Surgery xxx (2019) 1-8
ELSEVIER
Contents lists available at ScienceDirect
Surgery
journal homepage: www.elsevier.com/locate/surg
SURGERY NOWWEMEER 2018
$
- &
Comparison between functional and non-functional adrenocortical carcinoma
Alaa Sada, MDa,c, Malke Asaad, MDª, Katherine A. Bews, BAb,c, Geoffrey B. Thompson, MDª, William F. Young Jr., MDª, Irina Bancos, MDª, David R. Farley, MDª, Benzon M. Dy, MDª, Melanie L. Lyden, MDª, Elizabeth B. Habermann, PhD, MPHb,c, Travis J. Mckenzie, MDª,*
a Department of Surgery, Mayo Clinic, Rochester, MN
b Department of Health Services Research, Mayo Clinic, Rochester, MN
” Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MN d Division of Endocrinology, Mayo Clinic, Rochester, MN
ARTICLE INFO
Article history: Accepted 27 April 2019 Available online xxx
ABSTRACT
Background: While roughly half of adrenocortical carcinomas are functional, whether functional status impacts outcomes remains controversial. We compared presentation and survival for functional and nonfunctional neoplasms.
Methods: Adult patients presented with adrenocortical carcinomas at the Mayo Clinic were included. Tumor characteristics and outcomes were analyzed.
Results: The 266 identified patients presented with stage I (6%), II (33%), III (26%), and IV disease (32%); stage was unknown in 3%. Fifty-three percent of tumors were functional; patients with functional adrenocortical carcinomas were younger, more likely to be female, and more likely to present with metastatic disease. Surgical resection was undertaken in 84% of patients with 69% having R0 resection. While 30-day morbidity was similar between functional and nonfunctional adrenocortical carcinomas, median overall survival was better for nonfunctional adrenocortical carcinomas (median 66 vs 22 months, P = . 01). Functional adrenocortical carcinomas was independently associated with shorter survival after adjusting for age, sex, grade, stage, and resection attempt: hazard ratio = 1.5 (95% confi- dence interval, 1.04-2.14, P = . 03).
Conclusion: In our cohort, long term survival was worse for all patients with functional tumors. However, when analyzing patients with R0 resection, there was no survival difference between functional and nonfunctional adrenocortical carcinomas, signaling need for better understanding of adrenocortical carcinomas behavior to individualize and optimize treatment strategies.
@ 2019 Published by Elsevier Inc.
Introduction
Adrenocortical carcinoma (ACC) is a rare neoplasm with an estimated incidence of 0.5 to 2 cases per million person years.1 While the incidence has been stable since 1973, very limited progress has been achieved to improve survival despite new ad- vances in both medical and surgical care.1,2 Approximately 50% of ACC may present with clinical manifestations secondary to hor- monal excess. This can include autonomous secretion of
glucocorticoids, sex hormones, aldosterone, or mixed hormone secretion. Factors that determine prognosis include stage at pre- sentation, completeness of surgical resection, and other histologic factors. Studies have suggested that functional and nonfunctional ACC may have differing molecular profiles, but whether functional status has an impact on prognosis remains unclear.3 Herein, we sought to determine if functional status of ACC was independently associated with stage of disease at presentation, short term surgical morbidity and mortality, and long term prognosis.
Methods
The Mayo Clinic Cancer Registry was utilized to identify all pa- tients with ACC who presented to Mayo Clinic between January 1, 1997 and December 30, 2017. The Mayo Clinic Cancer Registry
Presented at the 40th Annual Meeting of the American Association of Endocrine Surgeons in Los Angeles, CA from April 7 to 9, 2019.
* Reprint requests: T.J. Mckenzie, MD, Department of Surgery, Mayo Clinic, 200 1st St SW, Rochester, MN 55905.
E-mail address: Mckenzie.Travis@mayo.edu (T.J. Mckenzie).
| All patients N = 266 | Functional ACC n = 140 | Nonfunctional ACC n = 126 | P value | |
|---|---|---|---|---|
| Age at diagnosis | .02 | |||
| Mean (SD) | 51.6 ± 15.5 | 49.2 ± 16.1 | 54.2 ± 14.4 | |
| Range | 18-84 | 18-83 | 26-84 | |
| Sex | .002 | |||
| Male | 105 (39.5%) | 43 (30.7) | 62 (49.2%) | |
| Female | 161 (60.5%) | 97 (69.3) | 64 (50.8%) |
SD, standard deviation.
