Short-term and long-term outcomes after robotic versus open hepatectomy in patients with large hepatocellular carcinoma: a multicenter study

Background: Robotic hepatectomy (RH) is currently widely accepted and it is associated with some benefits when compared to open hepatectomy (OH). However, whether such benefits can still be achieved for patients with large hepatocellular carcinoma (HCC) remain unclear. This study aimed to evaluate the short-term and long-term outcomes of patients undergoing RH or OH. Methods: Perioperative and survival data from patients with large HCC who underwent RH or OH between January 2010 and December 2020 were collected from eight centres. Propensity score matching (PSM) was performed to minimise potential biases. Results: Using predefined inclusion criteria, 797 patients who underwent OH and 309 patients who underwent RH were enroled in this study. After PSM, 280 patients in the robotic group had shorter operative time (median 181 vs. 201 min, P<0.001), lower estimated blood loss (median 200 vs. 400 ml, P<0.001), and shorter postoperative length of stay (median 6 vs. 9 days, P<0.001) than 465 patients in the open group. There were no significant differences between the two groups in overall survival and recurrence-free survival. Cox analysis showed AFP greater than 400 ng/ml, tumour size greater than 10 cm, and microvascular invasion were independent risk factors for overall survival and recurrence-free survival. After PSM, subgroup analysis showed that patients with a huge HCC (diameter >10 cm) who underwent RH had significantly lower estimated blood loss (median 200.0 vs. 500.0 min, P<0.001), and shorter length of stay (median 7 vs. 10 days, P<0.001) than those who underwent OH. Conclusion: Safety and feasibility of RH and OH for patients with large HCC were comparable. RH resulted in similar long-term survival outcomes as OH.


Introduction
Hepatocellular carcinoma (HCC) is a very common cancer in the world with a high rate of cancer-related mortality [1] .Hepatectomy is the main first-line treatment for HCC, which can offer patients the chance of long-term survival [2,3] .With advances in surgical techniques and knowledge related to liver surgery, tumour size is no longer considered to be a contraindication to hepatectomy.Many experienced centres have reported on hepatectomy for HCC larger than 5 cm, or even 10 cm [4,5] .Previous studies have classified liver tumours larger than 5 cm and 10 cm as 'large' or 'huge' HCC, respectively [6,7] .
The high recurrence rate of HCC after hepatectomy results in unsatisfactory long-term survival outcomes of these patients [8] .Tumour size is recognised to be a very important prognostic factor for survival and recurrence.It forms the basis of tumour staging systems, and at least to a certain extent, guide the treatment of HCC [6,9] .Furthermore, tumour size is also a potential risk factor affecting positive resection margins in hepatectomy [10] .Results for tumours that are larger than 5 cm, studies reported contradictory results.While some studies suggested that tumour size exceeding 5 cm to be a poor prognostic factor, others reported that there were no significant impact on long-term survival for patients with tumours larger than 5 cm [6,[11][12][13] .
Minimally invasive hepatectomy is now considered to be a safe and effective treatment for liver tumours in experienced hands [14] .In addition, it results in more rapid postoperative recovery, allowing patients to receive adjuvant oncological treatment without delay [15] .Rapid development of robotic hepatectomy (RH) has significantly changed the landscape of liver resection [16] .RH, including robotic donor hepatectomy, robotic major hepatectomy, and robotic minor hepatectomy, is increasingly used in many specialized centres [17][18][19][20][21] .The robotic system can provide better flexibility, ergonomics, and threedimensional magnified views of the surgical site [22,23] , contributing to lower conversion rates and shorter length of stay (LOS) in hospital [24] .Our centre is a high-volume centre with extensive experience in both RH and OH [20,[25][26][27] .Tumour size is known to be a factor impacting the difficulty of hepatectomy, especially in minimally invasive liver resection [28] , Furthermore, there is a lack of data on the short-term and long-term outcomes of RH in patients with large tumours compared with open hepatectomy (OH).
This multicenter study aimed to compare the short-term and long-term outcomes of RH and OH in patients with large HCCs in the different clinical subgroups to provide data to facilitate overall management of patients with large HCC.

Patients
This retrospective multicenter study was carried out on consecutive HCC patients with large HCCs who underwent RH or OH between January 2010 and December 2020 in eight centres.The inclusion criteria were patients: (1) with a resectable tumour 5 cm or greater which was histopathologically confirmed as HCC; (2) aged over 18 years; and (3) with no anaesthesia or surgical contraindications.The exclusion criteria were patients with: (1) other types of malignant tumours; (2) distant metastases; (3) unavailable perioperative data; (4) combined extrahepatic resection; and ( 4) intraoperative open conversion.The study was approved by all the Hospital's Institutional Review Boards.The ethics committee waived the informed consent requirement due to the anonymity of patient identities.This retrospective cohort study was registered with ResearchRegistry.com.The work was reported in accordance with the strengthening the reporting of cohort, cross-sectional and case-control studies in surgery (STROCSS) criteria [29] (Supplemental Digital Content 1, http://links.lww.com/JS9/B314).

