Received: 25 May 2020
Accepted: 14 July 2020
Check for updates
Critical analysis of lymph node examination in patients undergoing curative-intent resection for adrenocortical carcinoma
Benjamin W. Deschner MD İD Zachary E. Stiles DO İD Olivia M. DeLozier MD
Justin A. Drake MD Miriam Tsao MD @ | Evan S. Glazer MD, PHD, FACS
|
Jeremiah L. Deneve DO İD Danny Yakoub MD, PhD Paxton V. Dickson MD, FACS
Division of Surgical Oncology, Department of Surgery, University of Tennessee Health Science Center, Memphis, Tennessee
Correspondence
Paxton V. Dickson, MD, FACS, 910 Madison Ave., Suite 301, Memphis, TN 38163. Email: pdickso1@uthsc.edu
Abstract
Background: Adrenocortical carcinoma (ACC) is a rare tumor and the role of lymph node dissection remains ill-defined. This study evaluates the effect of no dal examination on prognosis and survival in patients undergoing curative-intent resection of ACC.
Methods: The National Cancer Database (2004-2015) was queried for patients undergoing margin-negative resection for ACC. Patients with distant metastases, neoadjuvant therapy, multivisceral resection and T4 tumors were excluded.
Results: Among 897 patients, 147 (16.4%) had lymph nodes examined. Factors as- sociated with lymph node examination included increasing tumor size (P <. 001), extra-adrenal extension (P <. 001), open operation (P <. 001), and resection at an academic facility (P = . 003). Lymph node metastasis was significantly associated with extra-adrenal tumor extension (P =. 04). Lymph node harvest, regardless of the number of nodes examined, was not associated with a survival benefit. Median overall survival was incrementally worse with increasing number of positive lymph nodes (88.2 months for NO, 34.9 months for 1-3 positive nodes, and 15.6 months for ≥4 positive nodes, P < . 001).
Conclusions: Lymph node harvest and lymph node metastasis were associated with more advanced tumors. Although nodal harvest did not offer a survival advantage, stratifying the nodal staging classification may provide important prognostic information.
KEYWORDS
adrenal cancer, lymphadenectomy, prognosis, staging
1 INTRODUCTION
Adrenocortical carcinoma (ACC) is an uncommon malignancy with an annual incidence of 1 to 2 per million.1-4 Margin-negative (RO) re- section is critical for potential cure in those with localized tumors. However, even among these patients, 5-year overall survival (OS) typically approaches only 60%.1,5-7 Given the rarity of ACC, several aspects of its surgical and medical management remain poorly un- derstood, including detailed prognostic and therapeutic implications of lymph node harvest during tumor resection.
Although a standard extent of lymph node retrieval is recommended for most intra-abdominal malignancies,8-10 it is not clearly defined for ACC.4,11 One limiting factor is the lack of a well-described and widely disseminated anatomically-based dissection to which surgeons should adhere. This is highlighted by the variety of definitions used to describe regional lymphadenectomy for ACC in studies evaluating its impact on prognosis and survival.5,12-15 Among these studies, reported rates of lymphadenectomy range from only 6% to 33%.3,5,12-17 Moreover, ex- amination of any lymph nodes has been alarmingly low, with reported rates less than 35%.3,5,12-17
Despite the low frequency of lymph node examination, nodal metastasis has been identified as an independent negative pre- dictor of survival.1,5,12-15,18 As such, node classification is a component of internationally accepted staging systems for ACC.19,20 However, the heterogeneity of inclusion criteria among studies evaluating lymph node examination in resection specimens has limited full appreciation of its influence on survival.1,5,12-15,18 The purpose of this study was to determine factors associated with lymph node retrieval and critically analyze the prognostic and therapeutic role of lymph node examination in patients undergoing curative-intent resection of ACC.
2 MATERIALS AND METHODS
2.1 Data source and patient selection
The National Cancer Database (NCDB) is a hospital-based, clinical database estimated to capture over two-thirds of annual cancer di- agnoses nationwide. Pertinent patient, diagnostic, pathologic, and treatment datapoints and outcomes variables are collected and available for analysis. This study was deemed exempt by the Uni- versity of Tennessee Health Science Center institutional review board as the data included is appropriately deidentified.
The 2004 to 2015 NCDB was queried based on histologic code for patients who underwent curative-intent resection of ACC. For pur- poses of this study, curative-intent was defined as margin-negative (R0) resection. Patients with tumors invading adjacent organs (T4), metastatic disease, multivisceral resection (MVR), or treatment with neoadjuvant radiation or systemic therapy were excluded. In addition, patients with missing values and those treated for other malignancies were excluded.
2.2 Lymph node examination
Patients meeting inclusion criteria were further analyzed to determine if lymph node examination (the identification of ≥1 lymph nodes on pa- thology) was performed. Patients with lymph nodes examined, regardless of the number, were contrasted to those patients without lymph node examination with regard to patient and tumor characteristics. Outcomes, measured as OS, were compared for the two groups.
Among patients who underwent lymph node examination, the presence of positive lymph node metastasis was further examined. Patients with positive nodes were compared with patients with ne- gative nodes to determine patient and tumor-related factors asso- ciated with nodal metastases. In addition, the effect of an increasing number of positive lymph nodes was further examined to investigate the utility of increased nodal sampling. A multivariable analysis was performed using a Cox proportional hazard model to better define the independent effect of lymph node examination and positivity when adjusted for known or suspected prognostic factors.
