Abstract
Background.
Adrenocortical carcinoma (ACC) is an aggressive, deadly malignancy. Resection remains the primary treatment; however, there is conflicting evidence regarding the optimal approach to and extent of surgery and the role of adjuvant therapy. We evaluated the impact of surgical technique and adjuvant therapies on survival in non-metastatic ACC.
Methods.
We performed a retrospective cohort study of subjects who underwent surgery for non-metastatic ACC between 2010 and 2019 utilizing the National Cancer Data-base. The primary outcome was overall survival. Cox proportional hazards models were developed to identify associations between clinical and treatment characteristics and survival.
Results.
Overall, 1175 subjects were included. Their mean age was 54 ± 15 years, and 62% of patients were female. 67% of procedures were performed via the open approach, 22% involved multi-organ resection, and 26% included lymphadenectomy. Median survival was 77.1 months. Age (hazard ratio [HR] 1.019; p < 0.001), advanced stage (stage III HR 2.421; p < 0.001), laparoscopic approach (HR 1.329; p = 0.010), and positive margins (HR 1.587; p < 0.001) were negatively associated with survival, while extent of resection (HR 1.189; p = 0.140) and lymphadenectomy (HR 1.039; p = 0.759) had no association. Stratified by stage, laparoscopic resection was only associated with worse survival in stage III disease (HR 1.548; p = 0.007). Chemoradiation was only associated with improved survival in patients with positive resection margins (HR 0.475; p = 0.004).
Conclusion.
Tumor biology and surgical margins are the primary determinants of survival in non-metastatic ACC. Surgical extent and lymphadenectomy are not associated with overall survival. In advanced disease, the open approach is associated with improved survival.
Adrenocortical carcinoma (ACC) is a rare, aggressive malignancy with a grim prognosis. The estimated annual incidence is between 0.72 and 2 cases per million.1-7 Patients are often diagnosed at an advanced stage with large, high-grade tumors and distant metastatic disease.1-7 Prognosis is exceedingly poor, with a median survival of 7–51.2 months and disease-specific 5-year survival of 11.6–60%.2,3,8,9 Given disease rarity and the dearth of prospective randomized trial data, many controversies remain regarding the optimal surgical and medical management of patients with ACC, especially for those without distant metastatic disease at diagnosis.
Surgical resection remains the mainstay of treatment for non-metastatic ACC, and margin-negative resection is strongly associated with improved survival. However, regardless of margin status, the relatively high rate of recurrence following resection suggests that these patients frequently have occult micrometastatic disease.3,9,10 However, controversy remains regarding surgical decision making in patients with non-metastatic ACC, including the extent of surgery required, the utility of regional lymphadenectomy, and the role of minimally invasive techniques.
Given the proven benefit of margin-negative resection, en bloc multi-organ resection is generally considered the standard surgical approach for locally invasive ACC. Nonetheless, data to support multi-organ resection are limited; multi-organ resection has not been shown to impact survival accounting for margin status,10-15 and whether universal en bloc resection improves rates of margin-negative resection has not been well-described in the existing literature. Similarly, there is no consensus regarding the appropriate extent of regional lymph node dissection during primary ACC resection, although a recent meta-analysis did identify an association between lymphadenectomy and improved survival in localized ACC.10,16,17 Beyond considerations about the extent of surgery, the topic of surgical approach has been hotly debated in ACC. Some studies have identified higher rates of recurrence and lower overall survival in patients undergoing laparoscopic resection, but others have found no difference between the open and minimally invasive approaches.7,10,17
Additionally, adjuvant chemotherapy and radiotherapy are often utilized in ACC treatment, although most supporting data are retrospective. Some studies demonstrate improvements in recurrence-free survival with mitotane, although perhaps only in certain populations.7,17,18 Despite some conflicting studies, adjuvant radiation appears to improve recurrence-free survival but not overall survival.7
Given the conflicting data on optimal treatment, this study aimed to evaluate the relationship between extent of surgery, surgical approach, adjuvant therapies, and survival in a large, national cohort of patients with non-metastatic ACC.
METHODS
Data Source
We utilized the National Cancer Database (NCDB) to perform a retrospective cohort study of all patients who presented with non-metastatic ACC. The NCDB is a joint initiative of the American College of Surgeons Commission on Cancer and the American Cancer Society and is one of the largest cancer registries in the world.19 The NCDB is a nationwide, facility-based dataset of all incident cancer diagnoses from over 1500 Commission on Cancer-accredited cancer programs, representing 72% of all newly diagnosed malignancies in the United States annually.20 This study was deemed exempt from University of Pennsylvania Institutional Review Board approval.
Study Cohort
All adult patients diagnosed with ACC from 2010 to 2019 were identified for inclusion by either International Classification of Diseases for Oncology, Third Revision (ICD-O-3) histology code for adrenal cortical carcinoma (8370 and 8373) or by ICD Tenth Revision (ICD-10) code for malignant neoplasm of the adrenal cortex (C74.0). Of 3251 patients with ACC, 1200 had documentation of metastatic disease and were excluded. Patients who did not have documentation of staging information (n = 416), did not receive upfront surgery (n = 114), did not undergo any surgery (n = 114), or underwent debulking surgery or unknown surgical intervention (n = 41) were also excluded, as were an additional 305 patients who underwent partial adrenalectomy, excisional biopsy, or other unspecified local excision, leaving a final cohort of 1175 patients. Data regarding adjuvant immunotherapy or hormonal therapy were excluded due to inadequate sample size.
