Introduction

Hepatocellular carcinoma (HCC) is a major global health burden contributing notably to the worldwide cancer-related mortality [1, 2]. While systemic and interventional therapies, e.g., trans-arterial chemoembolization (TACE) or radiofrequency ablation (RFA), are the main options in advanced tumor stages, liver resection remains the gold standard in earlier stages with preserved liver function [3]. Proper patient selection and the implantation of modern liver function assessment as well as minimal invasive liver resection did further allow to widen the patient population eligible for surgery improving outcome in individuals formerly treated by TACE or local ablative procedure [4,5,6,7]. While liver transplantation remains the treatment of choice in terms of recurrence for localized HCC, strict allocation rules and the limited availability of donor grafts preclude transplantation for a large proportion of HCC patients [8]. Therefore, liver resection is becoming increasingly popular across the whole spectrum of localized HCC requiring medical resources, e.g., surgical as well as intensive care capacity to conduct surgery in these patients [9, 10].

During the last 2 years, the global health systems have shifted resources to encounter the COVID-19 pandemic. Thus, curative intention surgery in oncological patients was frequently delayed and the corresponding impact on clinical outcome investigated [11]. Reduced overall survival (OS) of patients with different malignant diseases due to delayed time to surgery (TTS) in the scenario of surgical, systemic (adjuvant, neoadjuvant), and radiotherapy has been described [12]. Interestingly, in a recent international study of colorectal cancer patients, neither poorer outcomes nor compromised resectability was observed after a treatment delay during the COVID pandemic [13]. However, the role of TTS in the oncological outcome of HCC patients remains to be elucidated. Thus, the aim of this study was to investigate the impact of TTS on short- and long-term outcome in HCC patients.

Material and methods

Patients

Between 2009 and 2021, 240 patients underwent liver resection for HCC at the University Hospital RWTH Aachen (UH-RWTH). Of these, patients who underwent any neoadjuvant therapy (n=15) were treated as emergency cases due to active bleeding (n=3) and those who had no images in the radiological archives (n=5) were excluded from the study. As such, two hundred seventeen (n=217) patients were eligible for the TTS analysis (Supplementary Figure 1). All patients underwent detailed, internationally accepted staging. Therefore, only patients with localized HCC without distant metastasis were analyzed. The study was conducted at the UH-RWTH in accordance with the requirements of the Institutional Review Board of the RWTH-Aachen University (EK 22-342), the current version of the Declaration of Helsinki, and the good clinical practice guidelines (ICH-GCP).

Study definitions

TTS was calculated as the date difference from the date of diagnosis to the date of surgery. The date of diagnosis was set as the date of the first contrast-enhanced ultrasound, magnetic resonance imaging (MRI), or computed tomography (CT) indicating the presence of HCC. Imaging data of our center as well as external referring hospitals were used for this analysis depending on first hospital site of diagnosis. All imaging modalities were evaluated for diagnostic quality by a senior radiologist (PB).

Staging and surgical technique

Preoperatively, all patients were evaluated for general performance, operability, and liver function as previously described [14, 15]. Standard staging procedures were carried out by means of MRI or CT to define tumor burden and exclude distant metastasis. American society of anesthesiologists (ASA) and Eastern Cooperative Oncology Group (ECOG) performance status were used to evaluate the physical status of patients. Liver function was evaluated by standard laboratory parameters and the LiMAx test (Humedics® GmbH, Berlin, Germany) [7].

Patients staged Barcelona Clinic Liver Cancer (BCLC) A to C without signs of extrahepatic tumor burden and preserved liver function were considered to be surgical candidates and discussed within the institutional interdisciplinary tumor board. The indication for hepatectomy was finally made by an experienced hepatobiliary surgeon. In cases of HCC recurrence, operative resection was discussed within an interdisciplinary tumor board in cases with localized disease evaluating ECOG status, tumor morphology, and residual liver function. Patients considered no surgical candidates were referred to interventional therapies (TACE, RFA, microwave ablation, stereotactic radiation), systemic therapy, or best supportive care with respect to common international guidelines [9, 10].