2.3 | Statistical analysis
Categorical variables were reported as actual values with relative frequencies and compared with the x? test or Fisher’s exact test when appropriate. Continuous variables were given as the median along with the interquartile range (IQR) and compared using the Wilcoxon rank-sum test. Patients with absent survival or follow-up data were excluded from outcomes analysis. Survival was assessed on a univariate level via the Kaplan-Meier method and the log rank test. For Cox regression analysis, factors were chosen de novo based on known or suspected prognostic value, and hazard ratio (HR) with confidence intervals (CIs) were gener- ated for comparison. Hazards were assumed to be proportional over time and this was verified by assessing Martingale residuals. All hypothesis testing was two-sided and a P <. 05 was considered statistically significant. All statistical analysis was performed using SAS version 9.4 (SAS Institute, Cary, NC).
3 RESULTS
3.1 | Patient characteristics
A total of 897 patients met inclusion criteria. Most patients were female (61.1%) and Caucasian (88.1%), with a median age of 52 years (IQR: 42-63 years). The median tumor size was 10.0 cm (IQR: 6.5-13.5 cm) and the majority of tumors were confined to the adrenal gland (72.5%). Median follow-up was 37.5 months.
3.2 | Predictors of lymph node examination
Among the 897 patients selected, 147 (16.4%) had at least one lymph node examined (Table 1). Of these patients, the median number of
| Variable | All patients (n = 897) | Lymph nodes not examined (n = 750) | Lymph nodes examined (n = 147) | P value |
|---|---|---|---|---|
| Age, y, median (IQR) | 52 (42-63) | 52 (42-63) | 54 (43-63) | .52 |
| Gender, n (%) | .15 | |||
| Male | 349 (38.9) | 284 (37.9) | 65 (44.2) | |
| Female | 548 (61.1) | 466 (62.1) | 82 (55.8) | |
| Race, n (%) | .78 | |||
| Caucasian | 790 (88.1) | 662 (88.3) | 128 (87.1) | |
| African-American | 75 (8.4) | 63 (8.4) | 12 (8.2) | |
| Other | 25 (2.8) | 19 (2.5) | 6 (4.1) | |
| Unknown | 7 (0.8) | 6 (0.8) | 1 (0.7) | |
| Facility type, n (%) | .003 | |||
| Community cancer program | 29 (3.2) | 21 (2.8) | 8 (5.4) | |
| Comprehensive community program | 225 (25.1) | 204 (27.2) | 21 (14.3) | |
| Academic/research program | 387 (43.1) | 308 (41.1) | 79 (53.7) | |
| Integrated network cancer program | 70 (7.8) | 60 (8.0) | 10 (6.8) | |
| Unknown | 186 (20.7) | 157 (20.9) | 29 (19.7) | |
| Comorbidity index, n (%) | .92 | |||
| 0 | 688 (76.7) | 577 (76.9) | 111 (75.5) | |
| 1 | 159 (17.7) | 132 (17.6) | 27 (18.4) | |
| 2 or greater | 50 (5.6) | 41 (5.5) | 9 (6.1) | |
| Tumor size, cm, median (IQR) | 10.0 (6.5-13.5) | 9.2 (6.2-13.0) | 12.0 (9.2-15.3) | <. 001 |
| Tumor extent, n (%) | <. 001 | |||
| Confined to adrenal | 650 (72.5) | 567 (75.6) | 83 (56.5) | |
| Extends beyond adrenal | 196 (21.9) | 141 (18.8) | 55 (37.4) | |
| Unknown | 51 (5.7) | 42 (5.6) | 9 (6.1) | |
| Tumor grade, n (%) | .27 | |||
| Well/moderately differentiated | 77 (8.6) | 68 (9.1) | 9 (6.1) | |
| Poorly differentiated/undifferentiated | 123 (13.7) | 98 (13.1) | 25 (17.0) | |
| Not determined | 697 (77.7) | 584 (77.9) | 113 (76.9) | |
| Laterality, n (%) | .06 | |||
| Right | 403 (44.9) | 347 (46.3) | 56 (38.1) | |
| Left | 485 (54.1) | 394 (52.5) | 91 (61.9) | |
| Unknown | 9 (1.0) | 9 (1.2) | 0 (0.0) | |
| Approach, n (%) | <. 001 | |||
| Open resection | 241 (26.9) | 186 (24.8) | 55 (37.4) | |
| Minimally invasive resection | 152 (17.0) | 142 (18.9) | 10 (6.8) | |
| Unknown approach | 504 (56.2) | 422 (56.3) | 82 (55.8) | |
| Number of nodes examined, median (IQR) | 2 (1-6) | NA | 2 (1-6) | NA |
| Unplanned readmission within 30 d of surgical discharge, n (%) | 40 (4.5) | 35 (4.7) | 5 (3.4) | .50 |
| Length of stay, d, median (IQR) | 4 (2-6) | 4 (2-6) | 5 (4-7) | <. 001 |
| 90-d mortality, n (%) | 28 (3.1) | 25 (3.3) | 3 (2.0) | .60 |
| Adjuvant systemic therapy (≤90 d after resection), n (%) | 200 (22.3) | 165 (22.0) | 35 (23.8) | .63 |
| Adjuvant radiation (≤90 d after resection), n (%) | 78 (8.7) | 62 (8.3) | 16 (10.9) | .30 |
Abbreviations: IQR, interquartile range; NA, not applicable.