* Wilcoxon rank sum test.
t Pearson x2 test.
maintains annual follow-up information concerning recurrence and patient vital status. This study was approved by Mayo Clinic’s institutional review board.
Adult patients who presented with ACC over the study period were identified. We included only patients with documented functional status as reviewed by our internal endocrinologists. The Mayo Clinic Cancer Registry was utilized to abstract patient de- mographic characteristics, neoplasm characteristics, and survival. Medical records were reviewed to determine mode of presentation, stage based on the 8th edition of the TNM staging system, completeness of resection, and adjuvant chemotherapy and/or ra- diation status.
For patients who had surgery at our institution, we reviewed the 30-day surgical outcomes, including complications based on Clavien-Dindo classification, duration of procedure, estimated blood loss, postoperative admission to the intensive care unit, intra- or postoperative blood transfusion, and length of hospital stay. Long-term outcomes analyzed included overall survival.
Variables were analyzed using x2, Fisher exact, and Wilcoxon rank sum tests. Survival was assessed using Kaplan-Meier analysis and multivariable Cox proportional hazards regression models, using the duration between date of death or last patient contact and date of diagnosis. Statistical analysis was performed using SAS version 9.4 statistical software (SAS Institute, Inc, Cary, NC).
Results
Baseline characteristics
A total of 266 adult patients were identified meeting inclusion criteria. The average age was 51.6 ± 15.5 years with 60.5% of pa- tients being female. When comparing patient demographics be- tween functional versus nonfunctional ACC, patients with functional ACC were younger (49.2 ± 16.1 vs 54.2 ± 14.4, P = . 02) and more likely to be female (69.3% vs 50.8%, P = . 002). Baseline characteristics are summarized in Table I.
Neoplasm characteristics
Of the entire cohort of 266 total patients with ACC, 140 patients (52.6%) were found to have functional ACC. The most common type of functional ACC was isolated cortisol secreting in 66 patients (47.1%), followed by mixed hormone secreting ACC in 35 patients (25.0%), isolated sex hormone secreting in 27 patients (19.3%), and isolated aldosterone secreting in 11 patients (7.9%). Type of func- tional tumor was unclassified in a single patient. The mixed hor- mone secreting ACC produced mainly a combination of cortisol and androgen in all except 4 cases (2 of them secreted cortisol and estrogen, and 2 cases secreted cortisol and aldosterone) (Fig 1).
Mixed hormone 25%
Aldosterone 8%
Cortisol 47%
Isolated sex hormone 20%
Of the 140 patients with functional ACC, 68 patients (48.6%) presented with symptoms of hormone excess, while 34 (24.3%) presented with local compressive symptoms. Thirty cases of func- tional ACC (21.4%) were detected incidentally on imaging done for other reasons. The mode of presentation was undocumented for 8 patients (5.7%) due to limited records from the time of presenta- tion. Of the 126 patients with nonfunctional ACC, presentation due to symptoms of mass effect occurred in 75 cases (59.5%), while 39 (30.9%) were identified incidentally on imaging done for other reasons. The presentation was undocumented for 12 cases of nonfunctional ACC (9.6%).
Left-sided neoplasms were more common overall (56.4%), with no difference between functional and nonfunctional neoplasms by side (P = . 45). Overall average neoplasm size was 12.0 ± 5.6 cm; 83.4% of patients presented with neoplasm size ≥4 cm. There was no difference in neoplasm size between functional and nonfunc- tional ACC (P = . 24). Of the entire cohort of 266 patients, 5.6% presented with stage I disease, 33.1% presented with stage II dis- ease, 26.3% presented with stage III disease, and 32.0% presented with stage IV disease; stage was unknown in 3.0%. There was a significant difference in neoplasm stage between functional and nonfunctional ACC, as 41.4% of functional neoplasms presented with stage IV compared to 21.4% of nonfunctional neoplasms (P = .001) (Table II).
