Clinical outcomes and prognostic factors after adrenalectomy for adrenal metastasis from hepatocellular carcinoma

Abstract

Backgrounds/Aims

Adrenal metastasis from hepatocellular carcinoma (HCC) is uncommon, and the role of surgical resection remains undefined. This study analyzed clinical outcomes and prognostic factors in patients who underwent adrenalectomy for adrenal metastasis from HCC at a single tertiary center.

Methods

We retrospectively analyzed 31 patients who underwent adrenalectomy for adrenal metastasis from HCC (2002–2019), classified by initial HCC treatment: liver resection (n = 17), liver transplantation (n = 7), and non-surgical management (n = 7).

Results

The median age of patients was 58 years; 93.5% were male, and 90.3% had hepatitis B virus infection. The median interval from HCC diagnosis to adrenal metastasis was 20.5 months. After adrenalectomy, the median recurrence-free survival was 7.5 months, and the median overall survival (OS) was 34.3 months. The 5- and 10-year survival rates were 42.3% and 32.6%, respectively. OS differed significantly across treatment groups (p = 0.008): median OS was 75.5 months in the liver resection group, compared to 33.8 months in the non-surgical group and 20.7 months in the liver transplantation group. On multivariate Cox regression, prior liver resection was independently associated with reduced mortality.

Conclusions

Adrenalectomy was associated with prolonged OS in selected patients with adrenal metastasis from HCC, particularly those with prior liver resection. Given the high recurrence rate and the lack of a non-surgical comparator, these findings warrant cautious interpretation. Adrenalectomy may be considered within a multidisciplinary framework for carefully selected patients, pending confirmation by prospective multicenter studies.

INTRODUCTION

Hepatocellular carcinoma (HCC) ranks as the sixth most common malignancy and the third leading cause of cancer-related mortality globally [1,2]. Despite advancements in surveillance and treatment, recurrence and metastasis continue to hinder long-term survival. Extrahepatic metastasis occurs in 10%–37% of patients, with a median survival often under 12 months [3,4]. Common sites include the lung (39.5%–53.1%), lymph nodes (29.6%–34.2%), and bone (25.4%–43.2%). Adrenal metastases are relatively infrequent (8.0%–19.1% of HCC patients with extrahepatic manifestations) but are increasingly recognized due to improved imaging and prolonged survival [5-9].

Adrenal metastases may occur synchronously or metachronously after treatment of the primary tumor [8,10,11]. While sorafenib is the preferred first-line therapy for metastatic HCC [1], surgical intervention remains an option for isolated adrenal metastases. In particular, for metachronous lesions, adrenalectomy may improve overall survival (OS). Minor retrospective studies indicate that surgical resection can achieve 5-year OS rates of 20%–40%, in contrast to the suboptimal outcomes observed with systemic therapy alone [12-14]. Metachronous solitary adrenal metastasis often develops years after initial hepatic resection or transplantation and may signify more indolent tumor biology [8,9].

However, the optimal treatment method remains unclear, and there are no agreed-upon standards. Most data come from small, single-center studies with built-in selection bias [5-9]. Because there is limited evidence, the present study aimed to analyze clinical outcomes and prognostic factors in patients who underwent adrenalectomy for synchronous and metachronous adrenal metastasis from HCC at a single tertiary center.

PATIENTS AND METHODS

Patients

This retrospective cohort study included patients who underwent adrenalectomy for adrenal metastases from HCC at a tertiary referral center between May 2002 and July 2019. Patients were identified through institutional databases. The inclusion criteria were: (1) histologically confirmed adrenal metastases originating from HCC; (2) extensive clinical and follow-up data; and (3) curative-intent adrenalectomy with complete tumor resection. Patients with disseminated disease, uncontrolled intrahepatic recurrence, or substantial hepatic dysfunction were excluded.

Patients were classified according to their initial treatment for HCC: (1) non-surgical (transarterial chemoembolization or radiofrequency ablation without viable intrahepatic HCC); (2) liver resection; and (3) liver transplantation. The decision to perform an adrenalectomy was made by an interdisciplinary tumor board, based on radiological imaging and clinical progression.