| Variable | N0 (n = 123) | N1 (n = 24) | P value |
|---|---|---|---|
| Age, y, median (IQR) | 54 (41-61) | 51 (45-68) | .34 |
| Gender, n (%) | .47 | ||
| Male | 56 (45.5) | 9 (37.5) | |
| Female | 67 (54.5) | 15 (62.5) | |
| Race, n (%) | .13 | ||
| Caucasian | 107 (87.0) | 21 (87.5) | |
| African-American | 11 (8.9) | 1 (4.2) | |
| Other | 5 (4.1) | 1 (4.2) | |
| Unknown | 0 (0.0) | 1 (4.2) | |
| Facility type, n (%) | .72 | ||
| Community cancer program | 7 (5.7) | 1 (4.2) | |
| Comprehensive community program | 16 (13.0) | 5 (20.8) | |
| Academic/research program | 65 (52.9) | 14 (58.3) | |
| Integrated network cancer program | 9 (7.3) | 1 (4.2) | |
| Unknown | 26 (21.1) | 3 (12.5) | |
| Comorbidity index, n (%) | .09 | ||
| 0 | 97 (78.9) | 14 (58.3) | |
| 1 | 20 (16.3) | 7 (29.2) | |
| 2 or greater | 6 (4.9) | 3 (12.5) | |
| Tumor size, cm, median (IQR) | 12.0 (9.6-16.0) | 11.3 (8.5-14.3) | .54 |
| Tumor extent, n (%) | .04 | ||
| Confined to adrenal | 75 (61.0) | 8 (33.3) | |
| Extends beyond adrenal | 41 (33.3) | 14 (58.3) | |
| Unknown | 7 (5.7) | 2 (8.3) | |
| Tumor grade, n (%) | .07 | ||
| Well/moderately differentiated | 8 (6.5) | 1 (4.2) | |
| Poorly differentiated/undifferentiated | 17 (13.8) | 8 (33.3) | |
| Not determined | 98 (79.7) | 15 (62.5) | |
| Laterality, n (%) | .60 | ||
| Right | 48 (39.0) | 8 (33.3) | |
| Left | 75 (61.0) | 16 (66.7) | |
| Approach, n (%) | .19 | ||
| Open resection | 50 (40.7) | 5 (20.8) | |
| Minimally invasive resection | 8 (6.5) | 2 (8.3) | |
| Unknown approach | 65 (52.9) | 17 (70.8) | |
| Number of nodes examined, median (IQR) | 2 (1-5) | 5 (1-14) | .06 |
| Unplanned readmission within 30 d of surgical discharge, n (%) | 4 (3.3) | 1 (4.2) | .82 |
| Length of stay, median (IQR) | 6 (4-7) | 5 (4-7) | .40 |
| 90-D mortality, n (%) | 1 (0.8) | 2 (8.3) | .02 |
| Adjuvant systemic therapy (≤90 d resection), n (%) | 29 (23.6) | 6 (25.0) | .88 |
| Adjuvant radiation (≤90 d resection), n (%) | 14 (11.4) | 2 (8.3) | 1.0 |
Note: N0 = lymph nodes examined and were negative; N1 = lymph nodes examined with at least one positive.
Abbreviation: IQR, interquartile range.
WILEY
examined lymph nodes was 2 (IQR: 1-6). Factors associated with lymph node examination included facility type, tumor size, operative approach, and extra-adrenal extension (Table 1). Academic facilities examined lymph nodes in 79 out of 387 (20.4%) of specimens com- pared with 68 out of 510 (13.3%) at all other centers (P =. 003). Median tumor size was 12.0 cm in specimens with lymph nodes ex- amined versus 9.2 cm in those with lymph nodes not examined (P <. 001). Lymph nodes were detected more frequently in patients having open versus minimally invasive resection (55/241 [22.8%] vs 10/152 [6.6%], P <. 001) and in those with extra-adrenal extension (55/196 [28.1%] vs 83/650 [12.8%]; P <. 001). Lymph node ex- amination was associated with a greater length of stay (5 vs 4 days; P < . 001). There was no difference in 90-day mortality between pa- tients with and without lymph nodes examined.
3.3 | Predictors of lymph node positivity
Of the 147 patients who underwent lymph node examination, lymph node positivity was more common in tumors with extra-adrenal ex- tension when compared with tumors confined to the adrenal gland (14/55 [25.5%] vs 8/83 [9.6%]; P = . 04). No difference in either tumor size or laterality was detected when comparing node-positive (N1) and node-negative (NO) tumors (Table 2). When groups of patients with 1 to 3, 4 to 6, 7 to 9, and ≥10 lymph nodes were examined, a positive lymph node was detected in 11.8%, 15.0%, 20.0%, and 33.3% of these groups, respectively (Figure 1).
