Long-Term Survival After Complete Resection and Repeat Resection in Patients With Adrenocortical Carcinoma
Richard D. Schulick, MD and Murray F. Brennan, MD
Background: One of the key issues in the treatment of adrenocortical carcinoma is the efficacy of repeat resection of local recurrence and metastatic disease in affected patients. Options in the treatment of locally recurrent or metastatic disease are limited because chemotherapy and radio- therapy generally do not provide any significant prolongation in survival in treated patients.
Methods: A series of 113 patients who presented to Memorial Sloan-Kettering Cancer Center for treatment of adrenocortical carcinoma are presented.
Results: The median overall survival for all 113 patients was 38 months (5-year survival, 37%). Patients presenting with early stage I or II disease (n = 57) had a median survival of 101 months (5-year survival, 60%), whereas those with late stage III or IV disease (n = 56) had a median survival of 15 months (5-year survival, 10%). Patients who had complete primary resection (n = 68) had a median survival of 74 months (5-year survival, 55%), whereas those with incomplete primary resection (n = 45) had a median survival of 12 months (5-year survival, 5%). Resection of locally recurrent or distant metastatic disease was performed in 47 of these patients. Patients who had a complete second resection had a median survival of 74 months (5-year survival, 57%), whereas those with incomplete second resection had a median survival of 16 months (5-year survival, 0%).
Conclusions: Improved survival is seen in patients who present with early stage and have complete primary resection. Patients who undergo complete repeat resection of local recurrence or distant metastasis also have improved survival. Complete repeat resection was more readily accomplished in discrete distant metastatic lesions compared with bulky local recurrences.
Key Words: Adrenocortical carcinoma-Recurrence-Surgical resection-Survival.
Adrenocortical carcinoma is a rare tumor with an annual incidence of 100 to 200 cases per year in the United States. This is to be contrasted with the incidence of adenomas of the adrenal gland. As many as 2% of all autopsies show adenomatous change in the adrenal gland.1
Adrenocortical tumors are classified as either func- tional or nonfunctional. Patients develop symptoms be- cause of excess amounts of steroid, androgen, estrogen, or mineralocorticoid. Steroid, androgen, and estrogen excess are much more common than mineralocorticoid
Presented at the 52nd Annual Meeting of the Society of Surgical Oncology, Orlando, Florida, March 4-7, 1999.
excess in adrenocortical carcinoma. Adrenocortical car- cinomas are inefficient in steroidogenesis and may not show obvious clinical syndromes of excess even with large bulky disease.
Adrenocortical carcinoma is staged based on the tu- mor, node, metastasis (TNM) system. As in the original staging system popularized by MacFarlane,2 stage I dis- ease is a tumor less than or equal to 5 cm without local invasion and without nodal or distant metastases. Stage II disease is the same except the tumor is more than 5 cm. Stage III disease is with local invasion or positive lymph nodes. Stage IV disease is with local invasion and pos- itive lymph nodes or distant metastases (Table 1).
MATERIALS AND METHODS
Data for this analysis were obtained from a prospec- tive data base maintained on patients admitted to and treated at Memorial Sloan-Kettering Cancer Center
| Stage | Characteristics |
|---|---|
| Stage I | <5 cm, no local invasion, neg LN, no mets |
| Stage II | >5 cm, no local invasion, neg LN, no mets |
| Stage III | Local invasion or pos LN |
| Stage IV | Either local invasion and pos LN or mets |
neg LN, negative lymph nodes; pos LN, positive lymph nodes; mets, metastases.
(MSKCC). The median length of follow-up for all pa- tients was 28 months. Survival distributions were deter- mined by the method of Kaplan and Meier.3 All survival statistics in this report are disease-specific survival. Comparisons between groups of patients were made us- ing log-rank univariate analysis as well as Cox-regres- sion multivariate analysis.
RESULTS
Patient Characteristics
In this series, 113 patients were admitted to MSKCC for treatment. Of this group of patients, 107 patients had one or more operations for adrenocortical carcinoma at MSKCC and six patients did not have a surgical proce- dure. Of the 107 patients operated on at MSKCC, 62 (58%) had their primary resection at MSKCC and 45 (42%) had repeat resection(s) at MSKCC. The male to female ratio in this series was approximately 2:3 with 45 males and 68 females. Adrenocortical carcinoma was equally prevalent on the right and left side, with 56 patients presenting with lesions on the right side and 57 patients presenting with lesions on the left side. The median age at diagnosis was 43 years. The age distribu- tion of these patients is shown in Fig. 1.