Data collection

Demographic and clinical factors included age, sex, body mass index (BMI), comorbidities (hypertension and diabetes), viral hepatitis status, American Society of Anesthesiologists (ASA) score, and prior treatment history. Laboratory data at the time of adrenal metastasis diagnosis included aspartate aminotransferase, alanine aminotransferase, alpha-fetoprotein, and protein induced by vitamin K absence or antagonist-II (PIVKA-II). Evaluated tumor parameters included laterality, the maximum size of the adrenal lesion, and the diameter of the underlying hepatic tumor. Operative data were retrieved from prospectively maintained surgical records.

Surgical procedure

All adrenalectomies were performed by skilled endocrine surgeons. The choice of method (open vs. laparoscopic) depended on the size and location of the tumor and the surgeon’s preference. The transabdominal method was used for open surgeries. Laparoscopic adrenalectomy was chosen for small, well-defined solitary lesions. The goal was to remove the entire tumor with negative margins.

Follow-up and outcomes

After the adrenalectomy, the patient underwent a physical exam, liver function tests, and tumor marker tests every three months for two years, and then every six months thereafter. A contrast-enhanced computed tomography or magnetic resonance imaging was performed every six months or whenever recurrence was suspected.

Statistical analysis

Continuous data are presented as medians with ranges, while categorical variables are expressed as frequencies and percentages. Continuous variables were compared using Mann-Whitney U tests, and categorical data were compared using chi-square or Fisher’s exact tests. The primary outcome was OS, defined as the interval from adrenalectomy to death or last follow-up. The secondary outcome was recurrence-free survival (RFS), defined as the interval from adrenalectomy to confirmed recurrence. We used univariate and multivariate Cox proportional hazards regression analyses to identify prognostic factors for RFS and OS, reporting hazard ratios (HR) with 95% confidence intervals (CI). The proportional hazards assumption was verified using Schoenfeld residuals. Variables with p < 0.2 in univariate analysis were entered into multivariate models using forward stepwise selection. Survival curves were estimated using the Kaplan-Meier method and compared using the log-rank test. All analyses were performed using SPSS version 26.0 (IBM Corp.). A two-sided p-value of less than 0.05 was considered statistically significant.

Ethics

The study was approved by the Institutional Review Board of Samsung Medical Center (SMC-2025-11-028). Because it used anonymized data from the past, informed consent was not needed. All procedures followed the rules set by the Declaration of Helsinki.

RESULTS

Baseline characteristics

There were 31 patients in total: 29 males (93.5%) and 2 females (6.5%). The average age was 58 years (range: 41–72 years), and the average BMI was 22.5 kg/m² (range: 17.2–28.3 kg/m²). Hypertension (22.6%) and diabetes (16.1%) were two of the comorbidities. Most of the patients (90.3%) had hepatitis B virus (HBV). The percentages for ASA classes were as follows: I (9.7%), II (71.0%), and III (19.4%).

There were three treatment groups: non-surgical (n = 7, 22.6%), liver resection (n = 17, 54.8%), and liver transplantation (n = 7, 22.6%). Ninety-three percent of the patients had open adrenalectomy, while only three had laparoscopic surgeries. The average operative time was 240 minutes (range: 70–660 minutes). One patient had a bilateral adrenalectomy.

Twenty patients (64.5%) had right-sided adrenal tumors, ten (32.3%) had left-sided, and one patient (3.2%) had bilateral tumors. Twelve patients (38.7%) had synchronous metastases, defined as adrenal metastasis diagnosed concurrently with HCC (within 3 months). Of these, 8 (66.7%) underwent simultaneous hepatectomy and adrenalectomy, while the remaining 4 (33.3%) underwent staged procedures within 3 months. Nineteen patients (61.3%) had metachronous lesions. The median time from HCC diagnosis to adrenal metastasis was 20.5 months (range: 3.0–219.0). The median adrenal tumor size was 3.4 cm (range: 0.2–9.0), and the median primary hepatic tumor size was 6.8 cm (range: 1.2–25.0). The median interval from initial HCC treatment to adrenalectomy was 50.0 months (range: 12.6–219.1). Four patients (12.9%) required intraoperative transfusion (Table 1).