3.4 | Prognostic and therapeutic implications of lymph node examination
Multivariable analysis demonstrated that age, comorbid conditions, tumor size, extra-adrenal tumor extension, and not receiving adjuvant systemic chemotherapy were all significantly associated with decreased OS while N1 disease was the strongest independent predictor of decreased OS (HR: 2.00; 95% CI = 1.13-3.57; P =. 02) (Table 3). Notably, median OS was similar for patients who did not undergo lymph node examination (Nx) (81.3 months, 95% CI = 66.9- 91.5 months) and N0 patients (88.2 months, 95% CI = 60.2 months to not reached). This contrasted with just 23.4 months (95% CI = 13.2-40.9 months) OS for N1 patients (P =. 01) (Figure 2A). Among N0 patients, there was no difference in OS when greater numbers of nodes were sampled. Median OS for 1 to 3, 4 to 6, 7 to 9, and ≥10 nodes examined was 74.8 months (95% CI = 39.9 months to not reached), 91.2 months (95% CI = 63.9 months to not reached), 31.2 months (95% CI = 12.2 months to not reached), and 88.2 months (95% CI = 20.8 months to not reached), respectively (P =. 47) (Figure 2B).
When assigning an arbitrary number of ≥5 lymph nodes examined to define “lymphadenectomy,” and comparing this group to 1 to 4 lymph nodes examined, no difference in OS was identified (Figure 3A). Median OS for lymphadenectomy specimens was 62.0
35
30
25
20
%
15
10
5
0
1-3
4-6
7-9
≥10
Lymph nodes examined
months (95% CI = 20.8 months to not reached) compared with 74.8 months (95% CI = 40.1 months to not reached) for specimens with 1 to 4 lymph nodes examined (P = . 56). Furthermore, when stratifying into groupings by numbers of lymph nodes harvested, there was again no difference in OS (Figure 3B). Median OS for 1 to 3, 4 to 6, 7 to 9, and ≥10 lymph nodes harvested was 74.8 months (95% CI = 36.4 months to not reached), 91.2 months (95% CI = 32.1 months to not reached), 20.8 months (95% CI = 12.2 months to not reached), and 62.0 months (95% CI, 20.8 months to not reached; P= . 56), respectively.
3.5 Prognostic implications of resecting positive lymph nodes
When evaluating the impact of metastatic nodal burden, an increasing number of positive lymph nodes equated to a significant decline in OS
| Variable | 95% HR HR CI P value | ||
|---|---|---|---|
| Age | 1.02 | 1.01-1.03 | <. 001 |
| Male gender (vs female) | 0.82 | 0.66-1.01 | .06 |
| Comorbidity index ≥1 (vs 0) | 1.46 | 1.16-1.84 | .002 |
| Tumor size | 1.01 | 1.00-1.02 | .02 |
| Tumor extent beyond adrenal (vs confined to adrenal) | 1.57 | 1.24-1.99 | <. 001 |
| N1 (vs N0) | 2.00 | 1.13-3.57 | .02 |
| Lymph nodes not examined (vs N0) | 1.18 | 0.86-1.61 | .31 |
| Adjuvant systemic therapy (vs none) | 1.30 | 1.02-1.67 | .04 |
Note: N0 = Lymph nodes examined and were negative; N1 = Lymph nodes examined with at least one positive.
Abbreviations: CI, confidence interval; HR, hazard ratio.
(Figure 4A). Median OS in months was 88.2 (95% CI = 60.2 to not reached), 34.9 (95% CI = 13.2 to not reached), and 15.6 (95% CI = 3.4- 36.5) for patients categorized as N0, 1 to 3 nodes positive, and ≥4 nodes positive, respectively (P < . 001). Although 212 patients who underwent MVR were initially excluded from the dataset, when adding them to this specific analysis an even more pronounced effect was observed (Figure 4B). Median OS in months was then 79.5 (95% CI = 58.7-101.5), 29.2 (95% CI = 13.7-41.0), and 11.6 (95% CI = 3.8-27.5) for the three respective groups (P < . 001). When examined as a continuous variable, lymph node positivity translated to a HR of 1.09 (95% CI, 1.05-1.12; P < . 001) equating to a 9% increased hazard of death for each additional positive lymph node.
4 DISCUSSION
The existing literature regarding the role of lymphadenectomy for ACC is heterogenous and conflicting. To date, ten previous studies
have investigated lymph node examination in some capacity (Table 4).1,3,5,12-17,21 Definitions of lymphadenectomy have ranged from the examination of an arbitrary number of lymph nodes (≥4 or ≥5),12-15,21 to “surgeon intent,“5,12,22,23 to not being defined at all.3,16,17 In addition, inclusion criteria of patients evaluated among these studies differ widely by stage, presence of distant metastases, MVR, and margin status. Such variation likely contributes to the observed discrepancy in survival outcomes between these reports (Table 4). In this analysis, performance of lymph node examination and lymphadenectomy as well as the tumor-bearing status of these harvested nodes was comprehensively analyzed.
