Prof Bruno C Odisio MD, Jessica Albuquerque MD, Yuan-Mao Lin MD, Brian M Anderson PhD, Caleb S O’Connor MS, Bastien Rigaud PhD, Maria Briones-Dimayuga MD, Prof Aaron K Jones PhD, Bryan M Fellman MS, Prof Steven Y Huang MD, Joshua Kuban MD, Zeyad A Metwalli MD, Rahul Sheth MD, Peiman Habibollahi MD, Milan Patel MD, Ketan Y Shah MD, Veronica L Cox MD, Prof HyunSeon C Kang ,
Van K Morris MD, Prof Scott Kopetz PhD, Prof Milind M Javle MD, Prof Ahmed Kaseb MD, Ching-Wei Tzeng MD, Hop-Tran Cao MD, Timothy Newhook MD, Prof Yun Shin Chun MD, Prof Jean-Nicolas Vauthey MD, Prof Sanjay Gupta MD, Iwan Paolucci PhD, Prof Kristy K Brock PhD
Source: https://www.sciencedirect.com/science/article/abs/pii/S246812532500024X
Summary
Background
Tumour coverage with an optimal minimal ablative margin is crucial for improving local control of liver tumours following thermal ablation. The minimal ablative margin has traditionally been assessed through visual inspection of co-registered CT images. However, rates of local tumour control after thermal ablation are highly variable with visual assessment. We aimed to assess the use of a novel software-based method for minimal ablative margin assessment that incorporates biomechanical deformable image registration and artificial intelligence (AI)-based autosegmentation.
Methods
The COVER-ALL randomised, phase 2, superiority trial was conducted at The University of Texas MD Anderson Cancer Center (Houston, TX, USA). Patients aged 18 years or older with up to three histology-agnostic liver tumours measuring 1–5 cm and undergoing CT-guided thermal ablation were enrolled. Thermal ablation was performed with the aim of achieving a minimal ablative margin of 5 mm or greater. Patients were randomly assigned (1:1) to the experimental group (software-based assessment) or the control group (visual assessment) by use of dynamic minimisation to balance covariates. Randomisation was performed intraprocedurally after placement of the ablation applicator. Assessment of oncological outcomes and adverse events were masked to treatment allocation. All analyses were conducted on an intention-to-treat basis. The primary endpoint was the minimal ablative margin on post-ablation intraprocedural CT. A preplanned interim analysis for superiority was done at 50% patient enrolment. Adverse events were recorded with the Common Terminology Criteria for Adverse Events. This trial is registered with ClinicalTrials.gov (NCT04083378), and recruitment is complete.
Findings
Patients were enrolled and treated with thermal ablation between June 15, 2020, and Oct 5, 2023. 26 patients were randomly assigned to the control group (mean age 58·1 [SD 14·8] years; 18 [69%] male and eight [31%] female; 11 [42%] colorectal cancer liver metastasis; median tumour diameter 1·7 cm [IQR 1·3–2·3]) and 24 to the experimental group (mean age 60·5 [14·4] years; 16 [67%] male and eight [33%] female; ten [42%] colorectal cancer liver metastasis; median tumour diameter 1·8 cm [1·5–2·5]). The interim analysis showed a mean minimal ablative margin of 2·2 mm (SD 2·8) in the control group and 5·9 mm (2·7) in the experimental group (p<0·0001), prompting halting of enrolment in the control group. A further 50 patients were enrolled to a non-randomised experimental group (mean age 56·5 [SD 11·7] years; 27 [54%] male and 23 [46%] female; 30 [60%] colorectal cancer liver metastasis; median tumour diameter 1·5 cm [IQR 1·3–2·2]); among these patients, the mean minimal ablative margin was 7·2 mm (SD 2·8). Grade 1–3 adverse events were reported in five (5%) of 100 patients: three (12%) of 26 in the control group and two (3%) of 74 in the experimental groups. No grade 4–5 adverse events or treatment-related deaths were reported.
Interpretation
Software-based assessment during CT-guided thermal ablation of liver tumours is safe and significantly improves the minimal ablative margin compared to visual assessment. Adoption of software-based assessment as a standard component of thermal ablation should be considered to achieve the intended minimal ablative margin.
Funding
US National Institutes of Health and US National Cancer Institute.
Introduction
Primary liver cancer is the sixth most frequently diagnosed malignancy and the third leading cause of cancer-related deaths. Its incidence and mortality rates are expected to increase by more than 50% in the next two decades.1 Additionally, secondary liver cancers are increasing in incidence and are considered the primary cause of cancer-related mortality for several cancer types, including colorectal cancer, which is one of the most frequent sources of liver metastases.2 This underscores the need for effective curative-intent local therapies for the treatment of liver tumours. Thermal ablation has gained widespread acceptance as an alternative to surgical resection in selected patients, given that it is a minimally invasive, cost-effective, and safe technique, offering a shorter recovery time while providing comparable oncological outcomes.3, 4, 5, 6, 7, 8
– Research in context –
Evidence before this study
The reported rates of local tumour progression after thermal ablation of primary and secondary liver tumours are highly variable, with recent large retrospective studies showing rates ranging from 12% to 54%. The factor that most strongly influences local tumour progression is the minimal ablative margin, which has traditionally been evaluated by simple visual assessment. There is an unmet need for tools that provide accurate feedback on the minimal ablative margin to aid intraprocedural decision making. We searched PubMed on June 3, 2024, for articles published between Jan 1, 2014, and Jan 8, 2024, using the terms (“margin” OR” MAM”) AND (“liver” OR “hepat”) AND “ablat” AND “software”. We found two meta-analyses of retrospective studies showing that the minimal ablative margin can be quantified with software-based methods and that the minimal ablative margin is strongly linked to the risk of local tumour progression. However, we found no prospective studies evaluating the role of intraprocedural use of software-based ablation confirmation methods. Therefore, there is an unmet need for a randomised, prospective evaluation of the impact of software-based ablation confirmation methods during thermal ablation of primary and secondary liver tumours.
