Margins matter: raising the bar for thermal ablation through quantitative imaging validation in COVER-ALL

Thermal ablation is an established, curative-intent therapy for primary and metastatic liver tumors, offering comparable outcomes to surgical resection in well-selected patients when performed appropriately, while preserving parenchyma and reducing morbidity (1-3). Integrated into the Barcelona Clinic Liver Cancer (BCLC) guidelines for first-line treatment of early-stage hepatocellular carcinoma (HCC) in 2005 (4), randomized trials such as CLOCC (5) and COLLISION (6) have since provided strong evidence supporting its use for colorectal liver metastases. Alongside prospective multicenter studies like MAVERICC (7), ablation has been cemented within real-world modern multidisciplinary oncology practice.

While these trials represent major advances and will undoubtedly increase demand, the focus must now shift—from proving that ablation works, to understanding how to ablate most effectively. Unlike systemic therapies, ablation is not a uniform intervention. Success depends on the techniques used and the skill of the treating physician during all stages of the procedure, including planning, applicator placement, and treatment verification. The ultimate goal must be to make ablation behave as much like a drug as possible—standardized, reproducible, and scalable.

The Achilles’ heel remains local tumor progression (LTP), with heterogeneous outcomes reported across centers. Rates range from 5% with ≥5 mm margins in the COLLISION trial (6), to as high as 70% in earlier real-word studies (8,9), historically exceeding those of surgery (10,11). As in radiation oncology and surgery, margins matter (12-14)—and ablation, as a local therapy, is no exception. Hence, the most important predictor of local tumor control is the ablative margins (15-17), and one of the most pressing challenges now is how to best assess them.

The need for greater rigor in margin assessment is illustrated by a persistent contradiction in the ablation literature: many studies report high rates of “technical success”, yet disappointing LTP outcomes. If technical success has been achieved, why do tumors return so often? The answer lies in the subjectivity and ambiguity of visual assessment which unfortunately still dominates current clinical practice. Operators typically compare pre- and post-ablation scans side-by-side—an approach that is fundamentally flawed: subjective, unreliable, and prone to misjudgment (18). After a challenging procedure, there is a natural tendency to accept the result, to err on the side of reassurance. This human factor, familiar to anyone performing ablation, leads to hollow and biased conclusions.

Software-based quantification offers a solution. By incorporating image segmentation, registration, and fusion of pre- and post-ablation scans, these tools can calculate the minimal ablative margin (MAM)—defined as the shortest distance from the tumor boundary to the ablation zone edge. MAM is an objective, reproducible metric that anchors decisions in measurable reality. Retrospective studies have shown MAM offers superior performance versus visual assessment (18,19). However, no prospective randomized trial had evaluated the intraprocedural use of ablation confirmation software—until now.

COVER-ALL, a prospective single-center phase 2 trial led by Odisio et al. and published in The Lancet Gastroenterology & Hepatology, addresses this challenge directly (20). The trial randomized patients intra-procedurally to MAM assessment via side-by-side visual inspection of co-registered datasets versus software integrating deformable registration and AI-based segmentation. The software uses segmentation of target tumors and ablation zones on pre- and post- ablation images respectively, with image overlay (fusion), followed by quantitative assessment of MAM. This real-time feedback then guided the operator to re-ablate if margins were inadequate in the software group. The primary endpoint was MAM, with a pre-specified goal defined in the study protocol of ≥5 mm for each ablation, and secondary endpoints including workflow, safety, and LTP.

The results are compelling. The mean MAM in the software group was 5.9 versus 2.2 mm in the control group (P<0.0001), and margins >5 mm were achieved in only 15% of patients in the control arm versus 75% in the software arm. These findings prompted early termination of the control arm and expansion of the experimental cohort, where the margin improved further (mean 7.2 mm). Although not powered for oncologic endpoints, the 2-year cumulative incidence of LTP was numerically lower in the software group than in the control group (5% vs. 16%), without any increase in adverse events by 30 days, with a median software processing time of 3 minutes.

These findings position COVER-ALL as a landmark study, and implications go far beyond a single software platform. The authors tellingly refer to “an ablation confirmation methodology”, using the indefinite article “an”. Like any measurement, the validity of MAM depends entirely on how it is derived. Early studies proposed a 10 mm threshold using rudimentary methods (21), whereas more recent work suggests that margins of 5 mm or even 3 mm, may be sufficient (22,23). This is not a lowering of standards, but rather the refinement of methods (e.g., semi-automation and better registration for margin measurement, and more consistent needle placement using stereotactic or robotic systems). As such, without methodological standardization, MAM thresholds remain arbitrary, and comparisons between studies or centers lose meaning.

A panoply of questions follows. What is the image quality? Is image fusion rigid or deformable—and if deformable, are internal or external landmarks used? Should the pre- or post-ablation scan be used as the reference? Are vessels or liver surfaces used for alignment? Which imaging phase is used for segmentation—arterial or portal venous? Should segmentation, fusion, and MAM measurement be automated, semi-automated, or manual, and how is accuracy validated?