A. Sada et al. / Surgery xxx (2019) 1-8
| All patients N = 266 | Functional ACC n = 140 | Nonfunctional ACC n = 126 | P value | |
|---|---|---|---|---|
| Neoplasm size (cm) | ||||
| n | 227 | 117 | 110 | .24" |
| Mean (SD) | 12.0 ± 5.6 | 11.7 ± 5.4 | 12.4 ± 5.7 | |
| Range | 1.6-32.0 | 3.0-32.0 | 1.6-28.5 | |
| Laterality | .45 | |||
| Right | 116 (43.6%) | 58 (41.4%) | 58 (46.0%) | |
| Left | 150 (56.4%) | 82 (58.6%) | 68 (54.0%) | |
| Grade | .46 | |||
| Grade I | 12 (4.5%) | 8 (5.7%) | 4 (3.2%) | |
| Grade II | 12 (4.5%) | 6 (4.3%) | 6 (4.8%) | |
| Grade III | 32 (12.0%) | 13 (9.3%) | 19 (15.1%) | |
| Grade IV | 20 (7.5%) | 9 (6.4%) | 11 (8.7%) | |
| Unknown | 190 (71.4%) | 104 (74.3%) | 86 (68.3%) | |
| Lymphovascular invasion | .9+ | |||
| Absent | 102 (45.5%) | 51 (44.7%) | 51 (46.4%) | |
| Present | 77 (34.4%) | 39 (34.2%) | 38 (34.5%) | |
| Unknown | 45 (20.1%) | 24 (21.1%) | 21 (19.1%) | |
| Stage | .002 | |||
| Stage I | 15 (5.6%) | 9 (6.4%) | 6 (4.8%) | |
| Stage II | 88 (33.1%) | 45 (32.1%) | 43 (34.1%) | |
| Stage III | 70 (26.3%) | 25 (17.9%) | 45 (35.7%) | |
| Stage IV | 85 (32.0%) | 58 (41.4%) | 27 (21.4%) | |
| Unknown | 8 (3.0%) | 3 (2.1%) | 5 (4.0%) |
SD, standard deviation.
* Wilcoxon rank sum test.
+ Pearson x2 test.
There was no difference in neoplasm grade or intra/peritumor lymphovascular invasion between functional and nonfunctional ACC as shown in Table II.
Surgical resection
Attempt at surgical resection was performed in 224 patients (84%). This included 114 patients (81.4%) in the functional group and 110 patients (87.3%) in the nonfunctional group, P = . 23. Incomplete primary palliative resection due to the presence of distant metastatic disease was performed in 25 (17.8%) cases of functional ACC and 16 (12.7%) cases of nonfunctional ACC (P = . 21).
For all patients who underwent surgical resection with the intent of complete resection (181 patients), complete resection with negative margins (R0) was achieved in 154 patients (85%), including 76 patients with functional ACC and 78 patients with nonfunctional ACC, (P = . 21). Microscopic positive margins were present in 14 patients (8 had functional ACC, and 6 had nonfunc- tional, P = . 6), while the margin status was unknown for 13 pa- tients. Incomplete resection due to macroscopic margins despite intent for complete resection occurred in 2 cases of non-functional ACC (Fig 2). The proportions of patients who underwent resection for recurrent local, regional, or distant disease in the setting of functional and nonfunctional primary ACC was 15.8% and 25.5%, respectively (P = . 09).
A total of 123 patients had their first surgical resection of ACC at our institution. Among patients who had surgery elsewhere, 31 patients presented to our institution prior to recurrence and 61 patients presented when they developed local, regional, or distant recurrent disease. Among patients who underwent surgery at our institution, open approach was performed in 80.5% of cases (78.5% of resected functional and 82.8% of resected nonfunctional ACC) and minimally invasive approach was performed in 19.5%. The median size (interquartile range) of neoplasms resected though the minimally invasive approach was smaller than neoplasms resected with an open approach; 6.2 (5.0-7.7) cm compared to 13.5 (9.9-18.0) cm, respectively (P < . 0001).
On the other hand, when looking at surgical approach for all patients who had surgery regardless of institution, 61.6% were resected via open surgery and 20.1% via minimally invasive, while approach was not documented in 18.3%. There was no differences in the approach between resected functional and nonfunctional ACC (P> .05.).
