Safety and efficacy of stereotactic radiofrequency ablation for very large (≥8 cm) primary and metastatic liver tumors

To assess the safety and clinical outcomes of multi-probe stereotactic radiofrequency ablation (SRFA) for very large (≥8 cm) primary and metastatic liver tumors with curative treatment intent. A retrospective, single center study carried out between 01.2005 and 06.2018. 34 consecutive patients had a total of 41 primary and metastatic liver tumors with a median size of 9.0 cm (8.0–18.0 cm) at initial SRFA. Patients were treated under CT guidance using a 3D navigation system. Endpoints consisted of (i) technical efficacy; primary - requiring one treatment, and secondary – requiring two treatments (ii) complication and mortality rates (iii) local and distant recurrence (LR), (iv) disease free survival (DFS), (v) overall survival (OS). 33/41 tumors were successfully ablated at initial SRFA (80.5% primary technical efficacy rate (PTE)). Four tumors required repeat ablation, resulting in a secondary technical efficacy (STE) rate of 90.2%. Local tumor recurrence (LR) developed in 4 of 41 tumors (9.8%). The 30-day perioperative mortality was 2.3% (1/ 44 ablations). The total major complication rate was 20.5% (9 of 44 ablations). Three of nine (33.3%) major complications, such as pleural effusion, pneumothoraces or perihepatic hemorrhages were relatively easy to treat. The overall survival (OS) rates at 1-, 3-, and 5- years from the date of the first SRFA were 87.1%, 71.8%, and 62.8% for patients with hepatocellular carcinoma (HCC) and 87.5%, 70.0% and 70.0% for patients with intrahepatic cholangiocarcinoma (ICC) respectively. Patients with metastatic disease had OS rates of 77.8% and 22.2% at 1- and 3- years. The clinical results of SRFA in this study are encouraging and warrant a prospective multicenter study. SRFA may become one of the best therapeutic choices for a growing number of patients with primary and metastatic liver cancer.

Radiofrequency (RF) ablation has been increasingly accepted as a curative and cost-effective treatment for small liver tumors 1,2 . Whilst hepatic resection (HR) is still the preferred treatment in patients with preserved liver function, in clinical practice less than one-third of the patients are eligible for HR at diagnosis 3,4 . Historically, the reported local recurrence rates after conventional RF ablation are relatively high and vary from 10% to 39.1% by 5 years, depending on tumor size and number 5,6 . It has been shown that an ablation zone 'safety margin' is required around tumors by at least 5 mm to achieve local control and good clinical outcome 7 . However, achieving adequate treatment margins may be challenging in large and irregularly shaped tumors or in situations where the target tumor is either difficult to visualize, awkward to access, or adjacent to vulnerable structures. Similar considerations apply to HR, with larger tumors more likely to be unresectable due to small future liver remnant and close relationship with the major vessels or hepatic hilum 8 .
Transarterial chemoembolization (TACE) is currently the recommended treatment method in patients with impaired liver function and large (>5 cm) hepatocellular carcinoma (HCC) 9,10 and whilst a survival benefit has been demonstrated 10,11 , it is considered a palliative treatment 12 .
To overcome limitations in RF ablation zone size, other thermal ablation strategies have been developed including use of microwave (MW) energy and placement of multiple RF electrodes [13][14][15][16] . However, conventional multi-probe RF ablation using ultrasound or CT-fluoroscopy guidance lacks reliability for accurate three-dimensional RF probe placement and thus complete tumor coverage. Stereotaxy has proven useful for planning and executing complex or challenging ablations where access routes can be specifically facilitated and more precise coverage of the target tumor and safety margins can be accomplished 17,18 .
We have previously shown that SRFA is a viable treatment option in HCC 19 , although its role in very large lesions remains undefined.
The purpose of the current study was to assess the safety and clinical outcomes of multi-probe SRFA for very large (≥8 cm) primary and metastatic liver tumors, treated with curative intent. Specifically, our endpoints are primary (requiring one treatment) and secondary (requiring two treatments) technical efficacy, complication rates, recurrence, disease free and overall survival.

