Aldosterone-secreting adrenocortical carcinomas are associated with unique operative risks and outcomes
Michael L. Kendrick, MD, Kathleen Curlee, RN, Ricardo Lloyd, MD, David R. Farley, MD, Clive S. Grant, MD, Geoffrey B. Thompson, MD, Charles Rowland, William F. Young, Jr, MD, and Jon A. van Heerden, MD, Rochester, Minn
Background. Adrenocortical carcinoma (ACC) that produces aldosterone is an extremely rare, uncharac- terized endocrine malignancy. Our aim was to characterize this neoplasm in terms of its clinical behav- ior and patient outcomes.
Methods. A retrospective review was made of all patients who had operative management of aldosterone- secreting ACC from 1957 to 2000 at the Mayo Clinic. Comparisons were made to patients with non-aldosterone-secreting ACC treated during the same period.
Results. Of 141 patients with ACC, we identified 15 patients with aldosterone-secreting ACC. Isolated aldosterone hypersecretion was present in 10 patients, and mixed hormonal secretion was detected in 5. Mean tumor size and weight were 10.8 cm and 453 g, respectively. Surgical management included curative resection in 10 patients (67%). Perioperative mortality was 20%. Disease recurred in 7 patients (70%) with a median interval of 17 months. Five-year survival was 52%. Patients with aldosterone-secreting ACC had an increased risk of perioperative mortality (20% vs 5%), yet they had an overall survival of 63 months compared to 19 months for patients with non-aldosterone-secreting ACC.
Conclusions. Aldosterone hypersecretion occurs in 11% of all ACCs and is associated with unique oper- ative risk and outcome. Although patients harboring aldosterone-secreting ACC appear to have an increased risk of perioperative death, survivors may have an improved overall survival rate compared with patients with non-aldosterone-secreting ACC. (Surgery 2002;132:1008-12.)
From the Departments of Surgery, Endocrinology, Pathology, and Statistics, Mayo Clinic, Rochester, Minn
ADRENOCORTICAL CARCINOMA (ACC) is a rare endocrine malignancy exhibiting varied tumor characteristics, clinical presentation, and outcome. Characterizing the optimal management and clearly defining the prognostic factors and patient outcome have been arduous because of the infre- quent occurrence of this neoplasm. ACC is classi- fied as functional or nonfunctional, depending on the ability of the tumor to produce adrenocortical hormones. Approximately one half of ACCs are functional, with the most common secretory pat- tern being corticosteroid, followed by mixed hor- monal, sex hormone, and aldosterone.1-7
Presented at the 23rd Annual Meeting of the American Association of Endocrine Surgeons, Banff, Alberta, Canada, April 7-9, 2002.
Reprint requests: Jon A. van Heerden, MD, Department of Surgery, Mayo Clinic, 200 First Street, SW, Rochester, MN 55905.
@ 2002, Mosby, Inc. All rights reserved. 0039-6060/2002/$35.00 + 0 11/6/128476 doi:10.1067/msy.2002.128476
Whether the functional ability of the tumor affects the outcome of patients with ACC is not clearly defined. Although it has been reported that patients with functional tumors have a poorer prognosis than those with nonfunctional tumors,1,7 others have shown no difference in sur- vival.2 These reports are limited in that they com- pare all functional tumors as a group to those that are nonfunctional. It is not known whether the specific hormone secreted has unique prognostic significance in patients with ACC, or if these potential differences could account for the incon- sistent findings when considering functional tumors collectively.
Aldosterone-secreting ACC is extremely uncom- mon, comprising 0% to 7% of all functioning ACCs.1-7 The current literature regarding this neo- plasm is based exclusively on case reports.8-18 To our knowledge, this represents the first large col- lective series focused on the characterization of this rare neoplasm. Our aim was to determine whether aldosterone-secreting ACC is associated with unique risks and outcomes.
| Aldosterone (n = 15) | Non-aldosterone-secreting (n = 126) | P value | |
|---|---|---|---|
| Mean tumor size (cm) | 10.8 | 12.4 | .16 |
| Tumor grade III-IV | 47% | 50% | .9 |
| Stage III-IV | 40% | 59% | .17 |
| Resection with curative intent | 67% | 60% | .6 |
| Perioperative mortality | 20% | 5% | .02 |
| Adjuvant therapy | 42% | 46% | .68 |
| Kaplan-Meier 5-year survival | 52% | 28% | .08 |
METHODS
A retrospective review was performed for all patients diagnosed with ACC who underwent oper- ative management at the Mayo Clinic from 1957 to 2000. Data evaluated included presurgical and postsurgical history, results of serum and urine chemical analysis and endocrine studies, tumor imaging, pathologic findings, patient manage- ment, and outcome. Tumor functional status was determined by serum or urinary endocrine studies and when the clinical history and physical exami- nation were consistent with hormonal hypersecre- tion. All patients with evidence of aldosterone hypersecretion underwent biochemical analysis for aldosterone, cortisol, androgen, and estrogen hypersecretion. Follow-up data were obtained from patients by telephone, return visits, or from infor- mation generated by their primary physician and were complete in all but one patient for a median of 3.9 years.