Perioperative data
Guidelines on the diagnosis and treatment plan of HCC were designed by a multidisciplinary team and these were followed by each of the multiple centres.The data was collected from an electronic database and the data were subsequently analysed retrospectively.The data included perioperative baseline characteristics, pathology, and surgical outcomes.A tumour size of 5 cm or larger was considered to be a large HCC while a tumour size exceeding 10 cm was defined as a huge HCC [30][31][32] .Microvascular invasion was defined as presence of tumour emboli within the central vein, portal vein, or large capsular vessels or involvement of segmental or sectoral branches of portal or hepatic veins [33,34] .Morbidity grade was estimated using the Clavien-Dindo classification, and major complications were defined as a Clavien-Dindo grade ≥ 3 [35] .We used the BCLC staging system, which was widely used and had an updated version in 2022 [36] .The estimated blood loss was calculated based on the difference between the suction canister fluids and the abdominal irrigation fluids, plus the difference in weight between operative and dry gauze [37] .

Surgical procedures and follow-up
All RH procedures were performed by surgeons who had completed more than 30 robotic liver resections and had surpassed the associated learning curve [38] .Robotic surgical techniques, including patient positioning and robotic settings, have been described in our previous articles [25,39,40] .The follow-up strategy, including follow-up interval and content, was the same as in our previously reported study [8] .Overall survival (OS) was defined as the time interval from surgery to death or the last follow-up, and recurrence-free survival (RFS) was defined as the time from resection to the date of first diagnosis or the last follow-up for tumour recurrence.This study was censored on 30 June 2021.

Statistical analysis
Categorical variables were presented as numbers and percentages.Continuous variables were manifested as medians and interquartile ranges.Baseline, operative, and postoperative data comparison between the robotic and open groups were performed using the Mann-Whitney U test for continuous variables and the χ 2 test or Fisher's exact test for categorical variables.PSM was used to reduce selection bias between groups and subgroup analyses were performed based on tumour sizes to study the impact of surgical methods on selected patients.For PSM, the caliper width was set to a propensity score of 0.1 SD, and patients

HIGHLIGHTS
• This multicenter study aimed to evaluate the short-term and long-term outcomes of patients with large hepatocellular carcinoma undergoing robotic hepatectomy (RH) or open hepatectomy (OH).• Safety and feasibility of RH and OH for these patients were comparable.• RH resulted in similar long-term survival outcomes as OH.
were matched to controls in a 1:2 ratio.A standardized mean differences dot plot was used to display the results of balanced tests.Univariable and multivariable Cox proportional hazards models were used to analyse potential prognostic variables.Hazard ratios (HRs) and 95% CI were reported to measure the effects of potential prognostic variables.Survival analysis was calculated using Kaplan-Meier (K-M) analysis and compared using the log-rank test.A P < 0.05 was considered statistically significant.All statistical analyses were performed using SPSS software (version 22.0) and R software (version 4.1.1).

Patient characteristics
Of 797 patients who were included in the OH group and 309 patients who were included in the RH group; the huge HCC subgroup consisted of 347 patients, of which 299 underwent OH and 48 underwent RH.The patient characteristics in the unmatched cohorts differed significantly in AFP (P < 0.001), varices rates (11 vs. 4%, P < 0.001), tumour size (P < 0.001) and MVI positive rates (56 vs. 42%, P < 0.001).PSM was carried out to reduce selection bias.The matched results were not in an absolute ratio of 1:2 because there were no matched objects in some cases.A standardized dot plot of the mean differences for PSM is shown in Supplementary Figure 1 (Supplemental Digital Content 2, http://links.lww.com/JS9/B315).After PSM, the open and robotic groups featured 465 patients and 280 patients, respectively.There were no significant differences in the baseline characteristics between the two groups.The baseline characteristics of HCC patients with large tumours in the robotic and open groups before and after PSM are shown in Table 1.

Comparison of short-term outcomes between the RH and OH groups
The surgical outcomes of HCC patients with large tumours in the robotic and open groups before and after PSM are shown in Table 2.After PSM, there was no significant difference in total clamping time (median 21 vs. 26 min, P = 0.380).However, the robotic group had lower operative time (median 181 vs. 201 min, P < 0.001), lower estimated blood loss (median 200 vs. 400 ml, P < 0.001), decreased postoperative LOS (median 6 vs. 9 days, P < 0.001), and a lower major complication rate (2 vs. 6%, P = 0.014).Details on the short-term outcomes before and after PSM are shown in Table 2.