30-day Surgical Outcomes
When comparing 30-day morbidity between functional and nonfunctional ACC patients who underwent surgery at our insti- tution (n = 123), there was no significant difference in incidence or severity. Overall complications rate in functional and non- functional ACC were 36.9% and 41.4%, respectively (P = . 61). Clavien-Dindo class I rate was 31.3%, class II 47.9%, class III 16.7%, and class IV 4.2% and was similar between groups. The most common complications among each Clavien-Dindo class were nausea/vomiting in class I, blood transfusion (n = 16), pneumonia (n = 2), and atrial fibrillation (n = 2) in class II complications, and abscess requiring drainage in class III (n = 4). There were 2 class IV complications, myocardial infarction and septic shock.
There was no difference in estimated blood loss, surgery length, intra- or postoperative transfusion, and admission to the intensive care unit between functional and nonfunctional ACC as shown in Tables III and IV. However, in patients who underwent adrenalec- tomy without multi-visceral resection, length of hospital stay was longer for nonfunctional lesions compared to functional (6.0 days vs 3.7 days, P =. 03). When we compared length of stay for patients who had surgery before 2007 and those who had surgery after 2007, there was no difference in length of stay for patients under- going adrenalectomy without multi-visceral resection.
Adjuvant therapy
Adjuvant local radiation therapy was administered to 44 pa- tients; 12.9% of the functional group compared to 20.6% of the nonfunctional group (P = . 09). Adjuvant chemotherapy was
TNM I=9
TNM II=40
R0=76
TNM III=21
R1=8
TNM IV=6
Functional N=140
Surgery=114
R2=0
No surgery=26
Incomplete=25
Unknown=5
All patients, N=266
Unknown=8
Incompete=16
Non-functional N=126
Surgery=110
R2=2
TNM I=4
No surgery=16
R1=6
TNM II=39
R0=78
TNM III=34
TNM IV=0
TNM n/a =1
administered to 155 patients, 72.6% of the functional group and 61.9% of the nonfunctional one (P = . 09).
Among patients who received chemotherapy, 86.7% in the functional group either had Mitotane only or Mitotane with other systematic chemotherapy and 13.3% received multi-agents chemotherapy without Mitotane. On the other hand, 84.6% of nonfunctional group received Mitotane only or Mitotane with other systematic chemotherapy, and 15.4%% received multi-agent chemotherapy without Mitotane. There were no differences in chemotherapy regimens between functional and nonfunctional groups (P > .05).
Long term survival
The median follow-up duration in our study was 21.6 (9.7-56.2) months (range, 0.6-242.9). The median overall survival for ACC was 35 months. Overall survival for all patients was longer for nonfunctional ACC (median 66 months vs 22 months, P = . 01) (Fig 3). On multivariable analysis including all patients (N = 266), functional ACC was associated with worse survival after adjusting for age, sex, grade, stage, and resection status, hazard ratio = 1.5 (95% confidence interval, 1.04-2.13, P = . 03) (Table V).
Subgroup analysis including only patients with complete (R0) resection showed that survival was similar between functional and nonfunctional ACC (45 months vs 77 months) (P =. 43) as shown in Fig 4. When including only patients with R0 resection in the multi-
variable analysis, we did not find a statistical difference in survival between functional and nonfunctional ACC (Table VI).
Discussion
We sought to determine if the functional status of ACC was independently associated with the stage of disease at presentation, short-term surgical morbidity and mortality, and long-term sur- vival. Our study highlights that functional and nonfunctional neo- plasms differ in many respects. Patients with functional ACC tended to be younger and were more likely to be female compared to those with nonfunctional ACC. Furthermore, functional ACC was more likely to present with metastatic disease. When comparing all pa- tients who were evaluated at our institution with a diagnosis of ACC, those with functional neoplasms had shorter survival compared to those with nonfunctional ACC. Subgroup analysis comparing only patients with complete surgical resection (R0) demonstrated no difference in survival between functional and nonfunctional ACC.