Study Measures
Demographic characteristics (age, sex, race, ethnicity, insurance status, Charlson–Deyo score), tumor characteristics (clinical and pathologic staging, tumor size and grade, margin status after resection), and hospital type were abstracted. For analysis, pathologic stage was utilized if available, otherwise clinical stage was used.
Treatment characteristics were also abstracted. Surgical procedures were classified as adrenal resection (total adrenalectomy) or en bloc multi-organ resection (defined by the NCDB as any surgery involving ‘partial or total removal of the primary site with a resection in continuity … with other organs’). Patients who underwent partial adrenalectomy were excluded as this does not meet standard of care for ACC. Procedures were also classified by surgical approach (laparoscopic or open) and whether lymphadenectomy was performed. For the surgical approach, intention-to-treat (ITT) analysis was utilized such that patients converted from laparoscopic to open procedure were defined as ‘laparoscopic’. Lymphadenectomy was also defined via ITT, per the NCDB definition. In this data-set, there was strong correlation between number of nodes reported excised in pathology and operative reports (correlation 99.3% for no nodes, 97.8% for one to three nodes, 97.9% for four or more nodes). There is no preset definition for number of nodes to count as formal lymphadenectomy in ACC, and prior studies have used definitions ranging between any purposeful removal of lymph nodes to removal of five or more lymph nodes.21-23 In the NCDB, the operative report is used to define the purposeful extent of lymph node dissection. For this study, lymphadenectomy was defined as any attempt at regional nodal resection regardless of the number of nodes removed. Subanalyses were performed using more stringent definitions of lymphadenectomy. Patients treated with mitotane were classified as receiving chemotherapy according to the NCDB. The primary outcome was overall survival, defined as the length of time from diagnosis to mortality.
Statistical Analysis
Descriptive statistics were reported as means with standard deviations (SDs) or medians with interquartile ranges (IQRs) for continuous variables, and categorical variables were reported as frequencies with percentages. Group comparisons were performed using Chi-square tests, Student’s t-tests, and Wilcoxon rank-sum tests, as appropriate. Logistic regression was utilized to determine traits associated with treatment strategies.
Survival analysis was performed using the Kaplan–Meier method. Median survival time and 1-, 3-, and 5-year survival rates were calculated; Mantel–Haenszel tests were used to compare differences in survival. Univariable and multivariable Cox proportional hazards models were created to determine associations between clinical and treatment characteristics and overall survival, with multiple imputation by chained equations utilized to account for missingness. All variables were integrated into the multivariable models except for insurance status, which was excluded due to collinearity with age, and pathologic grade due to a high degree of missingness. Multivariable models were stratified by stage and resection margin. Statistical inference was performed utilizing Stata, version 17.0 (Stata Corp LCC, College Station, TX, USA).24
RESULTS
Cohort Characteristics
A total of 1175 patients met the inclusion criteria. The mean age was 54.1 years (SD 15.2) and 62.1% were female. The majority of patients (86.0%) were White, 9.5% were Black, and 7.9% identified as Hispanic. Most patients were privately insured (56.5%) or insured by Medicare (29.3%). At the time of presentation, most patients had stage II (45.9%) or stage III (44.5%) disease, and the median tumor size was 10.3 cm (IQR 7.0–14.5). These characteristics are summarized in Table 1.
TABLE 1.
Demographic and clinical characteristics of patients with non-metastatic ACC, stratified by extent of surgery
| All patients (N = 1175) | Total adrenalectomy (n = 921) |
En bloc multi-organ resection (n = 254) |
p value | |
|---|---|---|---|---|
| Mean age, years (SD) | 54.1 (15.2) | 53.9 (15.3) | 54.7 (15.0) | 0.424 |
| Sex | ||||
| Male | 445 (37.9) | 325 (35.3) | 120 (47.2) | 0.001 |
| Female | 730 (62.1) | 596 (64.7) | 134 (52.8) | |
| Race | ||||
| White | 1011 (86.0) | 788 (85.6) | 223 (87.8) | 0.179 |
| Black | 111 (9.5) | 94 (10.2) | 17 (6.7) | |
| Other | 53 (4.5) | 39 (4.2) | 14 (5.5) | |
| Spanish/Hispanic origin | ||||
| Yes | 93 (7.9) | 78 (8.5) | 15 (5.9) | 0.180 |
| No | 1082 (92.1) | 843 (91.5) | 239 (94.1) | |
| Insurance status | ||||
| Private | 656 (55.8) | 523 (56.8) | 133 (52.4) | 0.337 |
| Medicare | 340 (29.9) | 259 (28.1) | 81 (31.9) | |
| Medicaid | 87 (7.4) | 70 (7.6) | 17 (6.7) | |
| Other government | 30 (2.6) | 23 (2.5) | 7 (2.8) | |
| Uninsured | 48 (4.1) | 38 (4.1) | 10 (3.9) | |