Selection criteria in this study were set in an attempt to provide the most accurate assessment of how lymph node examination im- pacts patients who undergo curative-intent resection. To this end, patients that underwent R1 resection were excluded, as margin positivity has been repeatedly demonstrated to be a dominant independent predictor of poor survival.1,14,15,18 Additionally, patients with T4 tumors or those that underwent MVR were excluded to
(A)
1.0
NO
N1
Survival probability
0.8
NX
0.6
0.4
0.2
0.0
NO
123
48
14
0
N1
24
6
2
0
NX
750
305
77
3
T
0
50
100
150
(B)
Months from diagnosis
1.0
1 - 3 Nodes Examined
4 - 6 Nodes Examined
0.8
Survival probability
7 - 9 Nodes Examined
≥ 10 Nodes Examined
L
0.6
0.4
0.2
0.0
1-3
82
48
28
15
7
3
0
4-6
17
14
10
6
3
1
0
7-9
8
4
3
2
1
0
10 or more
16
12
7
5
3
1
0
T
T
T
T
T
0
25
50
75
100
125
150
Months from diagnosis
avoid potential inclusion of non-regional lymph nodes that might dilute or alter the accuracy of benign lymph node harvest. Indeed, of the 212 patients with MVR excluded, lymph nodes were identified in 46.4%, compared with only 17.5% in non-MVR (P <. 001). This is si- milar to what Marincola Smith et al found in a 13-institution retro- spective database (47.1% vs 16.2%; P <. 001), again highlighting the possibility of non-regional lymph node harvest in these cases.24 A well-established consensus definition for anatomic regional lym- phadenectomy during resection for ACC is lacking, likely contributing to the low rates observed in both this and previous studies. In 1966, Merklin described regional nodal drainage patterns by injecting and tracing blue latex solution.25 More recently, radiographic patterns of regional nodal recurrences have been documented as indirect methods of mapping a target nodal basin.26,27 These reports have led to suggested definitions of lymphadenectomy to include periadrenal, renal hilar, ipsilateral para-aortic or paracaval nodes, those enlarged on preoperative imaging, and potentially superior mesenteric artery
or celiac axis lymph nodes.22,27-29 Collectively, these studies em- phasize the need for a multi-center investigation for definitive re- solution on the extent of lymphadenectomy.
This study presents a comprehensive analysis of clin- icopathologic factors associated with examination of ≥1 lymph nodes as well as lymph node positivity for ACC. These data corroborate prior reports that lymph node examination is associated with larger tumors,5,12-14 open resections,5 and extra-adrenal extension.14,21 Consistent with previous publications, this study also identifies the association of N1 disease with extra-adrenal extension and not with tumor size.14,15 The relationship of lymph node examination with treatment center type has not been described, and may be attribu- table to a distinct surgical approach at academic centers. Finally, given some controversy about the appropriateness of minimally in- vasive resection for ACC, it is a notable finding that lymph node detection was significantly higher for patients who underwent open resection.
(A)
1.0
1 - 4 Nodes Examined
0.8
5 or More Nodes Examined
Survival probability
0.6
0.4
0.2
0.0
1
101
61
38
21
10
4
0
2
46
27
16
10
6
2
0
0
25
50
75
100
125
150
(B)
Months from diagnosis
1.0
1 - 3 Nodes Examined
0.8
4 - 6 Nodes Examined
Survival probability
7 - 9 Nodes Examined
≥ 10 Nodes Examined
0.6
0.4
0.2
0.0
1-3
93
54
32
17
8
3
0
4-6
20
14
10
6
3
1
0
7-9
10
4
3
2
1
0
10 or more
24
16
9
6
4
2
0
T
T
0
25
50
75
100
125
150
Months from diagnosis
A recommended minimum number of harvested lymph nodes that defines adequate lymphadenectomy has been well-established in multiple intra-abdominal malignancies, and can affect prognosis and survival.8-10,30 These concepts are difficult to extrapolate to ACC given the lack of a defined nodal basin, as well as the uncertain indications and benefits of adjuvant chemotherapy. In a study by Panjwani et al, excision of ≥4 lymph nodes gained a significant in- crease in detection of a metastatic node when compared with exci- sion of 1 to 3 lymph nodes.15 Furthermore, when evaluating N0 patients, a survival benefit was also identified for those with ≥4 lymph nodes resected when compared with patients with 1 to 3 lymph nodes, signaling a target prognostic threshold for “adequate lymphadenectomy.” The current study did not exhibit these findings. The difference in the median number of lymph nodes examined in node-negative patients when compared with node-positive (2 vs 4) did demonstrate a nonsignificant trend. However, when further stratified into nodal harvest groupings, a directly proportional
relationship in the detection of a positive lymph node was revealed, with no plateau demarcating a minimum threshold for lymphade- nectomy (Figure 1). Similarly, no survival benefit among N0 patients was detected when stratifying these patients into nodal harvest groupings (Figure 2B). The difference in these results may partially be attributed to excluding R1 resections within the current study; these findings highlight the importance of patient selection in making nuanced conclusions with retrospective data. In the current dataset, there were only 15 patients with R1 resections that had lymph nodes examined, limiting meaningful analysis of how the number of lymph nodes impacted identification of positive nodes as well as OS in this group (data not shown).