Liver resection was carried out as described previously and carried out in accordance with department-specific surgical standards in every case [14, 15]. Intraoperatively, ultrasound was used to visualize tumor spread and exclude additional suspect lesions. For transection of liver parenchyma in open surgery, the Cavitron Ultrasonic Surgical Aspirator (CUSA®, Integra LifeSciences®, Plainsboro NJ, USA) was used. To avoid blood loss, low central venous pressure was maintained during transection and intermittent Pringle maneuvers were used if necessary. For parenchymal transection in laparoscopic hepatectomy, Thunderbeat ® (Olympus K.K., Tokyo, Japan), Harmonic Ace ® (Ethicon Inc. Somerville, NJ, USA), or laparoscopic CUSA (Integra life sciences, New Jersey, USA) in combination with vascular staplers (Echelon, Ethicon, Somerville, New Jersey, USA) or polymer clips (Teleflex Inc., Pennsylvania, USA) were preferably used.

Statistical analysis

The primary objective of this study was to investigate the oncological effect of TTS on OS and recurrence-free survival (RFS) in HCC patients undergoing surgical resection. OS was defined as the period from date of liver resection to the date of death from any cause or date of the last contact if the patient was alive. RFS was defined as the period from liver resection to the date of recurrence. Patients with no tumor recurrence were censored at date of death or at the last follow-up for RFS analysis. For group comparison, patient subgroups with respect to TTS were generated (1–30 days, 31–60 days, 61–90 days, and over 90 days). Chi-square test was used to compare categorial data, expressed with number and percentage. Continuous variables were expressed as median and interquartile range and compared by Kruskal-Wallis test. A p value <0.05 was considered to indicate statistical significance. Kaplan-Meier analysis was used to generate survival curves. Univariate cox regression was to determine variables associated with OS and RFS. Significant parameters (p<0.05) were proceeded to a multivariable cox regression model and analyzed within a backward selection. Median follow-up was assessed with the reverse Kaplan-Meier method. Complications are reported using the Clavien-Dindo scale [16]. Perioperative mortality (Clavien-Dindo V) was defined as in-hospital mortality. All data processing was conducted by SPSS Statistics 24 (IBM Corp., Armonk, NY, USA).

Results

Patient cohort

A total of 217 patients underwent liver resection for HCC in curative intention from 2009 to 2021 at our institution were included in this study. Most of the patients were male (71.4%), the median age in the overall cohort was 69 years. A major part of the cohort (65%) displayed an ASA score of III and more. Alcohol-induced (23.5%) and non-alcoholic fatty liver disease (26.9%) along with viral induced hepatitis (24.9%) were the main underlying liver diseases in the cohort; a subset of patients (14.7%) presented with either cryptogenic or a less common liver disease (e.g., hemochromatosis). The largest proportion of the cohort (56.7%) was BCLC stage A at time of resection, whereas a subset of patients was categorized CHILD Pugh B (8.3%). The median number of HCC nodules was 1 (interquartile range: 1–2) with a median diameter of 50 mm (interquartile range: 33–80) of the largest lesion. Also, a notable proportion of patients (25.8%) showed macrovascular invasion in preoperative imaging. The median operative time was 204 min (interquartile range: 137–270) and the most common operative procedure was atypical liver resection (37.3%), followed by left/right hepatectomy (22.6%). Red blood cells (24.4%) and fresh frozen plasma (FFP) (36.4%) were administered intraoperatively on demand. R0 resection was achieved in most cases (94.5%; reasons for R1 resection presented in Supplementary Table 1). Of all individuals, 24.5% experienced complications Clavien-Dindo > II and 5.1% of the cohort deceased during hospitalization (reasons for perioperative mortality presented in Supplementary Table 2). Detailed perioperative characteristics are depicted in Table 1.

Table 1 Study cohort
Full size table

Time-to-surgery with respect to different characteristics

Interestingly, the median TTS in the overall cohort was 49 days (interquartile range (IQR): 30–83). No statistical difference in TTS between patients diagnosed in our center (21.2%, 54 days (IQR: 35–84)) and externally diagnosed patients (78.8%, 47 days (IQR: 27–79)) has been found (p=0.15). Patients treated during the COVID period from year 2020 to 2021 (27.2%) had a median TTS of 70 days (IQR: 42–90), resulting to a statistically significant longer TTS than patients treated before 2020 (72.8%, 46 days (IQR: 24–73)) (p<0.001).