| Unknown | 14 (1.2) | 8 (0.9) | 6 (2.4) | |
| Charlson–Deyo score | ||||
| 0 | 875 (74.5) | 678 (73.6) | 197 (77.6) | 0.416 |
| 1 | 225 (19.2) | 185 (20.1) | 40 (15.8) | |
| 2 | 49 (4.2) | 39 (4.2) | 10 (3.9) | |
| ≥ 3 | 26 (2.2) | 19 (2.1) | 7 (2.8) | |
| Laterality | ||||
| Left | 646 (55.1) | 487 (53.1) | 159 (62.6) | 0.007 |
| Right | 526 (44.9) | 431 (47.0) | 95 (37.4) | |
| Stage of disease | ||||
| I | 113 (9.6) | 108 (11.7) | 5 (2.0) | < 0.001 |
| II | 539 (45.9) | 443 (48.1) | 96 (37.8) | |
| III | 523 (44.5) | 370 (40.2) | 153 (60.2) | |
| Median tumor size, cm (IQR) | 10.3 (7.5) | 9.9 (7.2) | 13.0 (8.8) | < 0.001 |
| Margin status | ||||
| R0 | 866 (80.9) | 679 (81.0) | 187 (80.6) | 0.885 |
| R1 or R2 | 204 (19.1) | 159 (19.0) | 45 (19.4) | |
| Adjuvant therapy | ||||
| None | 539 (48.3) | 422 (48.1) | 117 (49.2) | 0.462 |
| Radiation only | 104 (9.3) | 83 (9.5) | 21 (8.8) | |
| Chemotherapy only | 275 (24.6) | 210 (23.9) | 65 (27.3) | |
| Radiation and chemotherapy | 198 (17.7) | 163 (18.6) | 35 (14.7) |
Data are expressed as n (%) unless otherwise specified
Bold values indicate p < 0.05
ACC adrenocortical carcinoma, IQR interquartile range, SD standard deviation
Surgical Approach
Most procedures were performed via the open approach (67.2%). While 32.8% of procedures were attempted laparoscopically or robotically, 14.9% of those (4.9% overall) were converted to open procedures (Table 2). Most resections were restricted to the adrenal gland (78.4%); 21.6% included en bloc multi-organ resections. Lymphadenectomy was performed in 25.9% of procedures. Patients with more advanced disease more frequently underwent open surgery (stage I, 29.7%; stage II, 65.2%; stage III, 77.8%; p < 0.001), as well as en bloc multi-organ resection (stage I, 4.4%; stage II, 17.8%; stage III, 29.3%; p < 0.001) and lymphadenectomy (stage I, 4.2%; stage II, 18.5%; stage III, 37.7%; p < 0.001) [Table 2]. The rate of R0 resection was lower for patients with stage III (70.8%) disease compared with those with stage I (85.4%; p = 0.002) and stage II disease (89.8%; p < 0.001), but there was no significant difference between stage I and II (p = 0.195).
TABLE 2.
Treatment characteristics of patients with non-metastatic ACC, stratified by stage of disease
| All patients (N = 1175) |
Stage I (n = 113) | Stage II (n = 539) | Stage III (n = 523) | p value | |
|---|---|---|---|---|---|
| Surgical approach | |||||
| Minimally invasive | 288 (27.9) | 70 (69.3) | 138 (28.4) | 80 (17.9) | < 0.001 |
| MIS converted to open | 51 (4.9) | 1 (1.0) | 31 (6.4) | 19 (4.3) | |
| Open | 694 (67.2) | 30 (29.7) | 317 (65.2) | 347 (77.8) | |
| Extent of resection | |||||
| Adrenalectomy only | 921 (78.4) | 108 (95.6) | 443 (82.2) | 370 (70.8) | < 0.001 |
| En bloc multi-organ resection | 254 (21.6) | 5 (4.4) | 96 (17.8) | 153 (29.3) | |
| Lymphadenectomy performed | |||||
| No | 688 (74.1) | 91 (95.8) | 331 (81.5) | 266 (62.3) | < 0.001 |
| Yes | 240 (25.9) | 4 (4.2) | 75 (18.5) | 161 (37.7) | |
| Margin status | |||||
| R0 | 866 (80.9) | 88 (85.4) | 441 (89.8) | 337 (70.8) | < 0.001 |
| R1 or R2 | 204 (19.1) | 15 (14.6) | 50 (10.2) | 139 (29.2) | |
| Adjuvant therapy | |||||
| None | 539 (48.3) | 68 (60.7) | 276 (53.5) | 195 (40.0) | < 0.001 |
| Radiation only | 104 (9.3) | 11 (9.8) | 41 (8.0) | 52 (10.7) | |
| Chemotherapy only | 275 (24.6) | 21 (18.8) | 125 (24.2) | 129 (26.4) | |
| Radiation and chemotherapy | 198 (17.7) | 12 (10.7) | 74 (14.3) | 112 (23.0) |
Data are expressed as n (%)
Bold values indicate p < 0.05
ACC adrenocortical carcinoma, MIS minimally invasive surgery
Univariable and multivariable logistic regression were utilized to determine associations between patient demographic and tumor characteristics and selected surgical approach. On multivariable analysis, only stage (stage II odds ratio [OR] 0.569, 95% confidence interval [CI] 0.338–0.960, p = 0.034; stage III OR 0.303, 95% CI 0.177–0.518, p < 0.001) and tumor size (OR 0.984, 95% CI 0.981–0.9888; p < 0.001) were associated with the open surgical approach and higher Charlson–Deyo score with laparoscopy (OR 0.225, 95% CI 0.141–0.358; p < 0.001). For lymphadenectomy, female sex was associated with lower likelihood (OR 0.586, 95% CI 0.425–0.807; p = 0.001), while tumor size (OR 1.002, 95% CI 1.001–1.003; p = 0.001) and advanced stage (stage II OR 5.697, 95% CI 1.736–18.694, p = 0.004; stage III OR 14.688, 95% CI 4.528–47.641, p < 0.001) were associated with higher likelihood. For extent of resection, females were again less likely to undergo en bloc multi-organ resection (OR 0.611, 95% CI 0.455–0.820; p = 0.001), whereas patients with more advanced tumors were more likely (stage II OR 5.252, 95% CI 1.874–14.721, p = 0.002; stage III OR 10.069, 95% CI 03.611–28.073, p < 0.001). There were no differences in surgical approach by race or insurance status.