Comparison of long-term outcomes between the RH and OH groups
Patients in the robotic group and the open group had similar OS (P = 0.475) and RFS (P = 0.500) after PSM.The median OS was 64.4 months in the open group and 68.9 months in the robotic group.The 1-year, 3-year, and 5-year OS rates were 89.2, 68.9, and 53.2% in the open group, and 92.5, 71.9, and 55.9% in the robotic group, respectively.The median RFS was 20.0 months in the open group and 25.7 months in the robotic group.The 1-year, 3-year, and 5-year RFS rates were 59. 4

Univariable and multivariable Cox regression analyses on survival outcomes in all HCC patients after hepatectomy
Univariable and multivariable Cox regression analysis for OS of all the HCC patients in the study are shown in Table 3. AFP greater than 400 ng/ml (HR = 1.431,P < 0.001), HCC size greater than 10 cm (HR = 1.735,P < 0.001), and presence of MVI (HR = 1.194,P = 0.040) were identified as independent risk factors of OS.Univariable and multivariable Cox regression analyses of RFS for HCC patients with large tumours are shown in Supplementary Table 2 (Supplemental Digital Content 2, http://links.lww.com/JS9/B315).AFP greater than 400 ng/ml (HR = 1.393,P < 0.001), HCC size greater than 10 cm (HR = 1.381,P < 0.001), and presence of MVI (HR = 1.208,P = 0.011) were identified as independent risk factors of RFS in all HCC patients.The multivariable Cox regression analysis results are shown in Figure 2.

Subgroup analysis of patients with huge HCC (diameter > 10 cm)
The baseline characteristics of patients in the huge subgroup are shown in Supplementary For patients in the huge HCC subgroup, the robotic group had shorter operative time (median 220 vs. 250 min, P = 0.005), lower estimated blood loss (median 200 vs. 500 ml, P < 0.001) and shorter postoperative LOS (median 7 vs. 10 days, P < 0.001) when compared with the open group.However, unlike the shortterm prognosis of the large HCC subgroup, there was no significant difference in the major complication rate (P = 0.477).The details are shown in Table 4. On subgroup analysis, OS and RFS were similar for patients in the huge HCC subgroup who underwent the two different surgical approaches.K-M plots and details of the long-term outcomes in the subgroup analysis are shown in Figure 3 and Supplementary Table 4 (Supplemental Digital Content 2, http://links.lww.com/JS9/B315).