Comparative analysis of the patient cohort

Categorized by time to surgery, 55 patients underwent liver resection within 30 days after diagnosis, 79 patients between 31 and 60 days, 45 between 61 and 90 days, and 38 patients after 90 days. Extensive group comparisons revealed no differences in major demographic and oncological characteristics. Differences were observed in gender (p=0.020) and largest tumor diameter (p=0.020) with this difference being based on larger tumors in TTS 1–30 days group compared to TTS 61–90 days (p=0.004) and TTS > 90 days (p=0.015) group. Furthermore, the distribution of laparoscopic resections differed significantly between the subgroups (p=0.001). Other examined parameters showed no statistical differences in distribution, detailed perioperative results for the 4 subgroups are described in Table 2.

Table 2 Comparative analysis of patients undergoing liver resection for hepatocellular carcinoma
Full size table

Survival analysis

After a median follow-up of 59 months, the median OS of the cohort was 42 months (95% CI: 30–54 months; 3-year OS=58%, 5-year OS=43%) and the median RFS was 26 months (95% CI: 19–33 months; 3-year RFS=42%, 5-year RFS=32%; Fig. 1). Regarding the analysis investigating TTS, the median OS was 62 months (95% CI: 22–102 months) in patients with a TTS less than 31 days, while the median OS was 41 months (95% CI: 19–63 months) in patients with a TTS between 31 and 60 days, 38 months (95% CI: 21–55 months) in patients with a TTS between 61 and 90 days, and 40 months (95% CI: 15–64 months) in patients with a TTS more than 90 days (p=0.602 log rank, Fig. 2A). For RFS analysis, 5 patients were excluded from RFS analysis due to missing recurrence data. Here, no difference in RFS was detected regarding TTS, with a median RFS of 21 months (95% CI: 11–31 months) in patients with a TTS less than 31 days, a median RFS was 26 months (95% CI: 6–46 months) in patients with a TTS between 31 and 60 days, 26 months (95% CI: 14–38 months) in patients with a TTS between 61 and 90 days, and 25 months (95% CI: 18–32 months) in patients with a TTS more than 90 days (p=0.994 log rank, Fig. 2B).

Fig. 1
figure 1

Oncological survival in hepatocellular carcinoma of the study cohort. A Overall survival. The median OS of the cohort was 42 months. B Recurrence-free survival. The median RFS of the cohort was 26 months. OS, overall survival; RFS, recurrence-free survival

Full size image
Fig. 2
figure 2

Oncological survival in hepatocellular carcinoma stratified by time to surgery. A Overall survival. The median OS was 62 in patients with a TTS less than 31 days, while the median OS was 41 months in patients with a TTS between 31 and 60 days, 38 months in patients with a TTS between 61 and 90 days, and 40 months in patients with a TTS more than 90 days (p=0.602 log rank). B Recurrence-free survival. The median RFS was 21 months (95% CI: 11–31 months) in patients with a TTS less than 31 days, 26 months in patients with a TTS between 31 and 60 days, 26 months in patients with a TTS between 61 and 90 days, and 25 months in patients with a TTS more than 90 days (p=0.994 log rank)

Full size image

Univariate and multivariable Cox regressions

Cox regressions were used for OS and RFS to identify risk factors for impaired oncological outcomes. For OS, gender (p=0.002), age (p=0.031), ASA score (<0.001), MELD (p=0.002) and CHILD Pugh Score (p=0.005), and INR (p=0.001) as well as various other liver function parameters, number of nodules (p<0.001), and largest nodule diameter (p=0.013) next to various other preoperative imaging features, laparoscopic resection (p=0.001), additional procedures to resection (p=0.045), intraoperative red blood cells (p<0.001) and FFP (p=0.010) transfusion, R1 resection (p=0.012), pT category (p<0.001), microvascular invasion (MVI, p<0.001), and postoperative duration of hospitalization (p=0.014) and complications (p<0.001) as well as postoperative transfusion of red blood cells (p=0.047) and FFP (p=0.046) gained statistical significance in univariate analysis (Table 3). Subsequently, those parameters were transferred to multivariable analysis (194 patients (89.4%) included due to data availability). In here, age (p=0.009), ASA score (p=0.012), INR (p=0.008), number of nodules (p=0.017), MVI (p=0.016), and postoperative complications (p<0.001) were identified as independent predictors for OS (Table 3). TTS showed no statistical significance in OS (p=0.126). A similar approach was conducted for RFS (183 patients (91.0%) included due to data availability). Comparable to OS, some preoperative liver function values and various preoperative imaging features as well as R1 resection (p=0.018), pT category (p<0.001), and MVI (p<0.001) showed statistical significance in univariate analysis. Subsequently, a multivariable Cox regression was carried out with those parameters. Here, INR (p=0.011), number of nodules (p<0.001), and MVI (p<0.001) were independent prognostic factors for RFS (Table 4). As in OS, TTS was no relevant prognostic factor for RFS (p=0.759).