These are not merely academic details—they go to the heart of MAM’s credibility as a quantitative imaging biomarker. Like all imaging biomarkers, MAM must meet the criteria set out in the Imaging Biomarker Roadmap for Cancer Studies (24). These include biological plausibility, technical precision (i.e., low intra-observer, inter-observer, and inter-center variability), biological validity, clinical qualification, and cost-effectiveness. COVER-ALL contributes meaningfully to the translational development of MAM as a putative imaging biomarker—demonstrating feasibility, usability, and impact on intraprocedural decision-making, and addressing key pillars of the roadmap such as biological and clinical validation, repeatability, and clinical utility.

While COVER-ALL sets an important precedent, it remains a single-center, single-platform study. Encouragingly, several other platforms for ablative margin assessment are emerging, ranging from standalone software solutions to those integrated within stereotactic or robotic devices. Yet, for MAM to become a fully qualified imaging biomarker, validation must extend across platforms, operators, and institutions. Without shared benchmarks and open-access datasets, we risk a fragmented ecosystem of proprietary tools—lacking interoperability, comparability, and external validity. A potential model is the Quantitative Imaging Biomarkers Alliance (QIBA), which has developed consensus-driven technical profiles to promote standardization across imaging platforms (25). As a start, we recently convened the Innsbruck Consensus, which proposed an A0/A1/A2 classification based on margin adequacy—a pragmatic framework to interpret MAM results and support clinical decision-making across platforms, analogous to R0/R1/R2 status in surgical oncology.

Looking ahead, several challenges remain. Future studies must be adequately powered for oncologic outcomes, assess cost-effectiveness, and consider histology-specific differences in tumor biology and margin sensitivity—hence the importance of the forthcoming ACCLAIM (NCT05265169) and PROMETHEUS (NL9713) trials. Attention must be paid to the safety implications of pursuing wider margins, particularly the risk of delayed complications such as bilomas that may fall outside 30-day windows.

Beyond validation, COVER-ALL opens several critical questions. Can MAM serve as a surrogate endpoint in clinical trials? Can it be integrated into reimbursement, credentialing or quality assurance? As ablation technologies mature, so too must the methods we use to evaluate them—and our skill in using them effectively. Ablation and ablative margins must be treated with the same scientific rigor as drug development and clinical application. COVER-ALL is a step change in this regard—demonstrating not just feasibility but also the reproducibility and transparency expected in high-quality therapeutic trials. Until MAM is validated across platforms, tumor types, and outcomes—and meets criteria for cost-effectiveness and safety—it remains a promising but developmental imaging biomarker. As such, COVERALL is a critical foundation rather than a final answer.

The significance of COVER-ALL—and the broader promise of real-time margin quantification—extends beyond improved technical endpoints, heralding a cultural shift in how ablation is performed, supervised, and taught. By delivering immediate, objective feedback, ablation confirmation software could help standardize training, reduce reliance on intuitive judgement, and shorten the learning curve for junior operators while preserving safety and efficacy.

The path forward is not simply to perform more ablations, but to perform them better—and to know, with confidence, that we have. We commend Odisio and colleagues for raising the bar and demonstrating that how we ablate truly matters. COVER-ALL marks a pivotal step in the shift from intuitive practice to standardized, evidence-based therapy—bringing us closer to treating ablation more like a drug.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the Editorial Office, Translational Gastroenterology and Hepatology. The article has undergone external peer review.