Added value of this study
To the best of our knowledge, this is the first randomised clinical trial to evaluate the intraprocedural use of a dedicated software-based ablation confirmation method, and its impact on decision making and on the minimal ablative margin achieved during thermal ablation of liver tumours. Our results show that this method was safe and significantly improved the minimal ablative margin, supporting its role as a standard component of thermal ablation of liver tumours.
Implications of all the available evidence
Although retrospective studies have shown that the minimal ablative margin is strongly associated with local tumour progression following thermal ablation of liver tumours, the intraprocedural use of a software-based method and its direct impact on procedure safety and efficacy is unknown. The findings of this randomised clinical trial support the adoption of software-based assessment as a standard component of thermal ablation to achieve the intended minimal ablative margin and consequently improve local tumour control.
Despite the inherent advantages of thermal ablation in the treatment of liver tumours, it has historically been associated with higher rates of local tumour progression than surgical resection. Rates of local tumour progression in the literature are widely variable, ranging from 12% to 54% in large series.9, 10 These findings emphasise the need to standardise the thermal ablation technique and improve local tumour control. Several factors have been found to influence local tumour progression, including tumour size, number, and location, as well as RAS mutational status.9, 11, 12 Investigators have shown that achieving an optimal minimal ablative margin surrounding the tumour is the most relevant factor influencing local tumour control and the only one that is modifiable during the thermal ablation procedure.13, 14, 15, 16, 17, 18 The optimal minimal ablative margin, extrapolated from studies on hepatocellular carcinoma and colorectal liver metastases, ranges from 5 mm to 10 mm. A minimal ablative margin lower than 5 mm has been shown to have very high risk of local tumour progression, whereas a minimal ablative margin of 5 mm or greater is linked to a low risk of local tumour progression. Thus, a minimal ablative margin of 5 mm or greater is currently considered to be the optimal minimal ablative margin for thermal ablation.12, 14, 17, 18 Traditionally, the minimal ablative margin has been assessed by visually inspecting co-registered pre-ablation and post-ablation contrast-enhanced CT (CECT) images. However, this approach is time-consuming, prone to operator bias, often results in inaccurate assessment due to interoperator variability, and could explain the currently reported high rates of local tumour progression.16, 17, 18, 19 Methods of assessing the minimal ablative margin during thermal ablation that are time-efficient, accurate, and less prone to operator bias are currently an unmet need. Moreover, at the time of writing, none of the currently available software-based ablation confirmation methods have been prospectively validated.
To overcome these limitations, we created a software-based assessment tool that combines biomechanical deformable image registration with ablation-specific artificial intelligence (AI)-based autosegmentation algorithms.20, 21 This approach has been retrospectively validated with both single-institution and multi-institutional datasets, demonstrating its ability to quantify and correlate the minimal ablative margin with local tumour progression following liver tumour thermal ablation.17, 22 We aimed to assess the effectiveness of this novel software-based assessment tool in a prospective clinical trial (COVER-ALL). The overall hypothesis of the COVER-ALL trial was that the intraprocedural use of the proposed software-based assessment will significantly increase the minimal ablative margin following thermal ablation of liver tumours.
Section snippets
Study design and patients
COVER-ALL was a randomised, phase 2, superiority trial conducted at The University of Texas MD Anderson Cancer Center (Houston, TX, USA). Eligible participants were individuals aged 18 years or older with an Eastern Cooperative Oncology Group performance status of 0–2 who had up to three primary or secondary liver tumours of any histological subtype measuring 1–5 cm and who were referred for percutaneous thermal ablation per standard of care. To avoid within-patient correlation for patients
Results
Between June 15, 2020, and Oct 5, 2023, 100 of 117 screened patients were enrolled (figure 1). Mean age was 57·8 (SD 13·2) years, 61 (61%) patients were male, 51 (51%) had colorectal cancer liver metastases, and the median tumour diameter was 1·6 cm (IQR 1·3–2·3). Demographics and baseline clinical characteristics were balanced between the randomised groups (table 1). Detailed adjunctive techniques and tumour histological subtypes are provided in the appendix (pp 5, 6). Technical success of
Discussion
This is, to the best of our knowledge, the first randomised trial to establish the feasibility and efficacy of software-based assessment as an intraprocedural tool for optimising the minimal ablative margin during thermal ablation of liver tumours. Specifically, in patients with a variety of subtypes of liver tumours, the mean minimal ablative margin was 2·2 mm (SD 2·8) with visual assessment, versus 5·9 mm (2·7) with the use of our proposed software-based assessment (p<0·0001). Notably, use of
Declaration of interests
KKB reports grants from the US National Institutes of Health and Raysearch Laboratories; a licensing agreement with RaySearch Laboratories; travel support from the American Association of Physicists in Medicine and Raysearch Laboratories; and service on the clinical advisory board of Raysearch Laboratories. BCO reports research grants from the National Institutes of Health, Siemens Healthineers, and Johnson & Johnson. All other authors declare no competing interests.
Acknowledgments
We thank all the patients participating in this study and their families. We thank Mara Castaneda and Barom Aromin (researcher nurses) for their support. This study was funded by the US National Institutes of Health and the US National Cancer Institute (grants 1R01CA235564 and P30CA016672). Editorial support was provided by Stephanie Deming, Research Medical Library, MD Anderson Cancer Center.