery (1986–2005, 2006–2022)
P-values < 0.05 were considered statistically significant and are shown in bold
Tumour necrosis evaluation methods
Among the 50 samples with tumour necrosis, 25 patients (50%) were classified as having a high average necrosis percentage (≥ 5%.) and a high maximal length (≥ 6000 μm), while 30 patients (60%) had a high hotspot percentage (100%), reflecting a distribution skewed towards 100% for the hotspot approach. The evaluation of necrosis extent showed a strong correlation between the average percentage and hotspot method (rₛ=0.63, P < 0.001), the hotspot and linear method (rₛ=0.79, P < 0.001), and the average percentage and linear method (rₛ=0.85, P < 0.001). Baseline characteristics according to the different necrosis evaluation methods are provided in the supplementary material (Tables S2-S4).
As shown in the Kaplan–Meier curves, patients with low (but nonzero) necrosis had the worst 5-year survival in all three assessment methods, while those with higher necrosis levels showed intermediate outcomes (Fig. 3). In univariable and multivariable analysis, a low (but nonzero) necrosis level was associated with worse 5-year overall and disease-specific survival across all evaluation methods (Table 2). The hotspot method showed the strongest adjusted associations for the low necrosis category (< 100%) compared with absent necrosis, despite a ceiling effect in the high (100%) category (Table 2).
Fig. 3.
Kaplan–Meier survival for three tumour necrosis evaluation methods. Overall and disease-specific survival for the (A–B) average percentage method, (C–D) hotspot method, and (E–F) linear method. For the average percentage method, pairwise log-rank p-values for OS were: no vs. low necrosis, P = 0.003; no vs. high necrosis, P = 0.145; and low vs. high necrosis, P = 0.240. Corresponding pairwise p-values for DSS were: no vs. low necrosis, P = 0.004; no vs. high necrosis, P = 0.411; and low vs. high necrosis, P = 0.128. For the hotspot method, pairwise log-rank p-values for OS were: no vs. low necrosis, P = 0.004; no vs. high necrosis, P = 0.077; and low vs. high necrosis, P = 0.280. Corresponding pairwise p-values for DSS were: no vs. low necrosis, P = 0.003; no vs. high necrosis, P = 0.274; and low vs. high necrosis, P = 0.113. For the linear method, pairwise log-rank p-values for OS were: no vs. low necrosis, P = 0.014; no vs. high necrosis, P = 0.043; and low vs. high necrosis, P = 0.744. Corresponding pairwise p-values for DSS were: no vs. low necrosis, P = 0.011; no vs. high necrosis, P = 0.239; and low vs. high necrosis, P = 0.307
Discussion
In this study, tumour necrosis was associated with poorer long-term outcomes in resected HCC. Exploratory analyses further suggest that lower (but nonzero) levels of necrosis were more strongly associated with a worse prognosis when compared with extensive necrosis.
Several prior studies of resected HCC reported an association between tumour necrosis and adverse survival outcomes [5, 14–16], and tumour necrosis was also a poor prognostic marker in multiple solid tumours [9–13]. Most previous studies focused on microscopic tumour necrosis [5, 9–15], while macroscopic tumour necrosis [17, 18] and radiological findings suggestive of tumour necrosis [19] predicted worse prognosis in renal cell carcinoma. In addition to evaluation as a standalone variable, tumour necrosis was incorporated into prognostic scoring systems in renal cell carcinoma and HCC [10, 20, 21]. In our cohort, necrosis was present in 52.1% of cases, consistent with previous HCC series [5, 14–16, 20], and was associated with female sex, larger tumour size and poor tumour differentiation. The association with female sex was not consistently reported in earlier HCC studies [5, 14, 16] and may reflect cohort-specific differences.
The relationship between the extent of necrosis and long-term outcomes remains less well-defined [5, 12, 22–25]. In renal cancer [23], upper urinary tract transitional cell carcinoma [24] and colorectal cancer [25], a linear relationship between the extent of necrosis and mortality was observed. However, assessment techniques and cut-off values varied considerably across studies, limiting comparability [5, 12, 16, 22–25]. A colorectal cancer study evaluating multiple necrosis scoring approaches reported good prognostic performance and reproducibility across methods [9].
A previous Chinese study [5] reported a linear relationship between necrosis extent and mortality in HCC, whereas we observed the poorest outcomes in tumors with low, but nonzero, necrosis across all three assessment methods. That study evaluated only the lesion with the most severe necrosis [5], while we assessed all available tumour slides. In our cohort, the hotspot method provided the clearest prognostic stratification for the low necrosis category, although hotspot values showed a ceiling effect with many cases reaching 100%. Despite this limitation, the hotspot approach may be practical in routine assessment, because it uses a standardised region-of-interest and a simple scoring principle.
Tumour necrosis may reflect aggressive tumour biology associated with hypoxia and inflammatory signalling [26, 27], which could partly explain its association with adverse outcomes, although such mechanisms were not directly assessed in this study. The observation that low, but nonzero, necrosis was associated with poorer survival may reflect differences in tumour biology, whereby limited necrosis occurs in actively proliferating tumours, whereas more extensive necrosis represents later-stage ischemic damage with reduced growth potential.