Peer Review File: Available at https://tgh.amegroups.com/article/view/10.21037/tgh-25-83/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://tgh.amegroups.com/article/view/10.21037/tgh-25-83/coif). E.W.J. reports honorarium from Varian to speak at the British Society of Interventional Radiology, 2024, and honorarium from Philips to attend a round table discussion regarding CT hepatic angiography. R.B. received funding for scientific studies from Siemens Healthineers (Erlangen, Germany). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. van Amerongen MJ, Jenniskens SFM, van den Boezem PB, et al. Radiofrequency ablation compared to surgical resection for curative treatment of patients with colorectal liver metastases - a meta-analysis. HPB (Oxford) 2017;19:749-56. [Crossref] [PubMed]
2. Loveman E, Jones J, Clegg AJ, et al. The clinical effectiveness and cost-effectiveness of ablative therapies in the management of liver metastases: systematic review and economic evaluation. Health Technol Assess 2014;18:vii-viii, 1-283. [Crossref] [PubMed]
3. Gurusamy K, Corrigan N, Croft J, et al. Liver resection surgery versus thermal ablation for colorectal LiVer MetAstases (LAVA): study protocol for a randomised controlled trial. Trials 2018;19:105. [Crossref] [PubMed]
4. Llovet JM. Updated treatment approach to hepatocellular carcinoma. J Gastroenterol 2005;40:225-35. [Crossref] [PubMed]
5. Ruers T, Van Coevorden F, Punt CJ, et al. Local Treatment of Unresectable Colorectal Liver Metastases: Results of a Randomized Phase II Trial. J Natl Cancer Inst 2017;109:djx015. [Crossref] [PubMed]
6. van der Lei S, Puijk RS, Dijkstra M, et al. Thermal ablation versus surgical resection of small-size colorectal liver metastases (COLLISION): an international, randomised, controlled, phase 3 non-inferiority trial. Lancet Oncol 2025;26:187-99. [Crossref] [PubMed]
7. Tinguely P, Ruiter SJS, Engstrand J, et al. A prospective multicentre trial on survival after Microwave Ablation VErsus Resection for Resectable Colorectal liver metastases (MAVERRIC). Eur J Cancer 2023;187:65-76. [Crossref] [PubMed]
8. Chinnaratha MA, Chuang MY, Fraser RJ, et al. Percutaneous thermal ablation for primary hepatocellular carcinoma: A systematic review and meta-analysis. J Gastroenterol Hepatol 2016;31:294-301. [Crossref] [PubMed]
9. Gillams A, Goldberg N, Ahmed M, et al. Thermal ablation of colorectal liver metastases: a position paper by an international panel of ablation experts, The Interventional Oncology Sans Frontières meeting 2013. Eur Radiol 2015;25:3438-54. [Crossref] [PubMed]
10. Li L, Zhang J, Liu X, et al. Clinical outcomes of radiofrequency ablation and surgical resection for small hepatocellular carcinoma: a meta-analysis. J Gastroenterol Hepatol 2012;27:51-8. [Crossref] [PubMed]
11. Wu YZ, Li B, Wang T, et al. Radiofrequency ablation vs hepatic resection for solitary colorectal liver metastasis: a meta-analysis. World J Gastroenterol 2011;17:4143-8. [Crossref] [PubMed]
12. Vandeweyer D, Neo EL, Chen JW, et al. Influence of resection margin on survival in hepatic resections for colorectal liver metastases. HPB (Oxford) 2009;11:499-504. [Crossref] [PubMed]
13. Bodingbauer M, Tamandl D, Schmid K, et al. Size of surgical margin does not influence recurrence rates after curative liver resection for colorectal cancer liver metastases. Br J Surg 2007;94:1133-8. [Crossref] [PubMed]
14. Jones D. ICRU report 50—prescribing, recording and reporting photon beam therapy. Medical Physics 1994;21:833.
15. Laimer G, Schullian P, Jaschke N, et al. Minimal ablative margin (MAM) assessment with image fusion: an independent predictor for local tumor progression in hepatocellular carcinoma after stereotactic radiofrequency ablation. Eur Radiol 2020;30:2463-72. [Crossref] [PubMed]
16. Verdonschot KHM, Arts S, Van den Boezem PB, et al. Ablative margins in percutaneous thermal ablation of hepatic tumors: a systematic review. Expert Rev Anticancer Ther 2023;23:977-93. [Crossref] [PubMed]
17. Laimer G, Verdonschot KHM, Kopf L, et al. Multicenter and inter-software evaluation of ablative margins after thermal ablation of colorectal liver metastases. Eur Radiol 2025;35:1046-56. [Crossref] [PubMed]
18. Laimer G, Schullian P, Putzer D, et al. Can accurate evaluation of the treatment success after radiofrequency ablation of liver tumors be achieved by visual inspection alone? Results of a blinded assessment with 38 interventional oncologists. Int J Hyperthermia 2020;37:1362-7. [Crossref] [PubMed]
19. Shur JD, Porta N, Kafaei L, et al. Evaluation of Local Tumor Outcomes Following Microwave Ablation of Colorectal Liver Metastases Using CT Imaging: A Comparison of Visual versus Quantitative Methods. Radiol Imaging Cancer 2025;7:e230147. [Crossref] [PubMed]
20. Odisio BC, Albuquerque J, Lin YM, et al. Software-based versus visual assessment of the minimal ablative margin in patients with liver tumours undergoing percutaneous thermal ablation (COVER-ALL): a randomised phase 2 trial. Lancet Gastroenterol Hepatol 2025;10:442-51. [Crossref] [PubMed]
21. Wang X, Sofocleous CT, Erinjeri JP, et al. Margin size is an independent predictor of local tumor progression after ablation of colon cancer liver metastases. Cardiovasc Intervent Radiol 2013;36:166-75. [Crossref] [PubMed]
22. Laimer G, Jaschke N, Schullian P, et al. Volumetric assessment of the periablational safety margin after thermal ablation of colorectal liver metastases. Eur Radiol 2021;31:6489-99. [Crossref] [PubMed]
23. Chlorogiannis DD, Sotirchos VS, Georgiades C, et al. The Importance of Optimal Thermal Ablation Margins in Colorectal Liver Metastases: A Systematic Review and Meta-Analysis of 21 Studies. Cancers (Basel) 2023;15:5806. [Crossref] [PubMed]
24. O’Connor JP, Aboagye EO, Adams JE, et al. Imaging biomarker roadmap for cancer studies. Nat Rev Clin Oncol 2017;14:169-86. [Crossref] [PubMed]
25. Shukla-Dave A, Obuchowski NA, Chenevert TL, et al. Quantitative imaging biomarkers alliance (QIBA) recommendations for improved precision of DWI and DCE-MRI derived biomarkers in multicenter oncology trials. J Magn Reson Imaging 2019;49:e101-e121. [Crossref] [PubMed]