Adrenal Adenocarcinoma: A Review of 53 Cases
GEORGE C. ZOGRAFOS, MD, DEBORAH L. DRISCOLL, BA, CONSTANTINE P. KARAKOUSIS, MD, PHD, AND ROBERT P. HUBEN, MD From the Departments of Surgical Oncology (G.C.Z., C.P.K.), Biomathematics (D.L.D.), and Urologic Oncology (R.P.H.), Roswell Park Cancer Institute, Buffalo, New York
Problem: Fifty-three patients (30 men, 23 women) with histologically proven adrenal carcinoma were reviewed. Nineteen (36%) had endocrine manifestations from functioning tumors. Arteriography was positive in 95% (19/20), CT scan in 94% (17/18), and ultrasound in 92% (12/13). Seventy-six percent of the patients, at the time of diagnosis, were stage III and IV. Most common metastatic sites were the liver, lymph nodes, bone, and lungs. Local recurrence developed in 39% of cases (15/38).
Method: Forty-one patients underwent an operation. Complete surgical removal of all gross tumor was achieved in 24 patients.
Result: The overall median survival time was 8 months, and the esti- mated 5-year survival rate 19%. There were significant differences in survival between the various stages (P = 0.01) and between the group of patients who underwent complete excision of the tumor and those with incomplete resection (P = 0.002).
Conclusions: Complete surgical excision offers the best prospect for long-term survival in localized adrenal carcinoma. @ 1994 Wiley-Liss, Inc.
KEY WORDS: adrenal adenocarcinoma, treatment, prognostic variables
INTRODUCTION
Adrenal carcinomas are relatively uncommon, with a reported annual incidence of 1 patient per 1 million pop- ulation. The diagnosis is often delayed and difficult be- cause of the retroperitoneal location of these tumors, the late appearance of symptoms and possibly a low index of suspicion by physicians. Most frequently, these tumors reach a conspicuous size without clinical syndromes of cortisol, aldosterone or sex hormone hypersecretion, and the diagnosis is made at an advanced stage [1,2]. The following report presents the totality of patients with this disease seen at our Institute during the past 40 years.
The duration of presenting signs and symptoms, the accuracy of the initial diagnosis, as well as the biochemi- cal, hormonal, and radiographic changes were examined. Furthermore, the surgical treatment, postoperative com- plications, survival, and metastatic sites that developed during the course of the disease or that were found at autopsy were recorded.
PATIENTS AND METHODS
During 1950-1990, 53 patients were treated for adre- nal carcinoma at Roswell Park Cancer Institute (RPCI). There were 30 (57%) men and 23 (43%) women, with a median age of 51 years (range 21-75).
Ten additional patients were referred to us as having adrenal carcinomas by other centers, but review of the slides, or autopsy material revealed other kinds of adrenal tumors, hence they were excluded from this study.
Tumors were classified as “functioning” if there was clinical evidence of endocrine disease confirmed by ap- propriate hormonal studies. Metastases detectable at the time of initial diagnosis, during the course of the disease and at autopsy, were recorded. The disease was staged according to the following criteria: stage I was tumor <5 cm, with negative nodes, no local invasion and no me-
Accepted for publication November 17, 1993.
Address reprint requests to Dr. R. Huben, Urologic Oncology, Roswell Park Cancer Institute, Elm & Carlton Streets, Buffalo, NY 14263.
| Type | No. of patients |
|---|---|
| Cushing syndrome | 14 (26%) |
| Adrenogenital | 1 (2%) |
| Mixed | 4 (7%) |
| Nonfunctioning | 30 (57%) |
| Unknown | 4 (8%) |
tastases; stage II, tumor >5 cm with negative nodes, no local invasion and no metastases; stage III, tumor with positive nodes or local invasion, and stage IV, tumor with positive nodes and local invasion or distant metastases [3].
Survival (in months) was calculated from the date of diagnosis to death or last seen and estimated survival distributions were calculated by the Kaplan-Meier method [4]. Cox’s Proportional Hazards model was used. to perform a multivariate analysis to determine the rela- tionship of several variables to survival [5]. Tests of significance with respect to survival distributions were based on the log-rank test.
RESULTS
The primary location of the tumor was the left adrenal gland in 32 patients and the right adrenal in 19, while 2 patients presented synchronous bilateral involvement. The family history for any type cancer was positive, for one or more members, in 15 (40%) of 38 patients with recorded history.