Table 3 Univariate and multivariable analysis of overall survival in hepatocellular carcinoma
Full size table
Table 4 Univariate and multivariable analysis of recurrence-free survival in hepatocellular carcinoma
Full size table

Discussion

Although improved therapy options with increased interdisciplinary approaches for patients with HCC have been implemented in the last decades, liver resection remains the first option for patients with early disease stage and preserved liver function [9]. However, liver resection in HCC which is usually accompanied by liver cirrhosis and other co-morbidities requires a notable amount of medical resources ranging from surgical theater to intensive care unit and normal ward capacities [17]. Due to the recent COVID pandemic, medical resources were sparse not only in western countries, but across the globe and usually shifted to treat COVID [18]. Therefore, we investigated the role of TTS in surgically resected HCC patients. Within a large European cohort, we demonstrated that TTS was no risk factor for reduced RFS and OS in HCC patients undergoing curative-intent surgery. Interestingly, we also could not identify major differences in perioperative characteristics of patients with different TTS intervals in our analysis. Furthermore, we determined age, ASA score, preoperative INR, multifocal disease, largest nodule diameter, MVI, and postoperative complications as independent prognostic factors of OS and INR, multifocal disease, and MVI as independent prognostic factors of RFS.

The currently available literature reveals conflicting results regarding the influence of TTS in HCC. While in a retrospective monocentric study by Signal et al. a worse survival due to delayed TTS was observed, a more recent multicentric study of Rao et al. showed no statistical significance of a treatment delay above 90 days on OS of HCC patients [19, 20]. Of note, both studies were not focused on surgically treated patients and included locoregional and systemic therapies across a large disease spectrum. In the cohort of Rao et al., only 31.3% were treated by liver resection, while in the publication of Signal et al. 28% of all patients did undergo surgery demonstrating a limited view on patients with early-stage HCC. Another large study by Govalan et al. demonstrated no association between delay in treatment for HCC and worse OS according to the data of 100,000 patients [21]. Although 38% of the included patients were treated with liver resection, non-curative modalities were also included in this investigation. While profiting from a large dataset, these multicenter datasets do only include a limited number of preoperative characteristics especially detailed tumor staging with associated risk factors, e.g., tumor spread and vascular invasion as well an undetailed view on patients’ performance. Thus, to the best of our knowledge, our study is the first report focusing on TTS in a cohort of surgical patients.

Interestingly, a large systematic review demonstrated a worsened OS after each 4 weeks of delay to definitive surgery in bladder, breast, colon, and head/neck cancer [12]. Regarding other carcinomas of the gastrointestinal tract, a 2020 published study showed an improved OS in pancreatic adenocarcinoma if surgery was conducted within 6 weeks after time of diagnose [22]. For gastric cancer on the other hand, a prolonged time to surgery seems to have no effect on OS according to a recent study [23]. In the case of colorectal liver metastasis undergoing liver resection, a larger monocentric retrospective study displayed a worse OS for patients undergoing liver resection with a time to surgery longer than 6 months [24]. Of note, a part of this cohort underwent neoadjuvant chemotherapy, whereas in our study, patients with any preoperative treatment were excluded to reduce bias in the cohort. Given these different findings for common visceral cancers, it is debatable to shift focus to tumors which are more prone to TTS-related effects due to their inherent malignant potential.

Interestingly, the median TTS was significantly higher during the COVID pandemic compared to the time interval before the COVID pandemic exemplifying the aforementioned shift in medical resources also in our university hospital. As TTS was not associated with OS or RFS in our analysis, it is assumable that this specific delay might not have an influence on long-term outcome. However, this research question must be readdressed and studied in detail after a sufficient follow-up time for these recent patients.