A single pathologist (R. L.) reconfirmed the his- tologic features of all surgical specimens. ACC was diagnosed by using established criteria.19-21 Tumors were classified into 4 grades by using criteria described previously.1 Stage of disease was assigned according to the criteria of MacFarlane22 as modi- fied by Sullivan et al.23 Comparisons were made between patients with and without aldosterone- secreting ACC in regards to tumor characteristics, patient management, and outcomes.
Statistical evaluation was performed with the Wilcoxon rank sum test to assess differences in con- tinuous measurements between groups, the chi- square test to compare categorical variables between groups, and the log rank test to compare distributions of survival.
RESULTS
Of 141 consecutive patients undergoing surgical management of ACC, we identified 15 patients (4 men, 11 women) with a mean age of 47 years (range, 16 to 77 years) having evidence of aldos- terone-secreting ACC. The mean duration of signs
or symptoms before operative intervention was 8.5 months (range, 1 to 109 months). Hypertension and hypokalemia were present in all patients. Presenting symptoms included muscle weakness (47%) and abdominal or back pain (20%). Laboratory studies consistent with aldosterone hypersecretion included hypokalemia (mean serum potassium, 2.0 mEq/L; range, 1 to 3.1 mEq/L; normal, 3.6 to 4.8 mEq/L), metabolic alkalosis (mean bicarbonate, 38 mEq/L; range, 30 to 48 mEq/L; normal, 22 to 29 mEq/L), elevated serum aldosterone concentration (mean, 78 ng/dL; range, 15 to 136 ng/dL; normal, 1 to 21 ng/dL), and elevated 24-hour urinary aldosterone excretion (mean, 72 µg; range, 17 to 249 ug; nor- mal, 2 to 16 µg). Pure aldosterone hypersecretion occurred in 10 patients. In the remaining 5 patients a mixed hormonal secretion consisting of both aldosterone and cortisol hypersecretion was observed. In addition, in 2 of these 5 patients androgen hypersecretion was also identified.
Operative management included resection with curative intent in 10 patients. Metastatic disease (n = 4) or advanced locally invasive tumor (n = 1) necessitated a palliative resection in 5 patients. Thirty-day perioperative death occurred in 3 patients (20%). One patient became coagulopath- ic immediately postoperatively and died of diffuse uncontrolled hemorrhage. A second patient died of sepsis with multisystem organ failure as a result of an infected subhepatic hematoma. The last patient was dismissed from the hospital on postop- erative day 7 in stable condition; 3 weeks later the patient’s condition rapidly deteriorated, and the patient died of unconfirmed causes. These patients were dissimilar with regards to age, comorbidities, and operative decade. Two of the patients had aldosterone and cortisol-secreting tumors, and 1 patient had pure aldosterone secretion tumor. All 3 patients had locally advanced or metastatic disease and underwent palliative resections.
Pathologic review was performed on all surgical specimens at the time of the original resection. In
100
90
80
70
Survival (%)
60
50
40
Aldosterone
30
20
10
Non-aldosterone
0
0
5
10
15
20
Years since surgery
addition, all specimens were reexamined at the time of this study to confirm the diagnosis and to establish tumor grade. The mean size and weight of the tumors were 10.8 cm (range, 4 to 18 cm) and 453 g (range, 18 to 1050 g), respectively. The tumor was located on the left in 11 patients. Grade I tumors were identified in 5 patients, grade II in 3 patients, grade III in 5 patients, and grade IV in 2 patients. There were no patients identified with stage I disease. Stage II, III, and IV tumors were identified in 9, 1, and 5 patients, respectively.
Adjuvant therapy (n = 3) or treatment for re- sidual disease (n = 2) consisting of mitotane in 4 patients and mitotane plus doxorubicin (Adriamycin) in 1 patient was instituted in 5 of 12 patients (42%) who survived the perioperative period.