Adjuvant Therapy
Overall, 24.6% of patients received adjuvant chemotherapy only, 9.3% received adjuvant radiation only, and 17.7% received both adjuvant chemotherapy and radiotherapy (Table 2). Utilization of adjuvant therapies varied based on patient and tumor characteristics. Treated patients were significantly younger (radiation: 51.8 vs. 54.9 years, p = 0.002; chemotherapy: 50.1 vs. 57.2 years, p < 0.001). Female patients were more likely to receive chemotherapy (45.6% of females vs. 37.7% of males; p = 0.009). Those receiving chemotherapy were more likely to have private insurance (chemotherapy: 68.1% vs. 48.4%; p < 0.001), presented with more advanced disease (p < 0.001) and more frequently had positive resection margins (23.7% vs. 16.1%; p = 0.002). Patients treated with adjuvant radiation presented with more advanced disease (p < 0.001) and more frequently had positive resection margins (33.3% vs. 13.8%; p < 0.001).
Margin Status by Surgical Approach
Logistic regression was utilized to determine associations between surgical technique and margin status as a potential driver of downstream survival. On univariable regression, there was no association between surgical approach (ITT, OR 0.991, 95% CI 0.699–1.407; p = 0.962) or extent of resection (OR 1.028, 95% CI 0.711–1.485; p = 0.885) with margin status. There was also no association between surgical approach and margin status when using an alternative ‘as treated’ definition of surgical approach, where laparoscopic to open conversions were counted as open procedures (OR 0.800, 95% CI 0.548–1.166; p = 0.246). Multivariable logistic regression incorporating stage, extent of resection, and surgical approach again found positive margin status was only associated with stage III disease (OR 2.575, 95% CI 1.356–4.891; p = 0.004). Similarly, there was no relationship on univariable or multivariable analysis between margin status and surgical approach or extent when stratified by stage.
Unadjusted Survival
The median follow-up time was 37.1 months (IQR 51.2), and the overall median survival was 77.1 months (95% CI 65.2–91.4). One-year survival was 85.9% (95% CI 83.6–87.8%), 3-year survival was 64.0% (95% CI 60.9–66.8%), and 5-year survival was 54.7% (95% CI 51.4–57.9%). At each timepoint, the survival rate decreased with increasing stage (p < 0.001). For instance, 3-year survival was as follows: stage I: 75.4% (95% CI 65.5–82.8%); stage II: 72.6% (95% CI 68.3–76.4%); stage III: 52.1% (95% CI 47.3–56.7%).
Kaplan–Meier survival curves were created, stratifying by stage (Fig. 1A-D), surgical approach (Fig. 2A-C), and adjuvant therapy. Stratifying by stage at presentation, there was no difference in unadjusted survival by extent of surgical resection (p = 0.226), lymphadenectomy (p = 0.815), use of adjuvant radiation (p = 0.436), or use of adjuvant chemotherapy (p = 0.329). Open surgery was associated with an unadjusted survival advantage over minimally invasive surgery (p = 0.002). Stratified by stage, there was no difference in unadjusted survival by surgical approach in stage I or II disease; however, in stage III disease, survival was significantly improved with open surgery over laparoscopic surgery (p = 0.001).
FIG. 1.
Kaplan–Meier curves for overall survival, stratified by adjuvant therapy, among patients with A stage I disease; B stage II disease; C stage III disease; and D all stages. ACC adrenocortical carcinoma
FIG. 2.
Kaplan–Meier curves for overall survival with 95% CIs, stratified by A surgical technique; B extent of surgery; and C lymphadenectomy. CI confidence interval
Cox Proportional Hazards Modeling
Univariable and multivariable Cox proportional hazards models were developed to determine associations between clinical and treatment characteristics and overall survival; missing values were imputed (Table 3). On univariable analysis, increasing age (hazard ratio [HR] 1.018, 95% CI 1.012–1.024; p < 0.001), Medicare insurance (HR 1.432, 95% CI 1.178–1.740; p < 0.001), higher Charlson–Deyo score (HR 1.247, 95% CI 1.103–1.411; p < 0.001), stage III disease (HR 2.326, 95% CI 1.584–3.240; p < 0.001) and poorly differentiated tumors (HR 2.084, 95% CI 1.054–4.119; p = 0.035) were associated with worse prognosis. Treatment with en bloc multiorgan resection (HR 1.230, 95% CI 1.055–1.601; p = 0.014), performance of lymphadenectomy (HR 1.245, 95% CI 1.004–1.543; p = 0.045), positive surgical margins (HR 1.822, 95% CI 1.476–2.348; p < 0.001), and adjuvant radiation (HR 1.359, 95% CI 1.005–1.837; p = 0.046) were also associated with worse survival. Adjuvant chemotherapy (HR 1.112, 95% CI 0.886–1.395; p = 0.362) and adjuvant chemoradiation (HR 0.987, 95% CI 0.761–1.279; p = 0.918) were not associated with survival.
TABLE 3.