In some circumstances, emergency surgery for HCC is necessary, e.g., because of acute tumor bleeding. Subsequently, these cases were also excluded from our analysis. However, in any other scenario, it seems debatable to delay TTS in the surgical candidates to preoperatively improve the performance status as our results suggest that this might not necessarily impair long-term oncological results. Moreover, in our cohort, a notable part of patients was assessed as ASA > III (65%, 141/217). Moreover, ASA score and postoperative complications were determined as independent factors for reduced OS as also demonstrated in other studies [25]. Thus, using the TTS to improve the patient’s condition prior to surgery appears reasonable. Prehabiliation is a health care intervention prior to surgery comprising lifestyle changes and training resulting in improved nutritional status and physical and mental fitness in the form of a multimodal and usually multidisciplinary concept [26, 27]. Previous meta-analyses already demonstrated reduced hospitalization [28] and complication rates [29] in patients undergoing prehabilitation prior to major abdominal surgery. Prehabilitation strategies include the improvement of aerobic fitness and body composition by physical therapy and correction of malnutrition by professional nutrition interventions as well as reduction of alcohol consumption, support for smoking cessation, and medical interventions to correct anemia as well as psychological support to improve preoperative anxiety, depression, and low self-efficacy [30]. With healthcare funding being a hotly debated subject in western society, structured prehabiliation programs have not widely been implemented. From a cost efficiency perspective, prehabiliation might not be implemented en masse but in selected patients benefiting most from preoperative exercise [31]. Given our data, as well as the high prevalence of sarcopenia in HCC and liver cirrhosis, HCC patients might be ideal candidates for structured prehabiliation programs, which is currently also enforced in our department [32].

Besides our primary observation regarding the oncological influence in HCC, we identified several prognostic factors in our cohort which are in line with the literature and indicate comparability of our data to other datasets. MVI has been identified as an important histological parameter and limitational factor for OS and RFS after liver resection and transplantation before [33, 34]. Although examination of suitable preoperative MVI prediction models is becoming more popular in recent years, postoperative histopathological examination currently seems to be the only valid option for proving MVI in HCC at current state [35, 36]. Further we could identify the number of nodules as independent predictor for OS as also commonly known risk factor for reduced OS [37, 38]. Interestingly, number of nodules as preoperative imaging parameter was described as prognostic preoperative imaging markers for appearance of MVI recently [39, 40]. INR has been identified as independent predictor for OS and RFS in our cohort which was also demonstrated in previous studies [41, 42]. Of note, all independent prognostic variables as defined by our multivariable models associated with OS and RFS were not different in the grouped analysis regarding TTS in our patients underlining the validity our results.

As with all retrospective analyses, our study has certainly limitations having to be considered when interpreting the results. Within the monocentric setting of our study, the data reflects the authors’ individual approach to HCC which might be different to clinical standards of other hepatobiliary centers. Also, due to etiological differences, our implications might not be transferable to Asian patients. While the focus of our study was to investigate the influence of TTS in surgically treated patients, we are not able to report on patients dropping from surgical treatment plans due to progression during waiting time as only a small subset of patients was diagnosed in our hepatobiliary center and most of the TTS interval was based on the time from diagnosis to initially presentation to our hepatobiliary unit and not on the waiting time for surgery. However, as HCC is usually slowly progressing which does also explain our findings, it is assumable that the proportion of patients showing a significant disease progression precluding surgical treatment during waiting time might be low. Of note, especially OS appeared numerically higher in patients with short TTS (1–30 days) compared to patients with longer TTS intervals but did not show statistical significance (p=0.602). It is debatable whether a statistically significant benefit would be detectable in a larger data set. However, generic cox regression gave no indication for a relevant effect of a shorter TTS and the better result was not replicable in the RFS analysis. Nevertheless, as with all monocentric analysis, our results warrant further investigations in larger, multicentric data sets.

Conclusion

Notwithstanding the mentioned limitations, we demonstrated that TTS does not influence OS and RFS in patients with HCC who underwent liver resection in curative intent. This finding might be used for prioritizing patients in the scenario of restricted medical resources. Further, our results suggest prehabilitation as important measure to improve short- and long-term outcomes in surgical candidates with HCC.