Assessing histological tumour necrosis is straightforward and does not require special staining techniques. Based on existing evidence, tumour necrosis holds promise as a prognostic tool for resected HCC. Our findings suggest that the presence of tumour necrosis, rather than its extent, might be the preferred prognostic marker in HCC. With validation in larger prospective studies, this could potentially help identify patients at higher risk of adverse outcomes and thereby inform decisions regarding closer postoperative surveillance or consideration of adjuvant therapy. Further work, including interobserver agreement analyses and external validation, is warranted before recommending a single method of tumour necrosis quantification as a universal standard in HCC.
The extended study period presents limitations, as data availability from earlier phases of the cohort was limited. In addition, the relatively small cohort size restricts statistical power and warrants cautious interpretation of the findings. The higher resection rate in the later part of the cohort resulted in shorter follow-up times, further reducing the ability to detect associations. The retrospective study design constitutes an inherent limitation that may affect the robustness of the findings. Although key factors were adjusted for in multivariable analyses, residual confounding cannot be fully excluded. Furthermore, the clinical practice has evolved, causing temporal heterogeneity. While advances in perioperative management may have improved outcomes in later years, increasing resection rates of more comorbid patients later in the cohort could potentially counterbalance these improvements. Finally, while the use of a single investigator for sample assessment ensured methodological consistency, it also represents a limitation by precluding formal evaluation of interobserver variability.
Tumour necrosis was independently associated with poorer 5-year overall and disease-specific survival in patients undergoing resection for HCC. However, the extent of necrosis did not correlate linearly with survival, with lower (but nonzero) levels of necrosis associated with the poorest outcomes. These findings suggest a complex role of tumour necrosis in HCC and underscore the need for further investigation in larger, independent cohorts.
Supplementary Information
Authors’ contributions
All authors contributed to the study’s conception and design. Material preparation and data collection from patient charts were performed by Niklas Sarelin, Valtteri Kairaluoma and Olli Helminen. Juha Väyrynen and Jan Böhm assisted in picking out retrieving and digitising the histological samples. Niklas Sarelin performed the histological analysis. Juha Väyrynen helped in assessing uncertain cases. The first draft of the manuscript was written by Niklas Sarelin and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding
Open Access funding provided by University of Oulu (including Oulu University Hospital). Financial support was provided by the Mary and Georg C. Ehrnrooth Foundation, Vaasan Lääkäriyhdistys rf, and The Swedish Cultural Foundation in Finland.
Data availability
The datasets analysed during the current study are available from the corresponding author upon reasonable request. Sharing the data will require additional ethical approval.
Declarations
Ethics approval and consent to participate
The Ethical Committee of the North Ostrobothnia Hospital District (EETTMK 81/2008) approved the study. The study was carried out in accordance with the Declaration of Helsinki. The need to obtain informed consent from the study participants was waived by the Finnish Medicines Agency (FIMEA/2021/004928).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–49. [DOI] [PubMed] [Google Scholar]
- 2.Zhang X, et al. Long-term survival outcomes after curative local therapies for hepatocellular carcinoma in a large national cohort. Global Health Med. 2019;1(1):55–60. [Google Scholar]
- 3.Galle PR, Forner A, Llovet JM, Mazzaferro V, Piscaglia F, Raoul JL, et al. EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol. 2018;69(1):182–236. [DOI] [PubMed] [Google Scholar]
- 4.Singal AG, Yarchoan M, Yopp A, Sapisochin G, Pinato DJ, Pillai A. Neoadjuvant and adjuvant systemic therapy in HCC: Current status and the future. Hepatol Commun. 2024;8(6):e0430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wei T, Zhang XF, Bagante F, Ratti F, Marques HP, Silva S, et al. Tumor Necrosis Impacts Prognosis of Patients Undergoing Curative-Intent Hepatocellular Carcinoma. Ann Surg Oncol. 2021;28(2):797–805. [DOI] [PubMed] [Google Scholar]