Our study detected a difference in patient age between func- tional and nonfunctional neoplasm, with functional ACC tending to affect younger patients. However, the reason and implication of this age difference is unclear. To our knowledge there is no existing study that addresses age difference between functional and nonfunctional ACC. Moreover, the national oncological databases, including the National Cancer Institute’s Surveillance,
A. Sada et al. / Surgery xxx (2019) 1-8
| Total (n = 51) | Functional (n = 26) | Nonfunctional (n = 25) | P value | |
|---|---|---|---|---|
| LOS, d | .03 | |||
| Mean (SD) | 4.8 (4.4) | 3.7 (2.2) | 6.0 (5.7) | |
| Range | 1.0-31.0 | 1.0-9.0 | 1.0-31.0 | |
| EBL, mL | .11* | |||
| n | 38 | 21 | 17 | |
| Mean (SD) | 578.2 (771.5) | 518.0 (824.3) | 652.6 (718.6) | |
| Range | 5.0-3,300.0 | 5.0-3,300.0 | 20.0-3,000.0 | |
| Surgery length, min | 1.00* | |||
| n | 35 | 19 | 16 | |
| Mean (SD) | 176.5 (67.8) | 179.1 (70.8) | 173.4 (66.3) | |
| Range | 72.0-316.0 | 72.0-316.0 | 75.0-289.0 | |
| Transfusion | .24+ | |||
| No | 20 (39.2%) | 13 (50.0%) | 7 (28.0%) | |
| Yes | 4 (7.8%) | 1 (3.8%) | 3 (12.0%) | |
| Missing | 27 (52.9%) | 12 (46.2%) | 15 (60.0%) | |
| ICU | .85+ | |||
| No | 44 (86.3%) | 23 (88.5%) | 21 (84.0%) | |
| Yes | 3 (5.9%) | 1 (3.8%) | 2 (8.0%) | |
| Missing | 4 (7.8%) | 2 (7.7%) | 2 (8.0%) | |
| Complication | .84+ | |||
| No | 34 (66.7%) | 17 (65.4%) | 17 (68.0%) | |
| Yes | 17 (33.3%) | 9 (34.6%) | 8 (32.0%) | |
| Clavien-Dindo | 1.00* | |||
| 0$ | 34 (66.7%) | 17 (65.4%) | 17 (68.0%) | |
| I | 8 (15.7%) | 4 (15.4%) | 4 (16.0%) | |
| II | 7 (13.7%) | 4 (15.4%) | 3 (12.0%) | |
| III | 2 (3.9%) | 1 (3.8%) | 1 (4.0%) |
EBL, estimated blood loss; ICU, postoperative admission to the intensive care unit; LOS, length of stay; SD, standard deviation.
* Wilcoxon rank sum test.
Pearson x2 test.
# Fisher exact test.
§ Clavien-Dindo class 0 = no complications.
| Total (n = 72) | Functional (n = 39) | Nonfunctional (n = 33) | P value | |
|---|---|---|---|---|
| LOS, d | .47 | |||
| Mean (SD) | 8.9 (10.1) | 9.1 (10.0) | 8.7 (10.2) | |
| Range | 2.0-62.0 | 3.0-60.0 | 2.0-62.0 | |
| EBL, mL | .60 | |||
| n | 56 | 34 | 22 | |
| Mean (SD) | 2,118.7 (3,390.03) | 2,464.0 (4,222.70) | 1,585.0 (1,259.80) | |
| Range | 20.0-25,000.0 | 250.0-25,000.0 | 20.0-4,000.0 | |
| Surgery length, min | .41* | |||
| n | 50 | 26 | 24 | |
| Mean (SD) | 274.4 (112.3) | 289.3 (120.69) | 258.3 (102.63) | |
| Range | 110.0-654.0 | 134.0-654.0 | 110.0-525.0 | |
| Transfusion | .65 | |||
| No | 23 (31.9%) | 11 (28.2%) | 12 (36.4%) | |
| Yes | 12 (16.7%) | 6 (15.4%) | 6 (18.2%) | |
| Missing | 37 (51.4%) | 22 (56.4%) | 15 (45.5%) | |
| ICU | .791 | |||
| No | 47 (65.3%) | 26 (66.7%) | 21 (63.6%) | |
| Yes | 25 (34.7%) | 13 (33.3%) | 12 (36.4%) | |
| Complication | .391 | |||
| No | 41 (56.9%) | 24 (61.5%) | 17 (51.5%) | |
| Yes | 31 (43.1%) | 15 (38.5%) | 16 (48.5%) | |
| Clavien-Dindo | .81+ | |||
| OS | 41 (56.9%) | 24 (61.5%) | 17 (51.5%) | |
| I | 7 (9.7%) | 3 (7.7%) | 4 (12.1%) | |
| II | 16 (22.2%) | 9 (23.1%) | 7 (21.2%) | |
| III | 6 (8.3%) | 2 (5.1%) | 4 (12.1%) | |
| IV | 2 (2.8%) | 1 (2.6%) | 1 (3.0%) |
EBL, estimated blood loss; ICU, postoperative admission to the intensive care unit; LOS, length of stay; SD, standard deviation.