There are conflicting reports regarding the therapeutic influence of lymphadenectomy for patients with ACC. At least three studies have documented improved OS for patients undergoing lymphade- nectomy versus those who did not.5,12,16 In two of these studies lymphadenectomy was defined- at least in part- by surgeon intent to
(A)
1.0
O Nodes Positive
1 - 3 Nodes Positive
Survival probability
0.8
≥ 4 Nodes Positive
0.6
0.4
0.2
0.0
0
123
78
48
28
14
5
0
1-3
18
9
6
3
2
1
0
4 or more
6
1
0
0
25
50
75
100
125
150
(B)
Months from diagnosis
1.0
O Nodes Positive
1 - 3 Nodes Positive
Survival probability
0.8
≥ 4 Nodes Positive
0.6
0.4
0.2
0.0
0
224
141
92
56
28
12
0
1-3
34
16
8
3
2
1
0
4 or more
14
4
1
0
0
25
50
75
100
125
150
Months from diagnosis
| Study | Year | # of patients | Median follow-up | Stages studied | Include MVR? | Include R1? | Include M1? | Definition of LAN | Rate with LNe (%) | Rate of N1 patients among those | Improved OS (RFS) (or DSS with LAN)? | DSS (LAN vs noLAN), mo | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Database | Rate of LAN (%) | Rate of N1 | who underwent LAN (%) | % 5Y OS LAN versus noLAN | ||||||||||||
| (LAN/ | patients among LNe (%) | |||||||||||||||
| noLAN), mo | ||||||||||||||||
| Icard et al3 | 2001 | 253 | AFCE | ⋯ | All | Yes | Yes | Yes | ND | ⋯ | 32.5 | ⋯ | ⋯ | No (P =. 06) | ⋯ | 42% versus |
| 36% | ||||||||||||||||
| (P = . 06) | ||||||||||||||||
| Bilimoria et al1 | 2008 | 3982 | NCDB | 24 | All | Yes | Yes | Yes | ND | 17.6 | ... | 26.5 | ... | ... | ... | ... |
| Reibetanz | 2012 | 283 | German ACC | 59/39 | ENSAT I-III | Yes | No | No | ≥5 LN and | 30.0 | 16.6 | 8.8 | 25.5 | Yes | 86.0 versus | ⋯ |
| et al12 | Registry | surgeon | 26.2 (N1 | |||||||||||||
| intent | group) | |||||||||||||||
| Tran et al | 2013 | 320ª | SEER | ... | AJCC III | Yes | Yes | No | ND | 29.1 | 26 | 38.7 | 35 | Yes (UVA of | ... | ... |
| and IV | Stage IV)b | |||||||||||||||
| Saade et al | 2015 | 259 | SEER | 64/62 | ENSAT I-III | Yes | Yes | No | ≥5 LN | 19.7 | 6.2 | 4.6 | 43.8 | No | 67.8 versus | ⋯ |
| 96.8 | ||||||||||||||||
| (P =. 73) | ||||||||||||||||
| Alanee | 2015 | 1732 | SEER | ⋯ | All | Yes | Yes | Yes | >5 LN | *** | 5.4 | ⋯ | *** | No | *** | *** |
| et al21 | ||||||||||||||||
| Nilubol | 2016 | 1525 | SEER | 13 | All (including | Yes | Yes | Yes | ≥4 LN | 16.0 | 8.4 | 12.8 | 33.8 | No | 94.5 versus | ⋯ |
| et al14 | debulking) | 121.9 | ||||||||||||||
| (P =. 30) | ||||||||||||||||
| Gerry et al5 | 2016 | 120 | 13-institution | 25/26 | All | Yes | No | No | Surgeon intent | 34.2 | 27 | 24.4 | 27.6 | Yes | *** | 76% |
| U.S. DB | versus | |||||||||||||||
| 59% | ||||||||||||||||
| Panjwani | 2017 | 156 | NCDB | ⋯ | All | No | Yes | No | ≥4 LN | 19c | 30 | 22.4 | 30 | Yes (in N0 | ⋯ | 79% versus |
| et al15 | patients) | 53% (N0 | ||||||||||||||
| group) | ||||||||||||||||
| Tella et al1 | 2018 | 1131ª | NCDB | *** | AJCC I-IV | Yes | Yes | Yes | NDe | 24.5 | 24.5 | ⋯ | ⋯ | No | *** | ⋯ |
| Deschner et al | 2020 | 897 | NCDB | ... | All | No | No | No | ≥5 LN | 16.6 | 5.1 | 16.8 | ⋯ | No | ... | ... |
Note: … Denotes areas where data is not provided. NO = lymph nodes examined and were negative; N1 = lymph nodes examined with at least 1 positive; R1 = positive pathologic margin.
Abbreviations: ACC, adrenocortical carcinoma; AFCE, French Association of Endocrine Surgeons Network; AJCC, American Joint Committee on Cancer; DB, database; DSS, disease-specific survival; ENSAT, European Network for the Study of Adrenal Tumors; LN , lymph node; LAN, lymphadenectomy; LNe, ≥1 lymph nodes examined; M1, metastatic disease; MVR, multivisceral resection; NCDB, National Cancer Database; ND, not defined; noLAN, no lymphadenectomy; OS, overall survival; RFS, recurrence-free survival; SEER, Surveillance, Epidemiology, and End Results Program; UVA, univariate analysis.
aOnly 88% of patients studied had surgery.
PStage IV in this study was per the seventh edition of the American Joint Committee on cancer staging guidelines (T3N1MO and T4NO/NXMO).