Table 1.

Baseline characteristics

Characteristic	Value
Sex (male/female)	29 (93.5)/2 (6.5)
Age (yr)	58 (41–72)
BMI (kg/m2)	22.5 (17.2–28.3)
Hypertension	7 (22.6)
Diabetes	5 (16.1)
HBV	28 (90.3)
Maximum HCC tumor size in liver (cm)	6.8 (1.2–25.0)
Groups
Non-surgical group	7 (22.6)
Liver resection group	17 (54.8)
Liver transplantation group	7 (22.6)
ASA
1	3 (9.7)
2	22 (71.0)
3	6 (19.4)
Time from HCC diagnosis to adrenal metastasis (mon)	20.5 (3.0–219.0)
Time from initial treatment to adrenalectomy (mon)	50.0 (12.6–219.1)
AST (IU/L)	33 (19–233)
ALT (IU/L)	33 (15–346)
AFP (ng/mL)	27 (1.3–304.0)
PIVKA-II (mAU/mL)	114 (12–6,572)
Adrenal tumor location
Right	20 (64.5)
Left	10 (32.3)
Both	1 (3.2)
Bilateral adrenalectomy	1 (3.2)
Operation time during adrenalectomy (min)	240 (70–660)
Maximum tumor size in adrenal HCC (cm)	3.4 (0.2–9.0)
Surgical approach for adrenalectomy (open/laparoscopy)	28 (90.3)/3 (9.7)
Transfusion	4 (12.9)
HCC recurrence after adrenalectomy	28 (90.3)
RFS after adrenalectomy (mon)	7.5 (0.2–230.0)
HCC recurrence sites after adrenalectomy
Liver	9 (29.0)
Bone	7 (22.6)
Lung	4 (12.9)
Adrenal gland	3 (9.7)
Others	5 (16.1)
None	3 (9.7)
Death	18 (58.1)
OS after adrenalectomy (mon)	34.3 (6.3–230.0)

Values are presented as number (%) or median (range).

BMI, body mass index; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; ASA, American Society of Anesthesiologists; AST, aspartate transaminase; ALT, alanine transaminase; AFP, alpha-fetoprotein; PIVKA-II, protein induced by vitamin K absence or antagonist-II; RFS, recurrence-free survival; OS, overall survival.

Recurrence after adrenalectomy

Recurrence of HCC was noted in 28 patients (90.3%). The RFS rates at one, three, five, and ten years were 45.2%, 21.8%, 14.5%, and 10.9%, respectively (Fig. 1A). Recurrence was observed in the liver (29.0%) and bone (16.1%), among other locations. Three patients (9.7%) remained recurrence-free throughout the follow-up period. The median RFS for the entire cohort was 7.5 months (95% CI 0.6–14.3). Stratified by treatment group, the median RFS was 6.0 months for non-surgical treatment, 21.9 months for liver resection, and 4.7 months for liver transplantation (p = 0.108) (Fig. 2A).

Fig. 1.

Survival. (A) Recurrence-free survival and (B) overall survival after adrenalectomy. HCC, hepatocellular carcinoma.

Fig. 2.

Survival according to initial treatment groups. (A) Recurrence-free survival and (B) overall survival after adrenalectomy. HCC, hepatocellular carcinoma.

In univariate Cox regression analysis, female sex showed an association with elevated recurrence risk (HR = 7.201; 95% CI 1.524–34.025; p = 0.024). However, given that only 2 female patients were included, this estimate is highly unstable and likely reflects sparse data bias; thus, it was excluded from the multivariate model. Elevated BMI and treatment modality showed trends toward significance (p = 0.100 and 0.123, respectively). In multivariate Cox regression, elevated BMI (HR = 1.231; 95% CI 1.018–1.490; p = 0.031) and liver transplantation as initial treatment (HR = 3.228; 95% CI 1.064–9.793; p = 0.038) were identified as independent prognostic factors for recurrence (Table 2).

Table 2.