* Wilcoxon rank sum test.
+ Pearson x2 test.
# Fisher exact test.
§ Clavien-Dindo class 0 = no complications.
Survival Functional vs Non Functional Tumor
100
Functional
Events/Total
Median (95% CI)
HR (95% CI)
functional
87/140
22.4 (18.6-32.9)
1.50 (1.10-2.05)
non-functional
73/126
65.5 (38.8-90.1)
Reference
90
Logrank P-value: 0.0108
+ Censor
80
70
60
Survival %
50
40
30
20
10
0
0
12
24
36
48
60
Survival Time (Months)
| Variable (ref) | HR (95% CI) | P value |
|---|---|---|
| All patients (N = 266) TNM stage (8th edition) | ||
| I | reference | |
| II | 0.85 (0.39-1.85) | .68 |
| III | 1.12 (0.51-2.45) | .78 |
| IV | 2.93 (1.37-6.28) | .006 |
| Unknown | 2.64 (0.93-7.51) | .07 |
| No surgery | 2.30 (1.56-3.39) | <. 0001 |
| Functional ACC | 1.49 (1.04-2.13) | .028 |
| Age ≥60 y | 2.16 (1.53-3.05) | <. 0001 |
| Female sex | 1.30 (0.91-1.85) | .15 |
| Grades | ||
| I | reference | |
| II | 1.12 (0.37-3.40) | .84 |
| III | 2.21 (0.86-5.70) | .10 |
| IV | 1.05 (0.38-2.89) | .93 |
| Unknown | 1.51 (0.65-3.51) | .34 |
CI, confidence interval; HR, hazard ratio.
Epidemiology, and End Results set of cancer registries and the Commission on Cancer’s National Cancer Database, do not include data regarding the functional status of ACC. This impedes their utility in the study of clinical differences between functional and nonfunctional ACC.
Our results demonstrate that, in addition to being younger, patients with functional ACC were more likely to be female. Existing studies show that ACC in general is more common in fe- males compared to males.1,4 However, when analyzing ACC func- tional status based on sex, interestingly we found that only functional ACC are more common in females, while nonfunctional ACC have an almost equal distribution of males and females. A study from Cleveland Clinic demonstrated that patients with functional ACC were more likely to be female while nonfunctional
ACC were more likely to be male.5 These clinical findings may be a result of differences in molecular pathogenesis between functional and nonfunctional ACC. In vitro studies suggest that estrogen and its receptors play a role in the growth and differentiation of ACC cells.6,7
There was no difference in neoplasm size between functional and nonfunctional ACC in our cohort. This finding is interesting as one could hypothesize that functional ACC presenting with excess hormone production would theoretically lead to earlier develop- ment of symptoms and, therefore, possible detection at a smaller size. This was not the case. Both functional and nonfunctional ACC tended to present at large size and at advanced stages. Despite similar size at presentation, there was a difference in TNM stage between functional and nonfunctional ACC. Among our cohort of patients, functional ACC was more likely to present with stage IV disease, while nonfunctional ACC was more commonly stage III at presentation. The importance of stage differences between func- tional and nonfunctional ACC is underlined through the difference in survival when all patients are evaluated, including those with incomplete resection. This would be an expected finding in patients with stage IV disease, where disease not resected may have detri- mental effects if autonomously producing glucocorticoids. Inter- estingly, there was no difference in survival when patients with complete surgical resection (R0) were analyzed alone utilizing multivariable analysis. It remains unclear if this survival difference is due to the detrimental effects of hormone overproduction or if there was a higher recurrence rate in the functional group, even after R0 resection.