“Study identified 827 patients with information on lymph node examination; they focused only on those with >1 LN examined (156 or 19%).
aStudy analyzed 3185 patients; only 1131 underwent surgical resection.
eLymphadenectomy in this study appears to be defined as examination of any number of lymph nodes.
remove lymph nodes, with only R0 resections included.5,12 To our knowledge, an exploration of the therapeutic implications of lym- phadenectomy during R0 resection for ACC has not been performed using the NCDB. Because surgeon intent cannot be extracted from the NCDB, lymphadenectomy was defined as examination of ≥5 lymph nodes, similar to previous studies.12,13,21 However, a survival benefit was not identified using this definition, similar to findings in two other administrative database studies using arbitrary lymph node counts of ≥4 and ≥5.13,14 A notable difference between the current study and prior investigations is that patients undergoing lymphadenectomy defined as ≥5 lymph nodes were compared with patients having 1 to 4 lymph nodes removed, rather than including those with 0 nodes removed. Analysis was performed in this manner given that 84% of specimens had no lymph nodes examined, a figure similar to existing reports.12,13,21 Furthermore, there was no OS difference between Nx and N0 patients (Figure 2A). Ultimately, the true definition of lymphadenectomy for ACC and its therapeutic ef- fect remain ill-defined, with a persistent need for standardization of dissection and subsequent prospective evaluation.
The presence of metastatic lymph nodes has been consistently associated with worse survival.1,12-15,18 As such, both the American Joint Committee on Cancer/Union for International Cancer Control and the European Network for the Study of Adrenal Tumors include node classification (N0 or N1) within the staging schema for ACC.19,20 However, to our knowledge, this is the first study reporting significant worse survival outcomes when stratifying by number of positive lymph nodes. The current analysis demonstrates that among patients undergoing R0 resection, each identified posi- tive lymph node results in a 9% increased risk of death (Figure 4A). This finding was preserved when including the 212 additional pa- tients with MVR, as patients with 1 to 3 positive lymph nodes had a median OS over double that of patients with ≥4 positive lymph nodes (Figure 4B). Although increasing metastatic nodal burden is unlikely to alter adjuvant therapy decisions, it does offer more granular prognostic information similar to the N-class stratification of other aggressive malignancies, such as gastric and pancreatic cancer.31
Limitations of the current study include the retrospective design as well as the absence of relevant data points within the NCDB, such as surgeon volume and intent to perform lymphadenectomy. Fur- thermore, it is not known how many patients had a presumed diag- nosis of ACC preoperatively, as the surgeon may not have been inclined to perform lymphadenectomy if malignancy was not sus- pected. Although nodes in the database are labeled “regional nodes,” the lack of consensus definition as well as specific knowledge of their location prevents definitive distinction between locoregional and distant lymph nodes. Attempts to control for this variable were made by excluding patients with T4 tumors and MVR. There is also no precise definition for “examined,” as we cannot distinguish which patients had nodes that were resected and simply not detected on pathologic review. The NCDB also does not capture important ACC tumor-related factors, such as Ki-67 index and hormone production.
Finally, while overall survival is reported, disease recurrence and disease-specific survival are not.
5 CONCLUSION
Lymph node examination was performed in a minority of patients undergoing curative-intent resection of ACC. Lymph nodes were more likely to be examined in patients treated at academic centers and in those with more advanced tumor characteristics. The in- dependent negative prognostic impact of lymph node metastasis was confirmed on multivariable analysis. Moreover, progressively worse survival with increasing metastatic nodal burden was demonstrated, suggesting the potential for further N class stratification in the sta- ging of ACC. Although an increasing number of examined lymph nodes improved the likelihood of detecting N1 disease, a set threshold for an “adequate” number of excised lymph nodes could not be recommended. Furthermore, when using an arbitrary defini- tion of lymphadenectomy as ≥5 lymph nodes removed, no ther- apeutic benefit was detected, regardless of nodal status. Standardization of a recommended lymph node dissection for curative-intent resection of ACC is imperative for future study of its prognostic and therapeutic implications.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author upon reasonable request.
ORCID
Benjamin W. Deschner D http://orcid.org/0000-0001-5775-8259 Zachary E. Stiles ID http://orcid.org/0000-0002-6393-5079 Miriam Tsao ID http://orcid.org/0000-0003-4139-9962 Evan S. Glazer (D http://orcid.org/0000-0002-5796-0542 Jeremiah L. Deneve ID http://orcid.org/0000-0001-7664-7386 Paxton V. Dickson D http://orcid.org/0000-0002-2345-983X
REFERENCES
1. Bilimoria KY, Shen WT, Elaraj D, et al. Adrenocortical carcinoma in the United States: treatment utilization and prognostic factors. Cancer. 2008;113(11):3130-3136.
2. Schteingart DE, Doherty GM, Gauger PG, et al. Management of patients with adrenal cancer: recommendations of an interna- tional consensus conference. Endocr Relat Cancer. 2005;12(3): 667-680.
3. Icard P, Goudet P, Charpenay C, et al. Adrenocortical carcinomas: surgical trends and results of a 253-patient series from the French Association of Endocrine Surgeons study group. World J Surg. 2001; 25(7):891-897.
4. Dickson PV, Kim L, Yen TWF, et al. Evaluation, staging, and surgical management for adrenocortical carcinoma: an update from the SSO endocrine and head and neck disease site working group. Ann Surg Oncol. 2018;25(12):3460-3468.
5. Gerry JM, Tran TB, Postlewait LM, et al. Lymphadenectomy for adrenocortical carcinoma: is there a therapeutic benefit? Ann Surg Oncol. 2016;23(Suppl 5):708-713.