Materials and Methods
Selection criteria. The Institutional Review Board of the Medical University of Innsbruck approved this single-center, retrospective study. All participants gave their informed consent to collect their data. All treatment plans were consensually agreed in multidisciplinary tumor board meetings.
Between 01.2005 and 06.2018, 943 consecutive patients were treated by SRFA. Seventy-seven patients with portal venous invasion, extensive tumor with subsequent palliative intention to treat, or benign liver tumors were excluded (Fig. 1). Thirty-four patients with tumors ≥8 cm at initial SRFA were included. Baseline characteristics of included patients are shown in Table 1.
A platelet count of <50000/mm 3 and prothrombin activity <50% were exclusion criteria for SRFA. Tumor diagnosis was confirmed by typical imaging appearances on multiphasic contrast MRI or CT, with histopathological confirmation before or during SRFA procedure.
Multi-probe stereotactic radiofrequency ablation -procedure. The basic principle of multi-probe RF ablation is the simultaneous usage of multiple RF probes to overcome ablation size limitation of single probe techniques by creating multiple overlapping ablation zones (Fig. 2). "Stereotaxy" derives from two Greek roots -"stereos" meaning solid, and "taksis" meaning arrangement. Stereotactic technique relates the position of targets and entrance points within the body to a Cartesian coordinate system.
The method of SRFA has previously been reported in detail [20][21][22] . In brief, the whole treatment is performed under general anesthesia with muscle paralysis. Patients are immobilized on the CT-table by a single (Bluebag, Medical Intelligence Schwabmünchen, Germany) or double vacuum fixation technique (BodyFix, Medical Intelligence Schwabmünchen, Germany). For image-to-patient registration, 10-15 fiducials, (X-SPOT, Beekley Corporation, Bristol, CT, USA) are broadly attached to the skin of the thorax and upper abdomen. A contrast-enhanced planning CT (Siemens SOMATOM Sensation Open, sliding gantry with 82 cm diameter, Siemens AG, Erlangen, Germany) is obtained with 3 mm slice thickness. To enable precise stereotactic conditions, the endotracheal tube (ETT) is temporarily disconnected during the planning CT, during each stereotactic needle placement and for the final control CT. The CT dataset is transferred to an optical-based navigation system     www.nature.com/scientificreports www.nature.com/scientificreports/ follow-up contrast-enhanced CT or MR scans performed at 1-month. The imaging tests were evaluated by two board certified abdominal radiologists by consensus. (II). Considering the complexity of the procedures and for better comparison to surgery, the study endpoint major complications were defined according to the Clavien-Dindo classification (Grade III+) 23 . Mortality was defined as death within 30 days after SRFA treatment. (III). Local recurrence was specified as appearance of new enhancing nodules (CT of MR scans at 3-6 months intervals) within or directly adjacent to the initially tumor-free ablation zone. New enhancing nodules distant to the ablation zone and or to the initial tumor location were specified as distant tumor recurrence. (IV). (IV, V) Survival was calculated from the date of initial SRFA to the date of death attributable to malignancy or other causes (i.e., event) or to the most recent follow-up visit (i.e., censoring).

Statistical analysis. The statistical analysis was performed with IBM SPSS version 24 (IBM, Armonk, New
York). The Kolmogorov-Smirnov-Test, tested normality of distribution. Data were expressed as total numbers, median and range. Survival data (OS and DFS) were analyzed with the Kaplan Meier method. The X² test was used to determine differences between categorical variables, and the Mann-Whitney U test between independent continuous variables. A two-tailed p-value of <0.05 was considered as statistically significant.
Ethical approval. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent. Informed consent was obtained from all individual participants included in the study.

Consent for publication. Consent for publication was obtained for every individual person's data included
in the study.
One patient developed acute liver failure after treatment of 3 HCCs (largest tumor 8 cm) requiring salvage liver transplantation. Thermal injuries of the bowel (1 patient) and gallbladder (1patient) had to be surgically treated. One liver abscess required CT guided drainage. Other complications included transient liver failure (1 patient) and pleural effusions (3 patients) requiring thoracoenteses respectively. Additionally, two perihepatic hemorrhages and one pneumothorax were treated within the same session by embolization and chest drain which did not influence the postoperative course. 3/9 (33%) major complications were successfully treated by the interventional radiologist within the next days by placing a pleural drainage in case of major pleural effusions (3 patients).   Technical Success. SRFA was successfully completed according to plan in all 41 tumors (technical success rate 100%) Table 3. 33/41 tumors were successfully ablated at initial SRFA (80.5% primary technical efficacy rate). Two very large tumors with 18 cm and 10 cm in diameter required three and two ablation sessions, respectively. All other tumors were treated in one session. After the first follow-up, 4 tumors were retreated, resulting in a secondary technical efficacy rate of 90.2%. 6-31 (median 12) RF electrodes were inserted in each tumor.
A sub analysis based on tumor type (HCC, ICC and Metastasis) showed the highest PTE rates for HCC tumors with 88.2% (15/17) and the lowest for ICC with 75% (9/12). However, the differences between groups were non-significant (p = 0.607).