That raises important and intriguing questions: Do functional and nonfunctional ACC simply have similar long-term survival? Are surgeons more likely to offer surgery for metastatic ACC for palli- ative reasons if the lesion is functional? The burden of hormone excess in functional ACC can lead to more aggressive surgical resection even in the presence of distant metastases and in fact a few studies have advocated offering palliative resection to
Survival Functional vs Non Functional Tumor
100
Functional
Events/Total Median (95% CI) HR (95% CI)
functional
33/76
45.0 (28.8-NE)
1.20 (0.76-1.91)
non-functional
40/78
77.3 (72.9-107.5)
Reference
90
Logrank P-value: 0.4336
+ Censor
80
70
60
Survival %
50
40
30
20
10
0
0
12
24
36
48
60
Survival Time (Months)
| Variable (ref) | HR (95% CI) | P value |
|---|---|---|
| TNM stage (8th edition) | ||
| I | reference | |
| II | 0.77 (0.32-1.87) | .56 |
| III | 1.19 (0.48-2.97) | .71 |
| IV | 5.31 (1.45-19.44) | .01 |
| Unknown | 2.21 (0.24-20.13) | .48 |
| Functional ACC | 1.28 (0.75-2.19) | .37 |
| Age ≥60 y | 2.15 (1.25-3.70) | .006 |
| Female sex | 1.52 (0.86-2.68) | .15 |
| Grades | ||
| I | reference | |
| II | 1.34 (0.36-5.08) | .66 |
| III | 1.64 (0.48-5.61) | .43 |
| IV | 0.58 (0.12-2.69) | .48 |
| Unknown | 1.16 (0.40-3.35) | .78 |
CI, confidence interval; HR, hazard ratio
functional ACC.8,9 Unfortunately, our study cannot address if palli- ative resection of functional ACC for patients with metastatic dis- ease improves survival. However, a combined series from our institution and MD Anderson Cancer Center evaluated ACC patients with distant metastatic disease and found that complete surgical resection of ACC along with synchronous metastasis improves survival.10,11
Our study has some limitations given the nature of a retro- spective study at affiliated tertiary referral centers. The majority of our ACC cases reflect more challenging and advanced stages, which may explain the limited number of patients presenting with early- stage disease. However, ACC stage distribution is similar to what is reported in population-based ACC studies.2,12 While we were able to review pathology reports for most patients, some were missing for patients who underwent surgical resection at outside hospitals, which explains some of the missed data especially for tumor grade.
We used overall survival instead of cancer specific survival as we were not able to confirm the cause of death in all patients. While this study is one of the largest institutional cohorts, our patient numbers remain relatively small and hinder sub-group analysis. Thus, we were not able to include the surgical approach in the multivariable analysis, even though the minimally invasive approach has been associated with higher recurrence rate and worse survival.
In conclusion, approximately half of ACC present with either isolated or mixed hormone secretion. In our cohort including all patients with ACC, long term survival was worse for functional tumors. However, when analyzing only patients with complete surgical resection, there was no survival difference between func- tional and nonfunctional ACC, signaling the need for better un- derstanding of ACC behavior to individualize and optimize treatment strategies.
Funding/Support
This publication was made possible by CTSA Grant Number UL1 TR002377 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes of Health (NIH). Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NIH.
Conflict of interest/Disclosure
The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article.
References
1. Sharma E, Dahal S, Sharma P, et al. The characteristics and trends in adreno- cortical carcinoma: a United States population based study. J Clin Med Res. 2018;10:636-640.
A. Sada et al. / Surgery xxx (2019) 1-8
2. Kebebew E, Reiff E, Duh Q-Y, Clark OH, McMillan A. Extent of disease at pre- sentation and outcome for adrenocortical carcinoma: have we made progress? W J Surg. 2006;30:872-878.
3. Ilvesmäki V, Kahri AI, Miettinen PJ, Voutilainen R. Insulin-like growth factors (IGFs) and their receptors in adrenal tumors: high IGF-II expression in functional adrenocortical carcinomas. J Clin Endocrinol Metab. 1993;77: 852-858.
4. Dackiw AP, Lee JE, Gagel RF, Evans DB. Adrenal cortical carcinoma. W J Surg. 2001;25:914-926.
5. Bodie B, Novick AC, Pontes JE, et al. The Cleveland Clinic experience with ad- renal cortical carcinoma. J Urol. 1989;141:257-260.