Univariable and multivariable Cox proportional hazards models evaluating the associations between clinical and treatment characteristics and survival in non-metastatic ACC
| Univariable |
Multivariable |
|||||
|---|---|---|---|---|---|---|
| HR | 95% CI | p value | HR | 95% CI | p value | |
| Age | 1.018 | 1.012–1.024 | < 0.001 | 1.019 | 1.013–1.026 | < 0.001 |
| Sex | ||||||
| Male | Ref | – | – | Ref | – | – |
| Female | 0.908 | 0.757–1.088 | 0.296 | 0.917 | 0.761–1.106 | 0.365 |
| Race | ||||||
| White | Ref | – | – | Ref | – | – |
| Black | 0.843 | 0.607–1.170 | 0.307 | 0.872 | 0.626–1.215 | 0.419 |
| Other | 1.025 | 0.654–1.605 | 0.915 | 1.121 | 0.712–1.767 | 0.621 |
| Insurance | ||||||
| Private | Ref | – | – | Ref | – | – |
| Medicare | 1.432 | 1.178–1.740 | < 0.001 | – | – | – |
| Medicaid | 0.988 | 0.677–1.441 | 0.950 | – | – | – |
| Other government | 1.832 | 1.010–3.053 | 0.020 | – | – | – |
| Uninsured | 1.001 | 0.613–1.634 | 0.998 | – | – | – |
| Charlson–Deyo Score, per point | 1.247 | 1.103–1.411 | < 0.001 | 1.121 | 0.986–1.272 | 0.075 |
| Stage of disease | Ref | – | – | |||
| I | Ref | – | – | |||
| II | 1.173 | 0.814–1.689 | 0.391 | 1.294 | 0.891–1.879 | 0.176 |
| III | 2.266 | 1.584–3.240 | < 0.001 | 2.421 | 1.649–3.555 | < 0.001 |
| Pathologic grade | ||||||
| Well-differentiated | Ref | – | – | – | – | – |
| Moderately differentiated | 0.951 | 0.404–2.241 | 0.909 | – | – | – |
| Poorly differentiated | 2.084 | 1.054–4.119 | 0.035 | – | – | – |
| Undifferentiated | 1.562 | 0.758–3.216 | 0.227 | – | – | – |
| Surgical approach | – | – | ||||
| Open | Ref | – | – | Ref | – | – |
| MIS | 1.101 | 0.908–1.335 | 0.328 | 1.329 | 1.072–1.648 | 0.010 |
| Type of resection | ||||||
| Adrenalectomy | Ref | – | – | Ref | – | – |
| En bloc multi-organ resection | 1.230 | 1.055–1.601 | 0.014 | 1.189 | 0.945–1.495 | 0.140 |
| Lymphadenectomy | ||||||
| No | Ref | – | – | Ref | – | – |
| Yes | 1.245 | 1.004–1.543 | 0.045 | 1.039 | 0.812–1.330 | 0.759 |
| Margins | ||||||
| R0 | Ref | – | Ref | – | – | |
| R1 or R2 | 1.822 | 1.476–2.248 | < 0.001 | 1.587 | 1.260–1.999 | < 0.001 |
| Adjuvant therapy | ||||||
| None | Ref | – | – | Ref | – | – |
| Radiation | 1.359 | 1.005–1.837 | 0.046 | 1.117 | 0.814–1.532 | 0.494 |
| Chemotherapy | 1.112 | 0.886–1.395 | 0.362 | 1.190 | 0.937–1.512 | 0.154 |
| Radiation and chemotherapy | 0.987 | 0.761–1.279 | 0.918 | 0.892 | 0.674–1.180 | 0.424 |
Bold values indicate p < 0.05
ACC adrenocortical carcinoma, HR hazard ratio, CI confidence interval, Ref reference
On multivariable modeling, increasing age (HR 1.019, 95% CI 1.013–1.026; p < 0.001), increasing stage at presentation (stage II: HR 1.294, 95% CI 0.891–1.879, p = 0.176; stage III: HR 2.421, 95% CI 1.649–3.555; p < 0.001), minimally invasive approach (HR 1.329; 95% CI 1.072–1.648; p = 0.010), and positive surgical margins (HR 1.587, 95% CI 1.260–1.999; p < 0.001) portended worse survival. Extent of surgery, lymphadenectomy, and adjuvant chemotherapy and radiotherapy were not significantly associated with survival.
Stratification by stage yielded slight variation in these results. In stage I disease, only age was associated with survival (HR 1.041, 95% CI 1.005–1.078; p = 0.026). In stage II disease, age (HR 1.021, 95% CI 1.011–1.031; p < 0.001) and Charlson–Deyo score (HR 1.393, 95% CI 1.153–1.684; p = 0.001) were associated with survival. In stage III disease, age (HR 1.016, 95% CI 1.007–1.026; p = 0.001), positive margins (HR 1.584; 95% CI 1.193–2.102; p = 0.001), and minimally invasive approach (HR 1.548, 95% CI 1.126–2.130; p = 0.007) were associated with worse survival. Neither lymphadenectomy nor extent of surgery were significantly associated with survival at any stage of disease.
Additional analyses were performed to determine the association between the number of lymph nodes resected and survival outcomes. Resection of four or more lymph nodes was not significantly associated with survival on univariable analysis (HR 1.131, 95% CI 0.828–1.544; p = 0.438) or on multivariable analysis (HR 0.901, 95% CI 0.636–1.276; p = 0.556). When lymphadenectomy was categorized as resection of no nodes, resection of one to three nodes, and resection of four or more nodes, both resection of one to three lymph nodes (HR 1.254, 95% CI 0.944–1.666; p = 0.117) and resection of four or more lymph nodes were not significantly associated with survival compared with no lymph node resection (HR 1.223, 95% CI 0.892–1.677; p = 0.211). On multivariable analysis with both granular alternative definitions of lymphadenectomy, only age, advanced stage, positive margins, and laparoscopic approach otherwise remained significantly associated with worse survival, similar to the original model.