- 6.Syrtveit AA, Ingebriktsen LM, Tegnander AF, Akslen LA, Wik E, Hoivik EA. Tumor necrosis associates with aggressive breast cancer features, increased hypoxia signaling and reduced patient survival. Sci Rep. 2025;15:45518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bredholt G, Mannelqvist M, Stefansson IM, Birkeland E, Hellem Bø T, Øyan AM, et al. Tumor necrosis is an important hallmark of aggressive endometrial cancer and associates with hypoxia, angiogenesis and inflammation responses. Oncotarget. 2015;6(37):39676–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Grivennikov SI, Greten FR. Inflammation and cancer: triggers, mechanisms and consequences. Immunity. 2019;51(1):27–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kastinen M, Sirniö P, Elomaa H, Ville K, Karjalainen H, Tapiainen V, et al. Establishing Criteria for Tumor Necrosis as Prognostic Indicator in Colorectal Cancer. Am J Surg Pathol. 2024;48(10):1284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Khor LY, Dhakal HP, Jia X, Reynolds JP, McKenney JK, Rini BI, et al. Tumor Necrosis Adds Prognostically Significant Information to Grade in Clear Cell Renal Cell Carcinoma: A Study of 842 Consecutive Cases From a Single Institution. Am J Surg Pathol. 2016;40(9):1224–31. [DOI] [PubMed] [Google Scholar]
- 11.Gilchrist KW, Gray R, Fowble B, Tormey DC, Taylor. SG 4th. Tumor necrosis is a prognostic predictor for early recurrence and death in lymph node-positive breast cancer. J Clin Oncol. 1993;11(10):1929–35. [DOI] [PubMed] [Google Scholar]
- 12.Swinson DEB, Jones JL, Richardson D, Cox G, Edwards JG, O’Byrne KJ. Tumour necrosis is an independent prognostic marker in non-small cell lung cancer. Lung Cancer. 2002;37(3):235–40. [DOI] [PubMed] [Google Scholar]
- 13.Hiraoka N, Ino Y, Sekine S, Tsuda H, Shimada K, Kosuge T, et al. Tumour necrosis is a postoperative prognostic marker for pancreatic cancer patients. Br J Cancer. 2010;103(7):1057–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Atanasov G, Dino K, Schierle K, Dietel C, Aust G, Pratschke J, et al. Angiogenic inflammation and formation of necrosis in the tumor microenvironment influence survival after radical surgery for de novo hepatocellular carcinoma in non-cirrhosis. World J Surg Oncol. 2019;17(1):139659377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ling YH, Chen JW, Wen SH, Huang CY, Li P, Lu LH, et al. Tumor necrosis as a poor prognostic predictor in solitary small HCC. BMC Cancer. 2020;20(1):607. [DOI] [PMC free article] [PubMed]
- 16.Wang Y, Ge H, Hu M, Pan C, Ye M, Yadav DK, et al. Histological tumor micronecrosis after R0 hepatectomy for HCC as a factor for adjuvant TACE. Int J Surg. 2022;105:106852. [DOI] [PubMed] [Google Scholar]
- 17.Pflanz S, Brookman-Amissah S, Roigas J, Kendel F, Hoschke B, May M. Impact of macroscopic tumour necrosis in renal cell carcinoma. Scand J Urol Nephrol. 2008;42(6):507–13. [DOI] [PubMed] [Google Scholar]
- 18.Lee SE, Byun SS, Oh JK, Lee SC, Chang IH, Choe G, et al. Significance of macroscopic tumor necrosis in renal cell carcinoma. J Urol. 2006;176(4):1332–8. [DOI] [PubMed] [Google Scholar]
- 19.Ooi GC, Sagar G, Lynch D, Arkell DG, Ryan PG. Cystic renal cell carcinoma: radiological and clinicopathological correlation. Clin Radiol. 1996;51(11):791–6. [DOI] [PubMed] [Google Scholar]
- 20.Yen YH, Kuo FY, Eng HL, Liu YW, Yong CC, Li WF, et al. Tumor necrosis as a predictor of early recurrence after resection in hepatoma. PLoS ONE. 2023;18(11):e0292144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Soini Y, Virkajärvi N, Lehto VP, Pääkkö P. Proliferation, apoptosis, and necrosis in hepatocellular carcinoma. Br J Cancer. 1996;73(9):1025–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pollheimer MJ, Kornprat P, Lindtner RA, Harbaum L, Schlemmer A, Rehak P, et al. Tumor necrosis as a prognostic factor in colorectal cancer. Hum Pathol. 2010;41(12):1749–57. [DOI] [PubMed] [Google Scholar]
- 23.Klatte T, Said JW, de Martino M, LaRochelle J, Shuch B, Rao JY, et al. Extent of tumour necrosis in renal cell carcinoma. J Urol. 2009;181(4):1558–64. [DOI] [PubMed] [Google Scholar]
- 24.Langner C, Hutterer G, Chromecki T, Leibl S, Rehak P, Zigeuner R. Tumor necrosis in upper urinary tract carcinoma. J Urol. 2006;176(3):910–4. [DOI] [PubMed] [Google Scholar]
- 25.Gao JF, Arbman G, Wadhra TI, Zhang H, Sun XF. Tumor inflammation and necrosis in colorectal cancer. World J Gastroenterol. 2005;11(14):2179–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Leek RD, Landers RJ, Harris AL, Lewis CE. Necrosis correlates with vascular density in breast carcinoma. Br J Cancer. 1999;79(5):991–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wykoff CC, Beasley NJ, Watson PH, Turner KJ, Pastorek J, Sibtain A, et al. Hypoxia-inducible expression of tumor-associated carbonic anhydrases. Cancer Res. 2000;60(24):7075–83. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets analysed during the current study are available from the corresponding author upon reasonable request. Sharing the data will require additional ethical approval.