Risk factors for HCC recurrence after adrenalectomy

	HR	95% CI	p-value
Univariate
Sex (female)	7.201	1.524–34.025	0.024
Age	1.024	0.978–1.072	0.318
BMI	1.155	0.973–1.372	0.100
HBV	2.302	0.670–7.908	0.185
AST	1.001	0.994–1.008	0.876
ALT	1.000	0.994–1.006	0.964
AFP	1.001	0.997–1.006	0.514
PIVKA-II	1.000	1.000–1.000	0.371
Group
Non-surgical group	1	1	0.123
Liver resection group	0.494	0.181–1.347	0.168
Liver transplantation group	1.338	0.435–4.114	0.611
Adrenalectomy (laparoscopy)	0.409	0.096–1.738	0.226
Operation time during adrenalectomy	0.999	0.995–1.002	0.507
Transfusion	0.756	0.261–2.192	0.607
Maximum tumor size in adrenal HCC	1.005	0.856–1.179	0.956
Maximum HCC tumor size in liver	1.015	0.968–1.063	0.537
Bilateral adrenalectomy	1.311	0.173–9.918	0.793
Multivariate
BMI	1.231	1.018–1.490	0.031
Group–liver resection group	1.000	1.000–1.000	0.028

HR, hazard ratio; 95% CI, 95% confidence interval; BMI, body mass index; HBV, hepatitis B virus; AST, aspartate transaminase; ALT, alanine transaminase; AFP, alpha-fetoprotein; PIVKA-II, protein induced by vitamin K absence or antagonist-II; HCC, hepatocellular carcinoma.

Overall survival after adrenalectomy

During follow-up, 18 patients (58.1%) passed away. The OS rates for 1, 3, 5, and 10 years were 90.3%, 58.5%, 42.3%, and 32.6%, respectively (Fig. 1B). The median OS was 33.8 months for the non-surgical group, 75.5 months for the liver resection group, and 20.7 months for the liver transplantation group (p = 0.008) (Fig. 2B).

In univariate Cox regression, HBV infection, elevated PIVKA-II, and treatment group were significantly associated with mortality. Female sex showed a trend in univariate analysis (p = 0.109) but was excluded from the multivariate model due to sparse data (n = 2). In multivariate Cox regression, liver resection as initial treatment was identified as an independent predictor of improved survival (HR = 0.183; 95% CI 0.042–0.796; p = 0.011) (Table 3).

Table 3.

Risk factors for mortality after adrenalectomy

	HR	95% CI	p-value
Univariate
Sex (female)	3.569	0.754–16.902	0.109
Age	1.044	0.984–1.108	0.155
BMI	1.123	0.914–1.381	0.268
HBV	5.080	1.307–19.747	0.019
AST	1.001	0.993–1.010	0.771
ALT	1.006	0.999–1.013	0.114
AFP	1.000	0.994–1.005	0.909
PIVKA-II	1.000	1.000–1.001	0.016
Group
Non-surgical group	1	1	0.017
Liver resection group	0.216	0.053–0.884	0.033
Liver transplantation group	1.325	0.346–5.065	0.681
Adrenalectomy (laparoscopy)	0.439	0.058–3.322	0.426
Operation time during adrenalectomy	1.000	0.996–1.004	0.986
Transfusion	0.755	0.172–3.314	0.709
Maximum tumor size in adrenal HCC	1.032	0.853–1.249	0.744
Maximum HCC tumor size in liver	1.025	0.972–1.081	0.362
Bilateral adrenalectomy	6.986	0.780–62.528	0.082
Multivariate
Group–liver resection group	0.183	0.042–0.796	0.011

HR, hazard ratio; 95% CI, 95% confidence interval; BMI, body mass index; HBV, hepatitis B virus; AST, aspartate transaminase; ALT, alanine transaminase; AFP, alpha-fetoprotein; PIVKA-II, protein induced by vitamin K absence or antagonist-II; HCC, hepatocellular carcinoma.

DISCUSSION

This study demonstrates that adrenalectomy for adrenal metastases arising from HCC can yield substantial long-term survival in carefully selected patients, particularly those initially managed with liver resection.