To isolate the effect of surgical treatment, a model was developed incorporating demographic and clinical characteristics, but excluding operative approach. This model incorporated the same patient demographic and diagnostic information as above (age, sex, stage, race, comorbidities) with information on adjuvant therapies (chemotherapy, radiation therapy, or chemoradiation), stratified by surgical margins, the primary factor by which teams will generally consider appropriate adjuvant therapies. In patients with negative resection margins, age (HR 1.021, 95% CI 1.013–1.029; p < 0.001), stage (stage II: HR 1.375, 95% CI 0.868–2.178, p = 0.175; stage III: HR 2.558, 95% CI 1.610–4.062, p < 0.001), and Charlson–Deyo score (HR 1.196, 95% CI 1.023–1.398; p = 0.025) were associated with worse survival. When resection margins were positive, female sex (HR 0.601, 95% CI 0.404–0.895; p = 0.012) and adjuvant chemoradiation (HR 0.475, 95% CI 0.288–0.783; p = 0.004) were associated with improved survival. Adjuvant chemotherapy alone (HR 0.749, 95% CI 0.458–1.225; p = 0.250) and adjuvant radiation alone (HR 0.614, 95% CI 0.353–1.068; p = 0.084) were not associated with improved survival in margin-positive resection, and nor were stage or Charlson–Deyo score.
Patients who received partial adrenalectomy (n = 253) were excluded from the original model, given it is not current standard of care. A separate multivariable survival analysis was performed to compare partial adrenalectomy outcomes with total adrenalectomy and en bloc multi-organ resection. Across the entire cohort, again age, stage, positive surgical margins, and open approach were significantly associated with worse survival. Partial adrenalectomy was not significantly associated with survival compared with total adrenalectomy across all patients (HR 1.179, 95% CI 0.954–1.456; p = 0.128), nor when the analysis was stratified by stage for stage I (partial adrenalectomy, n = 37 patients; HR 0.982, 95% CI 0.450–2.141; p = 0.962) or stage III disease (partial adrenalectomy, n = 134; HR 0.979, 95% CI 0.694–1.380; p = 0.903). However, partial adrenalectomy was associated with worse survival in stage II patients (partial adrenalectomy, n = 82; HR 1.529, 95% CI 01.135–2.059; p = 0.005).
DISCUSSION
In non-metastatic ACC, the optimal extent and approach to surgery and the role of adjuvant therapy remain controversial due to the lack of robust supporting data. As with other rare diseases, the dearth of prospective clinical trial data means that large retrospective cohort studies provide the highest-quality evidence for clinical management. While multiple recent studies have utilized the NCDB to analyze aspects of surgical management of ACC, most have focused on a single aspect of ACC care, such as laparoscopic versus open resection, the utility of lymphadenectomy, or postoperative radiation.16,25-29 Our study expands upon this prior work by holistically evaluating multiple pre- and postoperative decision points across all patients within the NCDB over the last 10 years. We found that the most important predictors of survival were patient age, stage at diagnosis, resection margins, and open surgical approach, consistent with prior studies. Multi-organ resection and lymphadenectomy were not associated with survival, and nor were adjuvant radiation or chemotherapy. However, two notable exceptions were that (1) in patients with stage I–II disease, open versus minimally invasive surgery was not associated with survival; and (2) in patients with positive resection margins, adjuvant chemoradiation was associated with improved survival.
We found that complete resection of the primary tumor was the most important modifiable risk factor in non-metastatic ACC. Compared with R0 margins, R1 or R2 margins were associated with a significantly worse median survival (34.6 vs. 97.2 months; p < 0.001). Few would debate the inextricable link between the extent of surgery and complete oncologic resection, however more aggressive and extensive surgery is also often linked to worse complication rates.30 Furthermore, frequent distant recurrence of ACC following reported R0 resections suggests that early micrometastatic disease may significantly impact outcomes despite effective surgery.3,9,10 Thus, we suggest that the ideal surgical decision making in non-metastatic ACC strives for the goal of achieving grossly negative margins while performing the least intensive operation possible. Some have advised en bloc resection as a default in the pursuit of negative mar-gins;31 however, our findings indicate that this likely constitutes overtreatment unless a tumor is clearly invading an adjacent structure. In our analysis, en bloc resection had no impact on survival, indicating that such aggressive surgery likely has no benefit unless it is clearly necessary to attain negative margins due to tumor extension into adjacent organs. Additionally, more extensive surgery and surgical approach had no relationship with margin status, indicating that differences in tumor biology and extent rather than technical differences in surgical management are the primary drivers of resectability.
In our cohort, we did find that the minimally invasive surgical approach was associated with worse overall survival. On analysis stratified by stage, we found no difference in survival based on open versus minimally invasive techniques in stage I and II disease. However, for stage III tumors, open adrenalectomy was associated with a survival advantage. Notably, minimally invasive surgery appeared to be associated with worse outcomes independent of margin status. Bearing in mind that this analysis is inherently limited to associations and does not establish causality, other studies have posited that the minimally invasive technique may be associated with a higher risk of capsular rupture and peritoneal carcinomatosis.7 Our finding is consistent with society guidelines, which recommend the open surgical technique for larger tumors with local invasion or when ACC is suspected, although suggested tumor size cut-offs vary.31,32 Nonetheless, the emphasis of most guidelines is on achieving an R0 resection with intact tumor capsule, regardless of technique.