6. Vassilopoulou-Sellin R, Schultz PN. Adrenocortical carcinoma. Clinical outcome at the end of the 20th century. Cancer. 2001;92(5):1113-1121.
7. Schulick RD, Brennan MF. Long-term survival after complete resec- tion and repeat resection in patients with adrenocortical carcinoma. Ann Surg Oncol. 1999;6(8):719-726.
8. Lykke J, Roikjaer O, Jess P Danish Colorectal Cancer Group. The relation between lymph node status and survival in stage I-III colon cancer: results from a prospective nationwide cohort study. Colorectal Dis. 2013;15(5):559-565.
9. Schwarz RE, Smith DD. Extent of lymph node retrieval and pancreatic cancer survival: information from a large US population database. Ann Surg Oncol. 2006;13(9):1189-1200.
10. Schwarz RE, Smith DD. Clinical impact of lymphadenectomy extent in resectable gastric cancer of advanced stage. Ann Surg Oncol. 2007; 14(2):317-328.
11. Datta J, Roses RE. Surgical management of adrenocortical carcinoma: an evidence-based approach. Surg Oncol Clin N Am. 2016;25(1):153-170.
12. 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):363-369.
13. Saade N, Sadler C, Goldfarb M. Impact of regional lymph node dis- section on disease specific survival in adrenal cortical carcinoma. Horm Metab Res. 2015;47(11):820-825.
14. Nilubol N, Patel D, Kebebew E. Does lymphadenectomy improve survival in patients with adrenocortical carcinoma? A population- based study. World J Surg. 2016;40(3):697-705.
15. Panjwani S, Moore MD, Gray KD, et al. The impact of nodal dissection on staging in adrenocortical carcinoma. Ann Surg Oncol. 2017;24(12): 3617-3623.
16. Tran TB, Liou D, Menon VG, Nissen NN. Surgical management of advanced adrenocortical carcinoma: a 21-year population-based analysis. Am Surg. 2013;79(10):1115-1118.
17. Tella SH, Kommalapati A, Yaturu S, Kebebew E. Predictors of survival in adrenocortical carcinoma: an analysis from the National Cancer Database. J Clin Endocrinol Metab. 2018;103(9):3566-3573.
18. Margonis GA, Kim Y, Prescott JD, et al. Adrenocortical carcinoma: impact of surgical margin status on long-term outcomes. Ann Surg Oncol. 2016;23(1):134-141.
19. Fassnacht M, Johanssen S, Quinkler M, et al. Limited prognostic value of the 2004 International Union Against Cancer staging classification for adrenocortical carcinoma: proposal for a revised TNM Classifi- cation. Cancer. 2009;115(2):243-250.
20. Amin MB, Greene FL, Edge SB, et al. The eighth edition AJCC cancer staging manual: continuing to build a bridge from a population-based
to a more “personalized” approach to cancer staging. CA Cancer J Clin. 2017;67(2):93-99.
21. Alanee S, Dynda D, Holland B. Prevalence and prognostic value of lymph node dissection in treating adrenocortical carcinoma: a na- tional experience. Anticancer Res. 2015;35(10):5575-5579.
22. Miller BS, Doherty GM. Regional lymphadenectomy for adrenocor- tical carcinoma. Ann Surg. 2013;257(4):e13-e14.
23. Reibetanz J, Jurowich C, Allolio B, Fassnacht M, Fassnacht M. Reply to letter: “regional lymphadenectomy for adrenocortical carcinoma”. Ann Surg. 2013;257(4):e15.
24. Marincola Smith P, Kiernan CM, Tran TB, et al. Role of additional organ resection in adrenocortical carcinoma: analysis of 167 Patients from the U.S. Adrenocortical Carcinoma Database. Ann Surg Oncol. 2018;25(8):2308-2315.
25. Merklin RJ. Suprarenal gland lymphatic drainage. Am J Anat. 1966; 119(3):359-374.
26. Polat B, Fassnacht M, Pfreundner L, et al. Radiotherapy in adreno- cortical carcinoma. Cancer. 2009;115(13):2816-2823.
27. Reibetanz J, Rinn B, Kunz AS, et al. Patterns of lymph node recurrence in adrenocortical carcinoma: possible implications for primary surgical treatment. Ann Surg Oncol. 2019;26(2):531-538.
28. Gaujoux S, Brennan MF. Recommendation for standardized surgical management of primary adrenocortical carcinoma. Surgery. 2012; 152(1):123-132.
29. Gaujoux S, Mihai R Joint Working Group of E, Ensat. European So- ciety of Endocrine Surgeons (ESES) and European Network for the Study of Adrenal Tumours (ENSAT) recommendations for the surgical management of adrenocortical carcinoma. Br J Surg. 2017;104(4): 358-376.
30. Bilimoria KY, Palis B, Stewart AK, et al. Impact of tumor location on nodal evaluation for colon cancer. Dis Colon Rectum. 2008;51(2): 154-161.
31. Amin MB, Greene FL. AJCC Staging Manual. 8th ed. Springer Inter- national Publishing; 2017.
How to cite this article: Deschner BW, Stiles ZE, DeLozier OM, et al. Critical analysis of lymph node examination in patients undergoing curative-intent resection for adrenocortical carcinoma. J Surg Oncol. 2020;1-11. https://doi.org/10.1002/jso.26138