For our group of 31 patients, the median RFS was 7.5 months, with 1, 3, 5, and 10-year rates of 45.2%, 21.8%, 14.5%, and 10.9%, respectively. Patients who underwent liver resection exhibited a numerically longer median RFS of 21.9 months compared to non-surgical or transplantation cohorts; however, this difference did not reach statistical significance (p = 0.108). This suggests that the primary tumor’s biology and the host liver’s condition continue to influence tumor behavior even after extrahepatic dissemination.

Despite the short median RFS, the median OS was 34.3 months, with 5- and 10-year OS rates of 42.3% and 32.6%, respectively. The apparent paradox between the short RFS (7.5 months) and prolonged OS can be attributed to several factors: active salvage locoregional therapies (transarterial chemoembolization [TACE], radiofrquency ablation, repeat resection) administered after recurrence, the favorable hepatic reserve within this highly selected cohort, and the inherently indolent tumor biology implied by patient selection. The treatment group significantly influenced OS (p = 0.008): patients who underwent liver resection achieved a median OS of 75.5 months, compared to 20.7 months for liver transplantation and 33.8 months for non-surgical patients. These results are comparable to or better than prior series, which reported a median survival of 11–26 months [15].

Korean studies demonstrated that surgical resection resulted in superior survival outcomes compared to TACE, radiation, or supportive treatment [9,11]. Our findings confirm that adrenalectomy is indicated in certain patients with effectively excised primary HCC and metachronous, localized extrahepatic disease. Numerous studies characterize adrenal metastasis as a delayed, metachronous occurrence after effective intrahepatic management, typically within 12–24 month intervals [15]. Our cohort exhibited a median interval of 20.5 months from HCC diagnosis to adrenal metastasis and 50.0 months from initial treatment to adrenalectomy, indicating favorable tumor biology and sufficient hepatic reserve. Most studies agree that complete adrenal resection offers better survival than non-operative management when there are only one or a few metastases. The five-year survival rate is 25%–35% [10,14].

Our series clearly shows distinct differences in survival rates based on the initial treatment for HCC. Patients who underwent liver resection exhibited the most favorable outcomes, while non-surgical patients demonstrated intermediate results, and those post-transplant experienced the least favorable outcomes. The mechanisms are as follows: (1) patients who have undergone resection demonstrate enhanced liver function, allowing them to tolerate multiple therapies; (2) adrenal metastases may indicate exclusively extrahepatic disease; and (3) a lack of extended immunosuppressive exposure. Post-transplant adrenal metastasis likely signifies aggressive biology (microvascular invasion, under-staged primary) exacerbated by immunosuppression, facilitating systemic dissemination and impeding antitumor immunity. This explains the short RFS (4.7 months) and poor OS (20.7 months).

Regarding sex as a prognostic variable, female sex showed an association with recurrence risk in univariate analysis but was excluded from the final multivariate Cox models due to sparse data bias. With only 2 female patients (6.5%), any point estimate is severely unstable. This association has not been consistently reported in prior studies, and current data do not support female sex as a reliable independent prognostic factor. Subsequent pooled multicenter analyses with adequate female representation are essential before conclusions can be drawn about sex-based risk stratification.

An elevated BMI was independently associated with a greater risk of recurrence after adrenalectomy (HR = 1.231; p = 0.031). Obesity and metabolic syndrome incrementally promote the onset and adverse outcomes of HCC through chronic inflammation, insulin resistance, and steatohepatitis. Although 90.3% of the patients were HBV-positive, the BMI observation may reflect a pro-inflammatory, pro-tumorigenic host environment, which is consistent with evidence that an elevated BMI predicts extrahepatic recurrence after hepatic resection [12].

The observed protective association of liver resection with mortality in multivariate Cox analysis (HR = 0.183; p = 0.024) within the treatment group underscores the critical importance of patient selection for adrenalectomy outcomes. The three treatment groups differ substantially in baseline characteristics and tumor biology, suggesting that the observed survival difference likely reflects patient selection rather than an independent effect of adrenalectomy. Recent evidence indicates that patients with solitary or limited extrahepatic metastasis, extended disease-free intervals, controlled intrahepatic disease, and absence of vascular invasion derive the greatest benefit [13]. Conversely, patients with multiple extrahepatic sites, uncontrolled recurrence, or rapidly progressive disease experience minimal benefit and may be more effectively managed with systemic therapy.