Interestingly, we found that lymphadenectomy was not associated with survival in ACC. This result contrasts findings of a recent meta-analysis of non-metastatic ACC, in which lymphadenectomy was associated with a significantly decreased likelihood of death.16,21-23 Current European Society of Endocrinology guidelines also suggest regional lymphadenectomy be routinely performed at the time of primary tumor resection for ACC.32 Our finding that lymphadenectomy does not improve overall survival may be due to differing database definitions of lymphadenectomy, however using alternative numeric cut-offs for what constitutes a true lymphadenectomy did not alter our results. The inefficacy of lymphadenectomy in our study possibly represents the presence of early distant micrometastasis in ACC, which may not be adequately addressed with this type of resection. Notably, we were unable to assess the impact of lymphadenectomy on locoregional control, as the NCDB does not capture data on recurrence or disease-free survival.
In addition to the extent of surgery, we evaluated the impact of adjuvant radiation and chemotherapy on survival. This is a particularly challenging area of investigation in retrospective cohort studies, as decisions about adjuvant therapy are influenced by tumor stage, margin status, and clinical details that may not be captured in a large database. We found no significant association between adjuvant chemotherapy and radiation therapy and survival, both overall and when stratified by stage. However, when stratified by margin status, adjuvant chemoradiation was associated with a significantly lower risk of mortality in patients with positive surgical margins. Adjuvant radiation and chemotherapy used alone did not show any significant survival benefit regardless of stage or margin status.
This study is limited primarily by the selection bias inherent to its retrospective design. While prospective randomized controlled trials would be ideal, they are generally not practical due to the rarity of ACC. Additionally, the NCDB is restricted to data collection from Commission on Cancer-accredited hospitals, which may limit the generalizability of these findings. Nevertheless, the NCDB represents a broad national sampling and captures 72% of new cancer diagnoses in the United States. Information regarding disease recurrence and specific chemotherapeutic regimens is not available within the NCDB, meaning that recurrence-free survival and specific adjuvant regimens cannot be assessed. Finally, disease-specific characteristics such as hormone production and tumor Ki-67 index, and treatment-specific characteristics such as chemotherapeutic regimen, mitotane levels, and duration of chemotherapy could not be evaluated.
CONCLUSION
This study demonstrated that disease biology and the oncologic integrity of resection dictate outcomes for patients with non-metastatic ACC. While open surgery is likely superior in patients with stage III disease, our data suggest that the extent of surgery and lymphadenectomy may have a minimal impact on disease trajectory in most patients. Adjuvant chemoradiation likely plays a role augmenting surgical resection in the setting of positive margins. In the absence of therapeutic advancements in the care of ACC, earlier identification and intervention on suspicious adrenal tumors likely represents the most effective means of improving overall survival for patients with ACC.
FUNDING
HW received funding from the National Institutes of Health, NCI grant #K08 CA270385. JMSB received funding from the NIH T32 Training Program in Surgical Oncology Research at Penn, grant #5T32CA251063-02
Footnotes
DISCLOSURE Jesse E. Passman, Wajid Amjad, Jacqueline M. Soegaard Ballester, Sara P. Ginzberg, and Heather Wachtel declare no conflicts of interest.
REFERENCES
- 1.Golden SH, Robinson KA, Saldanha I, Anton B, Ladenson PW. Prevalence and incidence of endocrine and metabolic disorders in the United States: a comprehensive review. J Clin Endocrinol Metab. 2009;94(6):1853–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sharma E, Dahal S, Sharma P, Bhandari A, Gupta V, Amgai B, Dahal S. The characteristics and trends in adrenocortical carcinoma: a United States population based study. J Clin Med Res. 2018;10(8):636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kebebew E, Reiff E, Duh Q-Y, Clark OH, McMillan A. Extent of disease at presentation and outcome for adrenocortical carcinoma: Have we made progress? World J Surg. 2006;30(5):872–8. [DOI] [PubMed] [Google Scholar]
- 4.Wooten MD, King DK. Adrenal cortical carcinoma. Epidemiology and treatment with mitotane and a review of the literature. Cancer. 1993;72(11):3145–55. [DOI] [PubMed] [Google Scholar]
- 5.Luton J-P, Cerdas S, Billaud L, et al. Clinical features of adrenocortical carcinoma, prognostic factors, and the effect of mitotane therapy. N Engl J Med. 1990;322(17):1195–201. [DOI] [PubMed] [Google Scholar]
- 6.Fassnacht M, Allolio B. Clinical management of adrenocortical carcinoma. Best Pract Res Clin Endocrinol Metab. 2009;23(2):273–89. [DOI] [PubMed] [Google Scholar]