The most common symptom was flank pain. Abdomi- nal and other palpable masses, abdominal pain, weak- ness, hypertension, and weight loss occurred frequently. Two patients with brain metastases and one with pulmo- nary metastases at the time of initial diagnosis, presented symptoms from the respective organs. Palpable masses often occurred in the supraclavicular, cervical, or axillary area, while the liver and spleen were also palpable in a fair number of patients.
In 16 of 32 patients, the diagnosis was made inciden- tally, in the process of evaluation of acute symptoms (e.g., chills and fever, nosebleeding, injury to the flank) or for such conditions as hypertension and amenorrhea. This information was unavailable for the remaining 21 patients. The initial diagnosis entertained before histo- logical proof, was inaccurate in 91% (30/33) of the pa- tients. Adrenal carcinoma coexisted with acute myelo- blastic leukemia in one case.
The mean time between first symptoms and diagnosis was 21 months (median = 8 mos). Nineteen (36%) pa- tients had endocrine manifestations and were found to have adrenocortical hormone (ACTH) excess and func- tioning tumors, as judged by serum and urine analysis (Table I). In 10 patients, plasma cortisol was elevated,
| Study | Total | Number | Sensitivity (%) | |
|---|---|---|---|---|
| Positive | Negative | |||
| IVP | 31 | 25 | 6 | 81 |
| US | 13 | 12 | 1 | 92 |
| CT | 15 | 14 | 1 | 94 |
| Arteriography | 20 | 19 | 1 | 95 |
IVP, intravenous pyelography; US, ultrasound; CT, computed tomog- raphy.
ACTH in three, renin in three, aldosterone in four, estro- gens in three, testosterone in two, and other hormones in four patients. The urine 17-ketosteroids were elevated in 13 patients, 17-OH steroids in 11, cortisol in eight, free corticol in six, and VMA in two.
Of the commonly performed laboratory tests, alkaline phosphatase, lactate dehydrogenase, sedimentation rate, transaminase and sugar were frequently increased. A de- crease of the hematocrit (Hct) was noted in 11 patients.
The diagnostic sensitivity of the various radiographic localization techniques is summarized in Table II. Arte- riography, computed tomography (CT), and ultrasound had the best sensitivity in this series.
Preoperative diagnostic biopsies of enlarged metastatic lymph nodes were performed in 10 patients and of other sites in four patients. Adrenal needle biopsy was per- formed in nine patients. The biopsy results proved the presence of malignancy (except in one case), without specifying the primary location in all the peripheral and in eight of nine adrenal biopsies.
According to the criteria mentioned above, six (11%) patients were stage I, seven (13%) stage II, 11 (21%) stage III, and 29 (55%) stage IV at diagnosis. The median diameter for the right adrenal tumors was 14 cm (range 8-30) and for the left 11 cm (0.3-35).
Forty-three of 53 (81%) patients underwent one or more operations. Extraadrenal extension of the tumor was found in 21 (66%) and localized disease in 11 (34%) of 32 patients for whom this information was recorded.
Adrenalectomy with complete surgical excision of all gross tumor was feasible in 15 of 43 patients. The surgi- cal approaches, described in detail in 32 of 43 operated patients were the thoracoabdominal (31%), anterior (28%), subcostal (25%), flank (13%), or other (3%).
Postoperative complications of varying severity oc- curred in 15 patients. In this analysis, both primary oper- ations, and secondary operations for the treatment of local or distant recurrence were considered. One of the patients developed a cardiac arrest and died at the com- pletion of the operation, immediately following resection of a large mass involving the right lobe of the liver and kidney. This was apparently due to hypovolemia because of excessive bleeding during the operation and inade-
162 Zografos et al.
quate blood replacement. A second patient died on the third postoperative day due to intraperitoneal bleeding, following resection of a left adrenal carcinoma en bloc with the ipsilateral kidney, distal pancreas, spleen, and left lateral segment of the liver. Two more patients died the 2nd and 16th postoperative days of adrenal insuffi- ciency: one of them following resection of metastasis from adrenal carcinoma causing bowel obstruction; the other patient following resection of a left adrenal tumor and kidney, and simultaneous median sternotomy for wedge resection of bilateral pulmonary metastases in- cluding a right lower lobectomy. Other recorded compli- cations were ileus (one), hematomas (two), anastomotic leak (one), sepsis (one), atelectasis (one), hypertension (one), and wound infection (one). Respiratory insuffi- ciency occurred in two patients. Of these patients, three had their operation at RPCI and eight in other centers prior to their referral.