6. Montanaro D, Maggiolini M, Recchia A, et al. Antiestrogens upregulate estrogen receptor ß expression and inhibit adrenocortical H295R cell proliferation. J Mol Endocrinol. 2005;35:245-256.
7. Sirianni R, Zolea F, Chimento A, et al. Targeting estrogen receptor-o. reduces adrenocortical cancer (ACC) cell growth in vitro and in vivo: potential
therapeutic role of selective estrogen receptor modulators (SERMs) for ACC treatment. J Clin Endocrinol Metab. 2012;97:E2238-E2250.
8. Bilimoria KY, Shen WT, Elaraj D, et al. Adrenocortical carcinoma in the United States: treatment utilization and prognostic factors. Cancer. 2008;113: 3130-3136.
9. Else T, Kim AC, Sabolch A, et al. Adrenocortical carcinoma. Endoc Rev. 2014;35: 282-326.
10. Dy BM, Strajina V, Cayo AK, et al. Surgical resection of synchronously meta- static adrenocortical cancer. Ann Surg Oncol. 2015;22:146-151.
11. Dickson PV, Kim L, Yen TW, et al. Evaluation, staging, and surgical manage- ment for adrenocortical carcinoma: an update from the SSO Endocrine and Head and Neck Disease Site Working Group. Ann Surg Oncol. 2018;25: 3460-3468.
12. Ayala-Ramirez M, Jasim S, Feng L, et al. Adrenocortical carcinoma: clinical outcomes and prognosis of 330 patients at a tertiary care center. Eur J Endo- crinol. 2013;169:891-899.
Discussion
Dr Richard Hodin (Boston, MA): I’m curious how you defined “nonfunctional” and how carefully those patients were actually tested for some subtle functionality of the tumor.
Dr Alaa Sada: That’s a good question. We defined functionality based on full biochemical evaluation. We started with around 300 patients, and we excluded those for whom a complete biochemical workup was not obtained.
Having said that, not all of these testings were performed at the Mayo Clinic. We included patients if they had functional testing performed outside Mayo if the results were scanned and docu- mented in our electronic medical records.
Dr Richard Hodin (Boston, MA): Specifically, when looking for Cushing’s Syndrome, for example, was it with blood testing, 24- hour urinary levels, ACTH, etc .? How carefully could you tease that out?
Dr Alaa Sada: That’s an excellent question. Basically, all these patients had serum cortisol, 24-hour urine cortisol, ACTH, and 1 milligram dexamethasone suppression testing for Cushing’s Syndrome.
Dr Matthew Nehs (Boston, MA): Your results are consistent with what we see at Brigham & Women’s Hospital in Boston. But it seems to be the opposite of what one might expect, because a nonfunctional tumor lies indolent in someone’s retroperitoneum presumably for years and years, whereas a functional tumor creates an endocrinopathy that’s detectable.
What do you surmise is the reason for this? Is it the functionality of the tumor such as cortisol driving Cushing’s Syndrome or
hyperglycemia that could account for this difference? Or are these tumors more aggressive independent of their hormonal function?
Dr Alaa Sada: That’s a good point. I agree with you. The litera- ture shows that functional tumors tend to have worse prognosis. The difference in prognosis might be related to tumor stage as we have seen that in functional lesions, the rate of stage 4 disease is two times higher compared to nonfunctional tumors or it may be related to the hormones themselves, like cortisol and its effect on physiology.
Another explanation for the difference in prognosis is that functional and non-functional ACC may have different molecular and biological profiles that can affect their recurrence and clinical behavior.
Dr Marybeth Hughes (Norfolk, VA): What’s fascinating and most surprising about your results is that you didn’t find a differ- ence in complications with the functional tumors. So the literature is pretty clear that hypercortisolism puts these patients at a lot higher risk of postoperative complications. How do you explain that you didn’t find that difference?
Dr Alaa Sada: Thank you for the interesting question. You are correct. It is well known that hypercortisolism increases post-sur- gical complications. I think as surgeons and health care providers typically expect higher complication rates for functional tumors, they tend to monitor these patients more closely. In addition, these patients tend to require multidisciplinary teams care specifically from endocrinology stand point so I think close monitoring from multiple levels may have helped improved these patients outcomes.