Seven patients (70%) had tumor recurrence with a median interval of 17 months (range, 4 to 211 months). Overall Kaplan-Meier 5-year survival was 52% with a median survival of 62 months (range, 4 to 291 months). Patients who under- went resection with curative intent had a greater median survival than those having palliative resec- tion (80 vs 6 months, respectively). Tumor charac- teristics, patient management, and outcomes were compared between patients with aldosterone- secreting ACC and those with non-aldosterone- secreting ACC (Table I). Patients with aldosterone-secreting ACC demonstrated in- creased perioperative mortality. Although these patients also showed a trend toward improved 5- year survival (52% vs 28%), this was not statistical- ly significant (Fig 1). There were no other significant differences between the 2 groups.
DISCUSSION
Approximately one half of all ACCs are hor- monally functional.1-7 Attempts to determine the prognostic significance of a tumor’s functional abil- ity has yielded inconsistent results.1-7 Although it might be expected that functional tumors would be more commonly well-differentiated than nonfunc- tional tumors and that this would lead to an improved survival in patients with functional tumors, this has not been observed. In fact, non- functional tumors have been shown in at least 2 studies to have a favorable prognosis.1,7 However, other studies have shown no prognostic signifi- cance based on tumor function.2,3,6 A possible con- founding factor in demonstrating differences in functioning versus nonfunctioning tumors is the rarity of ACC and thus the difficulty in comparing subsets of patients. Whereas cortisol hyperfunction is the most common scenario, aldosterone hyper- function is extremely uncommon, with slightly more than 30 clearly documented cases in the lit- erature.8-18 Whether aldosterone-secreting ACC exhibits unique prognostic implications has not been previously addressed. By using data from a single institutional experience with operative man- agement of ACC, our aim was to evaluate the effect of aldosterone secretion in regards to tumor behav- ior and patient outcomes. Although a series of 15 patients is a relatively large collective of this extremely rare neoplasm, a meaningful statistical analysis of such a small number is difficult. We have attempted to delineate the clinical and prognostic implications inherent to this unique and rare neo- plasm in the first such study.
We have shown that there is an increased risk of perioperative death in patients with aldosterone- secreting ACC. In fact, despite representing only 11% of all ACCs, patients with aldosterone-secret- ing ACC accounted for 33% of all perioperative deaths in our institutional experience during a period of 44 years. Two of the patients had bleed- ing complications that ultimately led to their demise, and the cause of death in the third patient was unknown. All 3 of these patients had advanced local or metastatic disease and underwent palliative resections. However, the stage of disease or inci- dence of resection for curative intent was not dif- ferent between patients with aldosterone-secreting and non-aldosterone-secreting ACC and is unlike- ly to account for the difference in mortality. In addition to aldosterone hypersecretion, 2 of these 3 patients also had cortisol hypersecretion. Whether cortisol secretion may have attributed to the mortality is not known. In a review of the cur- rent literature about aldosterone-secreting ACC,
an increased perioperative mortality is not evident. Whether this apparent discrepancy is due to selec- tive reporting in the literature or is a function of the small number of cases in this study is not known.
In multiple large series, the overall 5-year sur- vival in treated patients with ACC ranges from 16% to 35%.1-3,6,23 In this study, the 5-year survival for patients with aldosterone-secreting ACC was nearly double that of patients without aldos- terone-secreting neoplasms (52% vs 28%). Although these numbers did not reach statistical significance, this observation is consistent with our clinical impression.
Surgical resection is the mainstay of treatment in ACC and is the only modality with curative potential. Curative resection was possible in 67% of our patients with aldosterone-secreting ACC, which is consistent with the 42% to 72% reported for all patients with ACC.1,2,6
The role of adjuvant therapy in ACC is not clear- ly defined. Mitotane, an adrenolytic agent, has been the predominant therapeutic agent used in ACC. Although randomized controlled data are not available to validate a true benefit of mitotane, anecdotal and retrospective studies suggest a potential value and serve as the basis for its current use. The use of adjuvant therapy in aldosterone- secreting ACC in this study was not significantly dif- ferent from all patients with ACC. Small total numbers preclude a meaningful analysis of any potential benefit of adjuvant therapy in patients with aldosterone-secreting ACC.
In conclusion, aldosterone-secreting ACC is an extremely rare endocrine malignancy accounting for only 11% of all ACCs and is associated with unique operative risks and outcomes. Although patients harboring these malignancies have an increased risk of perioperative death, they may also have an increased long-term survival compared to non-aldosterone-secreting ACC. Treatment of these rare neoplasms should follow the standards of ACC, with surgical resection being the mainstay of treatment.