Nonsurgical treatment was administered to patients with inoperable tumors or with distant metastases and included OP’-DDD in 23 and radiotherapy in seven. Other chemotherapeutic agents used were, Adriamycin in 11, cisplatinum in six, Cytoxan in eight, 5-FU in six, methotrexate in three, vincristine in five, streptozotosin in two, and others in 13 patients.
Fifteen patients developed local recurrences, seven of whom developed one local recurrence, two developed two, and one four local recurrences. The number of local recurrences was unavailable for five patients. Surgical treatment was used in 10 patients for the management of local recurrence.
The most common metastases at diagnosis were in the live and lymph nodes, while during the course of the illness the bones and lungs were often affected.
Of the 53 patients, one is alive with disease 6 years following diagnosis, and two are alive and disease-free 8 and 13 years after the diagnosis.
The overall median survival was 8 months (range 0.03-372) and the estimated 5-year survival rate was 19%. Stage was found to be significantly related to sur- vival (P = 0.01) (Fig. 1). The medians and estimated 5-year survival rates for stages I, II, and III were 46 months and 33%, 84 months and 57%, and 8 months 18%, respectively. For stage IV, the median survival was 7 months, and the estimated survival rate at 2 years was 7%. Patients receiving a complete excision of the tumor (median = 13 months, 5-year rate = 38%) survived significantly longer than those with incomplete excisions (median = 2 months, 5-year rate = 0%) (P = 0.002) (Fig. 2).
A multivariate analysis considering stage at diagnosis, grade and completeness of the procedure as prognostic parameters for survival was done. Complete information was available for 29 patients. The results indicated that completeness of the procedure was significantly related
1.00
0.90
Stage
Estimated Percent Surviving
0.80
0.70
0.60
0.50
0.40
0.30
2
0.20
1
0.10
3
4
0
0
38
76
114
152
190
228
266
304
342
380
Months Since Diagnosis
1.00
0.90
Extent of Procedure
Estimated Percent Surviving
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
Incomplete
Complete
0
0
38
76
114
152
190
228
266
304
342
380
Months Since Diagnosis
to survival (P = 0.04). There were no significant correla- tions between survival and tumor diameter or time from the onset of symptoms to diagnosis.
DISCUSSION
Adrenocortical carcinoma is a rare malignant disease that accounts for only 0.02% of all cancers. Its deep retroperitoneal location makes early diagnosis difficult. In our series, symptoms lasted for a mean period of 21 months before medical attention was sought. At diagno- sis, 58% of the patients with available information (29/ 50) had metastases, a phenomenon also observed by other authors [6,7].
In a relatively large number of patients (16) in this series the diagnosis was made incidentally, while other
investigators have cited a lower incidence of this occur- rence [6]. In most series, approximately two-thirds of the patients are women [1,8], whereas 57% of our patients were men.
The initial diagnosis entertained, before histological proof, was inaccurate in 91% (30/33) of the patients in this series, which suggests a low index of suspicion for adrenal carcinomas.
The most common symptom and sign were flank pain and palpable mass(es), but hypertension, weakness, and weight loss also occurred frequently [2,6,7]. Hyperten- sion may be due to hormonal excess, activation of the renin-angiotensin system due to invasion or compression by the adrenal tumor, or concurrent systemic atheroscle- rotic disease [7]. The left adrenal preponderance ob- served by other investigators was also found in this series [2,6,9].
Nonfunctioning adrenal carcinomas predominate in most series [2,6], while of 14 patients with primary Cush- ing syndrome only one had this kind of malignancy [10]. It has also been reported that very rarely patients with adrenocortical carcinoma will have a hypertensive syn- drome caused by aldosterone excess [11,12]. Four pa- tients (of seven tested) had an elevated plasma aldoster- one level, two of whom also had an elevated urine level. The blood pressure in these patients was mildly elevated (150/100 mmHg).
Recently, the radiographic imaging of the adrenal gland has improved dramatically. Tumors as small as ≤1 cm may be readily detected. The adrenal magnetic reso- nance imaging (MRI) has been reported to facilitate the preoperative diagnosis of adrenal tumors [13]. No echo patterns are characteristic enough to permit differentia- tion of adrenal adenoma from carcinoma. Smaller lesions are more likely to be benign, and larger lesions with areas of necrosis, hemorrhage, and calcification are more likely to be malignant [14].