Rate Study Group
There was no significant difference between groups based on tumor entity (p = 0.72).

Discussion
The main finding of our study is that SRFA is a feasible and effective treatment strategy for tumors >8 cm. To the best of our knowledge, there are no reports available dealing with thermal ablation of this tumor size, partially because HR remains the recommended treatment in large tumors with preserved liver function. However, the minority of patients are suitable for resection 3,4 , meaning other curative treatment options are required.
Since RFA results in complete cure for small HCCs (<3 cm) and colorectal liver metastases (CRLM) in greater than 90% of cases [24][25][26] , it has been incorporated into international guidelines for small tumors 12,[27][28][29] . However, early studies demonstrated that complete cure falls to 40-70% 30,31 for medium HCCs (up to 5 cm) and to 23-45% for large HCCs 31,32 respectively, which is very likely to reflect the size limitations of RF ablation zones and the difficulties in producing overlapping ablations. To overcome size limitations, two main strategies were implemented. Firstly, increasing energy transfer in the form of microwave energy, and secondly using multiple RF probes to create overlapping ablation zones. More recent studies have consequently shown better PTE rates up to 86-97% in medium and large HCCs up to 10 cm 15,33,34 .
Whilst MWA can produce larger ablation zones faster and with less 'heat-sink' effect (i.e., large vessels reduce local tissue heating due to cooling) than RFA 35 , this higher thermal energy might injure adjacent vulnerable structures or create unnecessarily large ablation zones 36 . Whilst the lower energy transferred in RFA may result in more predictable ablation zones, the greater number of probes required considerably increases the complexity and difficulty of the procedure.
Accurate three-dimensional probe alignment to create sufficiently overlapping ablation zones demands a high level of operator experience using conventional US-and CT-guidance. However, SRFA offers three-dimensional ablation planning to achieve an optimal configuration of RF probes and create multiple overlapping coagulation volumes. Usage of an aiming device and triggering of respiratory motion 37 facilitate precise path needle and probe placement 17 . Immediate post ablation contrast-enhanced CT fusion with the planning CT allows for rapid, reliable intraoperative judgment of the ablation results with the option for re-ablation within the same session. In a recent histopathological validation study in explanted livers after bridging therapy with SRFA, we found no viable tumor in 183 of 188 treated lesions (97.3%) and in 50 of 52 lesions >3 cm (96.2%).
Despite our inclusion of tumors up to 18 cm in diameter in the present study, LR developed in only 4 of 41 tumors (9.8%). In contrast, best results from the US-or CT-guided multiprobe thermal ablation literature with 'freehand' targeting quote LR rates of 15.9-19.7% after thermal ablation of large HCCs (largest tumors 10 cm, mean 3-5.7 cm) and other large liver tumors up to 8 cm (mean tumor size 2.8 cm) 33,34,38 .
In the present study, distant tumor recurrence was found in 70.6% of patients and median DFS rates were 4.4-7.1 months. These unfavourable prognostic indicators despite high levels of effective local tumor control (90.8%) might be explained by large lesions having a greater chance of vascular invasion 46 and micrometastases 47 , and are line with the surgical literature at 57.1-82.4% 43,44 .
Our major complication rate of 20.5% is higher than in the study of Ma et al. 34 and Laeseke et al. 32 who reported a major complication rate of 16% following RFA for HCC >5 cm. One of the main reasons for that could be the high number of needles used (median 12 per tumor). However, our mortality rate compares favorably to Filmann et al. 48 who reported a mortality of 9.1-25.5% after major HR in large outcome analysis of 110 332 procedures. Zhou et al. 44 also published a systematic review after HR of large HCC (>10) a median perioperative morbidity of 29.2% (13.6-72%) and mortality of 3.5% (0-18.2%) and Nomit et al. 45 reported a major complication rate of 50.0% after laparoscopic HR of large CRLM.
Limitations. Study limitations include the retrospective design, heterogeneity of previous unsuccessful loco(regional) treatments (TACE or HR), and single treatment center bias. Comparisons with previous related studies are limited as stereotactic navigation systems were not employed in prior reports.
In summary, although all patients in this study were considered to be inoperable, we report lower mortality rates, lower complication rates and similar survival to the curative literature meaning SRFA may represent an important tool in the treatment of very large liver tumors.