REFERENCES
1. Kendrick ML, Lloyd R, Erickson L, Farley DR, Grant CS, Thompson GB, et al. Adrenocortical carcinoma: surgical progress or status quo? Arch Surg 2001;136:543-9.
2. Henley DJ, van Heerden JA, Grant CS, Carney JA, Carpenter PC. Adrenal cortical carcinoma: a continuing challenge. Surgery 1983;94:926-31.
3. Wooten MD, King DK. Adrenal cortical carcinoma. Cancer 1993;72:3145-55.
4. Schulick RD, Brennan MF. Adrenocortical carcinoma. World J Urol 1999;17:26-34.
5. Favia G, Lumachi F, D’Amico DF. Adrenocortical carcino- ma: is prognosis different in nonfunctioning tumors? Results of surgical treatment in 31 patients. World J Surg 2001;25:735-8.
6. Icard P, Goudet P, Charpenay C, Andreassian B, Carnaille B, Chapuis Y, et al. Adrenocortical carcinomas: surgical trends and results of a 253-patient series from the French associa- tion of endocrine surgeons study group. World J Surg 2001;25:891-7.
7. Tritos NA, Cushing GW, Heatley G, Libertino JA. Clinical features and prognostic factors associated with adrenocorti- cal carcinoma: Lahey Clinic Medical Center experience. Am Surg 2000;66:73-9.
8. Weingartner K, Gerharz EW, Bittinger A, Rosai J, Leppek R, Riedmiller H. Isolated clinical syndrome of primary aldos- teronism in a patient with adrenocortical carcinoma. Urol Int 1995;55:232-5.
9. Stone NN, Janoski A, Muakkassa W, Shpritz L. Mineral- ocorticoid excess secondary to adrenal cortical carcinoma. J Urol 1984;132:962-5.
10. Dixon AN, Bing RF. Two cases of adrenocortical carcinoma presenting as Conn’s syndrome. J Hum Hypertens 2001; 15:75-9.
11. Scott HW Jr, Sussman CR, Page DL, Thompson NW, Gross MD, Lloyd R. Primary hyperaldosteronism caused by adrenocortical carcinoma. World J Surg 1986;10:646-53.
12. Farge D, Chatellier G, Pagny JY, Jeunemaitre X, Plouin PF, Corvol P. Isolated clinical syndrome of primary aldostero- nism in four patients with adrenocortical carcinoma. Am J Med 1987;83:635-40.
13. Arteaga E, Biglieri EG, Kater CE, Lopez JM, Schambelan M. Aldosterone-producing adrenocortical carcinoma. Ann Intern Med 1984;101:316-21.
14. Fraser AG, Croxson MS, Espiner EA, Synek B. Adreno- cortical carcinoma presenting as primary aldosteronism in a young man. Aust N Z J Med 1987;17:60-2.
15. Taylor HC, Douglas JG, Berg GJ, Bravo EL. Primary aldos- teronism caused by adrenal cortical carcinoma. Endocrinol Japon 1982;29:701-8.
16. Slee PHTJ, Schaberg A, van Brummelen P. Carcinoma of the adrenal cortex causing primary hyperaldosteronism. Cancer 1983;51:2341-5.
17. Grim CE, Ganguly A, Yum MN, Donohue JP, Weinberger MH. Hyperaldosteronism due to unsuspected adrenal car- cinoma: discovery during investigation of hypertension in a young woman. J Urol 1981;126:783-6.
18. Deckers S, Derdelinckx L, Col V, Hamels J, Maiter D. Peritoneal carcinomatosis following laparoscopic resection of an adrenocortical tumor causing primary hyperaldos- teronism. Hormone Res 1999;52:97-100.
19. van Slooten H, Shaberg A, Smeenk D, Moolenaar AJ. Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 1985;55:766-73.
20. Weiss LM, Medeiros LJ, Vickery AL. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13:202-6.
21. O’Hare MJ, Monaghan P, Neville AM. The pathology of adrenocortical neoplasia: a correlated structural and func- tional approach to the diagnosis of malignant disease. Hum Pathol 1979;10:137-54.
22. MacFarlane DA. Cancer of the adrenal cortex: the natural history, prognosis and treatment in a study of fifty-five cases. Ann R Coll Surg Engl 1958;23:155-86.