- 7.Else T, Kim AC, Sabolch A, et al. Adrenocortical carcinoma. Endocr Rev. 2014;35(2):282–326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Vassilopoulou-Sellin R, Schultz PN. Adrenocortical carcinoma: clinical outcome at the end of the 20th century. Cancer. 2001;92(5):1113–21. [DOI] [PubMed] [Google Scholar]
- 9.Bilimoria KY, Shen WT, Elaraj D, Bentrem DJ, Winchester DJ, Kebebew E, Sturgeon C. Adrenocortical carcinoma in the United States: treatment utilization and prognostic factors. Cancer. 2008;113(11):3130–6. [DOI] [PubMed] [Google Scholar]
- 10.Datta J, Roses RE. Surgical management of adrenocortical carcinoma: an evidence-based approach. Surg Oncol Clin. 2016;25(1):153–70. [DOI] [PubMed] [Google Scholar]
- 11.Bellantone R, Ferrante A, Boscherini M, et al. Role of reoperation in recurrence of adrenal cortical carcinoma: results from 188 cases collected in the Italian National Registry for Adrenal Cortical Carcinoma. Surgery. 1997;122(6):1212–8. [DOI] [PubMed] [Google Scholar]
- 12.Porpiglia F, Fiori C, Daffara F, et al. Does nephrectomy during radical adrenalectomy for stage II adrenocortical cancer affect patient outcome? J Endocrinol Investig. 2016;39(4):465–71. [DOI] [PubMed] [Google Scholar]
- 13.Porpiglia F, Fiori C, Daffara F, Zaggia B, Scarpa RM, Terzolo M. Does nephrectomy during radical adrenalectomy for adrenocortical cancer affect oncological results? J Urol. 2010;183(4S):e11–e11. [DOI] [PubMed] [Google Scholar]
- 14.Marincola Smith P, Kiernan CM, Tran TB, et al. Role of additional organ resection in adrenocortical carcinoma: analysis of 167 patients from the US adrenocortical carcinoma database. Ann Surg Oncol. 2018;25:2308–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gaujoux S, Brennan MF. Recommendation for standardized surgical management of primary adrenocortical carcinoma. Surgery. 2012;152(1):123–32. [DOI] [PubMed] [Google Scholar]
- 16.Hendricks A, Müller S, Fassnacht M, Germer C-T, Wiegering VA, Wiegering A, Reibetanz J. Impact of lymphadenectomy on the oncologic outcome of patients with adrenocortical carcinoma: a systematic review and meta-analysis. Cancers. 2022;14(2):291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Jasim S, Habra MA. Management of adrenocortical carcinoma. Curr Oncol Rep. 2019;21(3):1–11. [DOI] [PubMed] [Google Scholar]
- 18.Terzolo M, Angeli A, Fassnacht M, et al. Adjuvant mitotane treatment for adrenocortical carcinoma. N Engl J Med. 2007;356(23):2372–80. [DOI] [PubMed] [Google Scholar]
- 19.Boffa DJ, Rosen JE, Mallin K, et al. Using the national cancer database for outcomes research: a review. JAMA Oncol. 2017;3(12):1722–8. [DOI] [PubMed] [Google Scholar]
- 20.Mallin K, Browner A, Palis B, et al. Incident cases captured in the National Cancer Database compared with those in US population based central cancer registries in 2012–2014. Ann Surg Oncol. 2019;26(6):1604–12. [DOI] [PubMed] [Google Scholar]
- 21.Reibetanz J, Jurowich C, Erdogan I, et al. Impact of lymphadenectomy on the oncologic outcome of patients with adrenocortical carcinoma. Ann Surg. 2012;255(2):363–9. [DOI] [PubMed] [Google Scholar]
- 22.Gerry JM, Tran TB, Postlewait LM, et al. Lymphadenectomy for adrenocortical carcinoma: Is there a therapeutic benefit? Ann Surg Oncol. 2016;23(5):708–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Saade N, Sadler C, Goldfarb M. Impact of regional lymph node dissection on disease specific survival in adrenal cortical carcinoma. Horm Metab Res. 2015;47(11):820–5. [DOI] [PubMed] [Google Scholar]
- 24.Stata Statistical Software. Release 17 [computer program]. College Station: StataCorp LLC; 2021. [Google Scholar]
- 25.Ginsburg KB, Chandra AA, Handorf EA, et al. Association of surgical approach with treatment burden, oncological effectiveness, and perioperative morbidity in adrenocortical carcinoma. Clin Genitourin Cancer. 2022;20(5):497.e491–497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Shah M, NeMoyer RE, Kashyap R, et al. Surgical resection for adrenocortical carcinoma: current trends affecting survival. J Surg Oncol. 2022;125:1224–30. [DOI] [PubMed] [Google Scholar]
- 27.Tseng J, DiPeri T, Chen Y, et al. Adrenocortical carcinoma: the value of lymphadenectomy. Ann Surg Oncol. 2022;29(3):1965–70. [DOI] [PubMed] [Google Scholar]
- 28.Hue JJ, Ahorukomeye P, Bingmer K, et al. A comparison of robotic and laparoscopic minimally invasive adrenalectomy for adrenal malignancies. Surg Endosc. 2022;36(7):5374–81. [DOI] [PubMed] [Google Scholar]
- 29.Delman AM, Turner KM, Griffith A, Schepers E, Ammann AM, Holm TM. Minimally invasive surgery for resectable adrenocortical carcinoma: a nationwide analysis. J Surg Res. 2022;279:200–7. [DOI] [PubMed] [Google Scholar]
- 30.Lee J, El-Tamer M, Schifftner T, et al. Open and laparoscopic adrenalectomy: analysis of the national surgical quality improvement program. J Am Coll Surg. 2008;206(5):953–9. [DOI] [PubMed] [Google Scholar]
- 31.Yip L, Duh Q-Y, Wachtel H, et al. American association of endocrine surgeons guidelines for adrenalectomy: executive summary. JAMA Surg. 2022;157(10):870–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Fassnacht M, Dekkers OM, Else T, et al. European society of endocrinology clinical practice guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European network for the study of adrenal tumors. Eur J Endocrinol. 2018;179(4):G1–46. [DOI] [PubMed] [Google Scholar]