In this series, arteriography was the most sensitive radiologic study, followed by CT scan with contrast, which is also useful in demonstrating bilateral renal func- tion in case nephrectomy becomes necessary. The scan may also give some idea as to whether the inferior vena cava contains tumor thrombus [9,15], something which may also be revealed by an ultrasound [16]. According to our experience, preoperative diagnostic biopsies do not help in identifying the primary site of origin.
The thoracoabdominal approach offers better exposure for large tumors. However, mobilization of the right lobe of the liver may provide direct access to the right adrenal gland and vein with a transabdominal incision [17]. Gen- erally, in order to avoid operative complications adequate exposure and control are of paramount importance. Flex- ibility of mind is required following the “path of least resistance” within the context of the operative plan, i.e., developing first those planes in which dissection is easy
and safe. Gentleness in dissection are obviously impor- tant and prospective anticipation of bleeding. For major vessels, e.g., renal vein and artery, unless they are easily and safely accessible at the beginning of the operation, it is generally best to delay their ligature and division until the tumor has been mobilized. Any bleeding then from these vessels can be more easily controlled with the appli- cation of clamps, the rapid removal of the tumor and individual ligation then of the vessels. At the completion of major extirpative surgery, extra time should be al- lowed for a meticulous survey of the operative field to make certain that both small and larger vessels are se- curely controlled and the field is completely dry. Adrenal insufficiency and electrolytic abnormalities are now bet- ter understood and should not cause a patient’s death in modern times. For patients with metastases from adrenal carcinoma and the primary tumor in situ, staged resection of the primary and later of the metastases, if resectable, is preferable.
Adrenal cortical carcinomas have been treated primar- ily with O,P’DDD (Mitotane) when unresectable or re- current and the objective response has varied from 34% to 61% [1]. The significant side effects of O,P’DDD and the slight prolongation in survival achieved with this agent have raised questions about its use, particularly in the patient with early, relatively asymptomatic, or stable le- sions [18].
CONCLUSION
Early and complete resection remains the key to cure. Complete resection was followed by a significantly pro- longed survival in comparison to that observed after in- complete resection.
REFERENCES
1. Karakousis CP, Uribe J, Moore R: Adrenal adenocarcinomas: Diagnosis and management. J Surg Oncol 16:385-389, 1981.
2. Grondal S, Cedermark B, Eriksson B, Grimelius L, Harach R, Kristoffersson A, Rastad J, Uden P, Akerstrom G: Adrenocortical carcinoma. A retrospective study of a rare tumor with a poor prognosis. Eur J Surg Oncol 16:500-506, 1990.
3. Norton JA, Doppman JL, Jensen RT: Cancer of the endocrine system. In De Vita T, Hellman S, Rosenberg SA (eds): “Cancer — Principles and Practice of Oncology.” 3rd Ed. Philadelphia: J.B. Lippincott Co., 1989, p 1269-1344.
4. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am State Assoc 53:457-486, 1958.
5. Cox DR: Regression models and life-tables. J R Stat Soc B 34: 187-220, 1972.
6. Venhatesh S, Hickey R, Sellin R, Fernandez J, Samaan N: Adre- nal cortical carcinoma. Cancer 64:765-769, 1989.
7. Bodie B, Novick A, Pontes E, Straffon R, Montie J, Babiak T, Sheeler L, Schumacher P: The Cleveland Clinic Experience with adrenal cortical carcinoma. J Urol 141:257-260, 1989.
8. Schwarz R, Schmidt N: Efficient management of adrenal tumors. Am J Surg 161:576-579, 1991.
9. Cohn K, Gottesman L, Brennan M: Adrenocortical carcinoma. Surgery 199:1170-1179, 1986.
10. Perry R, Nieman L, Cutler G, Chrousos G, Loriaux L, Doppman
J, Travis W, Norton J: Primary adrenal causes of Cushing’s syn- drome. Ann Surg 210:59-68, 1989.
11. Merrell R: Aldosterone-producing tumors (Conn’s syndrome). Semin Surg Oncol 16:66-70, 1990.
12. Greathouse D, McDermott M, Kidd G, Hoefeldt F: Pure primary
hyperaldosteronism due to adrenal cortical carcinoma. Am J Med 76:1132, 1984.
13. Doppmen J, Reinig J, Dwyer A, et al: Differentiation of adrenal masses by magnetic resonance imaging. Surgery 102:1018-1026, 1987.