23. Sullivan M, Boileau M, Hodges CV. Adrenal cortical carci- noma. J Urol 1978;120:660-5.
DISCUSSION
Dr Quan-Yang Duh (San Francisco, California). I have a couple of questions for you. One is regarding the diag- nosis. The case that you presented of an almost 5-cm lesion presumed to be an adrenal carcinoma was proba- bly a misdiagnosis from the beginning. Regarding the rest of the cases, was there any question about the diag- nosis being a benign or a malignant tumor in the first place?
I have a follow-up question related to the postoper- ative treatment for these patients. Because most of the patients with functioning tumors have Cushing’s syn- drome, do you have to treat them with steroids after- wards? Otherwise, you will get into trouble. How did you handle these patients? Was that part of the prob- lem with the subclinical function of these patients, secreting subclinical amounts of cortisone suppressing the other side and maybe not being replaced? Could that explain some of the postoperative complications that you have?
Dr Kendrick. As far as the diagnosis, the patient I have presented here was the only patient whom we did not operate on primarily. He was referred to us after his ini- tial diagnostic evaluation and operation. In other patients, the diagnosis of ACC was suggested preopera- tively based on the radiographic characteristics and clin- ical presentation. The functional status whether pure aldosterone or mixed secretory pattern was delineated prior to surgery in all patients. Management postopera- tively in these patients generally entails a prolonged cor- ticosteroid taper.
Dr Duh. Because some tumors can have subclinical function in terms of Cushing’s, a tumor may be hyper- secreting but the lab tests just don’t make it to whatever the level that you set as normal. So maybe a dexametha- sone suppression test should have been done to show that these are truly pure aldosterone-secreting tumors.
Dr Kendrick. That is very perceptive, and I agree. Most of these patients did not have dexamethasone sup- pression tests. They were purely based on the diagnosis of ACC and the functional studies that I have mentioned.
Dr Charles Proye (Lille, France). I have a very short question. Maybe I missed the point. What about your def- inition of malignancy? Is it grounded on the Weiss score for ACC or not?
Dr Kendrick. The criteria for diagnosis are based on histologic features as well as gross invasion of the tumor. Many of our patients presented with stage III or IV dis- ease; therefore, there were few circumstances where the malignancy of the tumor was in question.
Dr Proye. Did you stratify your patients according to Weiss score?
Dr Kendrick. No, we did not.
Dr L. Michael Brunt (St Louis, Missouri). I hope you will emphasize in the paper about the individual who had the laparoscopic approach. As Dr Duh said, this was a somewhat atypical lesion at 4.5 cm in diameter. But there has been one other case reported of disseminated carcinomatosis after resection of an apparently benign aldosteronoma. Since all of these tumors are now approached laparoscopically, I think it behooves us to realize that occasionally these lesions can be malignant.
My question for you is regarding the 20% mortality in the study carried out over 45 years. Where within that time frame did these perioperative deaths occur? Have you had any perioperative deaths in the last 10 or 15 years?
Dr Kendrick. That is an important question. Each peri- operative death occurred in a separate decade. The age of the patient and comorbidities were also seemingly unre- lated. However, one observation was that 2 of the 3 patients had mixed secretion, and only 1 had pure aldos- terone secretion. Thus, whether or not this mortality is truly related to the aldosterone secretion is difficult to ascertain from such a small number of patients. Therefore, because of small patient numbers, no observ- able pattern, and no other supportive studies, the conclu- sion of an increased perioperative mortality rate in patients with aldosterone-secreting tumors must be con- sidered with reservation. When looking at the survival of mixed versus pure aldosterone-secreting tumors, the over- all survival was 7 months versus 96 months. Based on these observations, it is plausible that the aldosterone-secreting tumors do portend an improved overall survival.
Dr Paolo Miccoli (Pisa, Italy). What impressed me very much is that in the first case there was a rupture of the capsule during the operation. One could think that this patient did so poorly for that reason. But I have a comment. I had a similar case where the patient was operated on at my institution, the tumor was smaller than this one, it was removed laparoscopically with no rupture, and he did exactly the same as in your case. So I think that laparoscopic adrenalectomy for ACC should be regarded with much, much caution.
Dr Kendrick. We agree emphatically.
Dr Rocco Bellantone (Rome, Italy). In the Italian series of ACC, we found that the stage is the most impor- tant factor for survival. Do you agree? The second ques- tion is, what do you think about operation for recurrence? In our series, reoperation can be useful in selected cases.
Dr Kendrick. We are with you on both counts. We have previously reported stage as a significant prognostic factor in ACC. In addition to Memorial Sloan-Kettering, we have also shown that operation for recurrent disease is warranted, in that it improves survival.