Discussion
Hepatectomy remains the best option for treatment of large liver cancers [32] .With development of the robotic surgical systems and optimisation of surgical techniques, RH is becoming increasingly deployed for such surgeries.As a considerable proportion of HCC patients have large tumours [6,7] , and a large tumour size increases the difficulty of hepatectomy, minimally invasive hepatectomy faces unique technical challenges including distortion of normal anatomy, compression of vessels, increased number of tumour-supplying vessels, and presence of tumour invasion [7] .When compared with OH, previous studies have indicated that minimally invasive hepatectomy to be associated with lower estimated blood loss and shorter hospital stay, and with similar safety and efficiency for patients with large HCCs [2,4,28,[41][42][43] .However, few studies have reported the shortterm and long-term outcomes of RH in patients with large tumours.It is still unclear as to whether these patients would benefit from robotic surgery.
In our study, the short-term and long-term outcomes of RH was evaluated to compare with OH in patients with large HCCs.This retrospective, multicenter study was used to study the surgical outcomes of RH versus OH, and PSM was carried out to reduce selection bias.Since the prognosis of large HCCs after hepatectomy is worse than that of smaller HCCs [6] , subgroup analysis was used to evaluate whether the short-term or long-term results differ significantly between these two groups in relation to tumour size.Our study demonstrated that RH had better shortterm outcomes, but with similar long-term outcomes when compared with OH.AFP greater than 400 ng/l, HCC size greater than 10 cm, and presence of MVI were found to be independent risk factors for OS and RFS.Subgroup analysis further demonstrated that RH showed better results for the RH group than the OH group in the huge HCC subgroup, implying that RH for these patients is worthy of consideration.
After PSM, the preoperative characteristics were similar in the two groups.The robotic group still showed significantly better short-term outcomes, including lower estimated blood loss and shorter postoperative LOS.These findings are consistent with the results reported in most previous studies.The decreased estimated blood loss may be related to the haemostatic effect of pneumoperitoneum and meticulous haemostasis under threedimensional visualisation.Furthermore, based on a previous   study, laparoscopic hepatectomy for large HCCs tended to have longer operative time than OH (297.5 vs. 205 min, P < 0.001) [5] .
Our study indicated that RH typically took less time to perform than laparotomic hepatectomy (181 vs. 201 min, P < 0.001), suggesting that the robotic systems can offer some benefits and can enable surgeons to carry out hepatectomy more efficiently and with better outcomes.Previous studies have identified tumour size and vascular invasion as independent predictors of survival for large HCCs [32,43] .Similar results were obtained in our present study.Furthermore, AFP levels, HCC size, and presence of MVI were shown to be independent risk factors for OS and RFS.Advanced HCCs with high AFP levels and presence of MVI have been shown to be associated with poor prognosis [44] .
Previous studies have shown a 5-year OS rate for large HCC to range from 32.3-77.9% [31,32,45,46].Our study showed that patients with large HCCs had a high recurrence rate.Thus, these patients require regular postoperative surveillance to provide early appropriate treatments when recurrence is detected.In our study, there was no significant difference between the robotic and open groups, suggesting that RH to be a good treatment option for patients with large HCC.Our results were consistent with the results obtained in previous studies in that the surgical approach (RH vs. OH) did not significantly affect OS and RFS in patients with large HCCs [5] .On subgroup analysis, RH was better in estimated blood loss, perioperative LOS, and decreased operative time.These outcomes indicated that RH to be safe and feasible to be used as a treatment option for patients with large HCCs.
The guidelines in different areas have some differences, and one of the main differences between Western and Eastern guidelines is the criteria for surgical resection [47][48][49] .The European Association for the Study of the Liver (EASL) Clinical Practice Guidelines state that only patients with a single lesion less than 5 cm in maximum diameter, or a maximum of three lesions with a maximum diameter of 3 cm each, are eligible for surgical resection.However, some Asian guidelines, such as Korean, Japanese, and Chinese guidelines, allow hepatectomy for patients with larger or multiple lesions, as long as they have well-preserved hepatic function, no main portal trunk invasion, and no extrahepatic spread.The differences between Western and Eastern guidelines for hepatectomy of HCC reflect the differences in the prevalence of HCC, local clinical practice, insurance system, and other availability of resources [50] .Therefore, the differences between Western and Eastern clinical practices may limit the use of our results.According to the guidelines, the treatment options for HCC depend on several factors, such as the size, number, and location of the lesions, the stage of the cancer, the liver function, the general health, and patients' preferences.The main aim of treatment is to achieve prolonged survival and improve quality of life.A recent study provides preliminary evidence for the safety and efficacy of combined therapy for huge HCC [51] .The treatment of large HCCs will involve multiple disciplines and become more personalised with the development of high-quality clinical studies.
This study has some limitations.First, this is a retrospective study with potential bias.Second, the majority of HCC patients in this study suffered from viral hepatitis B, the aetiological factor differs significantly from HCC patients in western countries.Therefore, our results need to be validated in western patient cohorts.Finally, the centres involved in this study are high-  Data are presented as n (%), and median (IQR).a because some cases could not simultaneously find effective matching objects, the matching result was not an absolute 1:2.HCC, hepatocellular carcinoma; IQR, interquartile range; LOS, length of stay; PSM, propensity score matching.
volume centres which may limit the use of our results to other centres.
In conclusion, this study demonstrated the feasibility and safety of RH in HCC patients with large tumours.RH showed comparable short-term and long-term outcomes when compared with OH.

Figure 1 .
Figure 1.Kaplan-Meier curves estimating OS and RFS of HCC patients after PSM.

Figure 2 .
Figure 2. Forest plots of multivariable analysis for OS and RFS for patients with large tumours.A. forest plot of multivariable analysis for OS; B. forest plot of multivariable analysis for RFS.

Table 1 (
, 37.1, and 25.6% in the open group, and 64.7, 38.8, and 26.0% in the robotic group, respectively.Detailed comparison of long-term outcomes between the robotic and open groups are shown in Figure1and Supplementary Supplemental Digital Content 2, http:// links.lww.com/JS9/B315).

Table 3 (
Supplemental Digital Content 2, http://links.lww.com/JS9/B315).There were 347 patients in this subgroup, with 299 patients in the open group, and 48 patients in the robotic group.After 1:2 PSM, 83 patients in the open group and 47 patients in the robotic group were matched.

Table 1
Baseline characteristics of HCC patients with large tumours in the robotic and open groups before and after PSM.

Table 2
Surgical outcomes of HCC patients with large tumours in the robotic and open groups before and after PSM.
Data are presented as n (%) or median (interquartile range).HCC, hepatocellular carcinoma; LOS, length of stay; PSM, propensity score matching.

Table 4
Surgical outcomes of HCC patients with huge tumours ( > 10 cm) in the robotic and open groups before and after PSM.