The right-sided predominance (64.5%) observed in our cohort aligns with prior literature and is likely attributable to anatomical factors. The right adrenal gland’s proximity to the right hepatic lobe and inferior vena cava facilitates direct hematogenous spread via the right adrenal vein. HCC in East Asian populations predominantly arises in the right hepatic lobe, and venous drainage from the right liver preferentially accesses the right adrenal gland via the short hepatic veins and inferior vena cava. These anatomical relationships likely explain the right-sided preponderance of adrenal metastasis in HCC.

To identify suitable candidates for adrenalectomy, we performed an exploratory analysis comparing patients with early recurrence (RFS ≤ 6 months; n = 17, 54.8%) to those with longer RFS (> 6 months; n = 14, 45.2%). Patients with early recurrence were more likely to have undergone liver transplantation (52.9% vs. 14.3%; p = 0.022) and had higher PIVKA-II levels at adrenal metastasis diagnosis (median 325 vs. 52 mAU/mL; p = 0.031). In multivariate Cox analysis, liver transplantation (HR = 3.2; 95% CI 1.1–9.4; p = 0.032) and elevated PIVKA-II (HR = 1.001; 95% CI 1.000–1.002; p = 0.044) were independently associated with early recurrence. Based on these findings, adrenalectomy may be most appropriate for patients who: (1) initially underwent liver resection rather than transplantation; (2) have normal or mildly elevated PIVKA-II at adrenal metastasis diagnosis; and (3) present with small, solitary, well-defined adrenal lesions with no evidence of concurrent extrahepatic dissemination.

Individually tailored concepts from multidisciplinary tumor boards should guide surgical decisions for adrenal metastases [7,9,14]. Our findings, in the absence of larger controlled trials, pertain to meticulously selected cohorts with well-maintained liver function and regulated intrahepatic tumors.

Several limitations of this study merit discussion. First, this was a single-center, retrospective study with a limited sample size (n = 31). This substantially constrained statistical power and contributed to imprecise estimates, particularly for subgroup analyses. Second, the absence of a matched non-surgical control cohort precluded formal causal inference regarding the survival benefit of adrenalectomy. All patients were selected by a multidisciplinary tumor board, which inherently enriched the cohort for favorable clinical characteristics. Third, sex could not be reliably evaluated as a prognostic factor given the very small number of female patients (n = 2). Fourth, the lack of standardization in staging protocols, imaging modalities, and systemic therapy regimens across the prolonged 17-year inclusion period may have introduced heterogeneity in recurrence detection. Fifth, the high prevalence of HBV infection (90.3%) may limit generalizability to Western populations where hepatitis C virus, alcohol-related liver disease, or metabolic-associated steatotic liver disease are predominant. Finally, the absence of detailed immunosuppressive regimen data for post-transplant patients limited full accounting for their suboptimal outcomes.

In this highly selected cohort, adrenalectomy was associated with prolonged OS in patients with adrenal metastasis from HCC, particularly those who had previously undergone liver resection. However, given the very high recurrence rate (90.3%), the short median RFS (7.5 months), and the absence of a non-surgical comparator group, these findings should be interpreted with considerable caution. The data indicate that adrenal metastasis from HCC most often represents a manifestation of systemic disease, and adrenalectomy is unlikely to be curative in the majority of patients. Therefore, adrenalectomy should be considered only within a comprehensive multidisciplinary framework that includes systemic therapy and rigorous oncological surveillance. It should be reserved for carefully selected patients with controlled intrahepatic disease, preserved hepatic function, and limited extrahepatic burden. Prospective multicenter studies with standardized eligibility criteria and modern systemic regimens are needed to validate these findings, refine patient selection criteria, and establish evidence-based indications for adrenalectomy in this rare clinical setting.