Signs and Symptoms
A majority of the patients presented with functional lesions. Functional lesions were defined as those that
14
n = 113
12
Number of Patients
10
8
6
4
2
0
0 - 5
5 - 10
10 - 15
15 - 20
20 - 25
25 - 30
30 - 35
35 - 40
40 - 45
45 - 50
50 - 55
55 - 60
60 - 65
65 - 70
70 - 75
75 - 80
Age at Diagnosis
Cushing’s Syndrome n = 35 (56%)
Virilization n = 13 (21%)
Cushing’s + Virilization n = 6 (10%)
Hyperaldosteronism n = 3 (5%)
Feminization n = 5 (8%)
showed signs and symptoms of excess adrenal hormone production. Nonfunctional lesions were defined as those that did not show signs and symptoms of excess adrenal hormone production even if laboratory studies demon- strated an elevation of one of the adrenal hormones. By these criteria, 62 patients (55%) had functional lesions (Fig. 2) and 51 patients (45%) had nonfunctional tumors (Table 2). The median size of the carcinomas at presen- tation was 14 cm (range, 4-25 cm) and the median weight at presentation was 750 g (range, 4-2600 g).
Stage at Diagnosis
Staging was performed on the basis of operative find- ings, pathologic specimens, and presence of metastatic disease. Patients whose pathology reports did not specify lymph node status and who did not have metastatic disease were classified as stage II as outlined by the criteria in Table 1. Very few patients presented with stage I disease (n = 5). Patients tended to present with later stage disease; 52 (46%) presented with stage II disease, 12 (11%) presented with stage III disease, and 44 (39%) presented with stage IV disease (Fig. 3).
Patients who initially presented with stage IV disease commonly had either liver or lung metastases. Other sites of metastatic disease included omentum, peritoneum, bone, distal lymph node basins, and the soft palate (Table 3).
Survival
The overall survival of patients with adrenocortical carcinoma was 37% at 5 years, with a median survival of
| Sign/symptom | nª | %ª |
|---|---|---|
| Pain | 33 | 65 |
| Palpable mass | 17 | 33 |
| Weight loss | 6 | 12 |
| Hematuria | 2 | 4 |
| Varicocele | 1 | 2 |
| Dyspnea | 1 | 2 |
| Asymptomatic | 3 | 6 |
ª Some patients presented with more than one major sign/symptom.
Stage I 5 (4%)
n =113
Stage IV 44 (39%)
Stage II 52 (46%)
Stage III 12 (11%)
38 months. If patients were stratified according to stage at presentation, patients presenting with stage I and II disease had similar survival curves as did patients with stage III and IV disease (Fig. 4). The median survival of patients presenting with stage I or II disease was 101 months (5-year survival, 60%) and the median survival of patients presenting with stage III or IV disease was 15 months (5-year survival, 10%; P < . 001). If patients were stratified according to completeness of primary resection, patients who underwent complete resection had improved survival (Fig. 5). Patients undergoing a complete primary resection had a median survival of 74 months (5-year survival, 55%), whereas patients under- going an incomplete primary resection had a median survival of 12 months (5-year survival, 5%; P < . 001). The functionality, age of the patient, side of the lesion, and sex of the patient were insignificant factors in sur- vival of the patient on univariate and multivariate anal- ysis (Table 4).
Repeat Resection
Of the 107 patients who were primarily resected, 47 underwent a second resection, 19 underwent a third resection, eight underwent a fourth resection, four un- derwent a fifth as well as sixth resection, and one patient underwent a seventh resection. There were a total of 83 repeat resections for local recurrence or metastatic dis- ease. The 30-day mortality of these repeat resections was 3 of 83 resections or 3.6%. Sixty-two of these repeat
| Organ/site | nª | %ª |
|---|---|---|
| Liver | 28 | 64 |
| Lung | 17 | 39 |
| Omentum/peritoneum | 8 | 18 |
| Bone | 3 | 7 |
| Distal LN | 3 | 7 |
| Palate | 1 | 2 |
LN, lymph node.
ª Some patients presented with more than one site of metastasis.
1
n total = 113
.9
Cumulative Survival
.8
.7
.6
.5
Stage II (n=52)
.4
Stage I (n=5)
.3
.2
Stage III (n=12)
.1
0
Stage IV (n=44)
0
12
24
36
48
60
72
84
96
108
120
Months after First Resection/Diagnosis
resections were complete and 21 were incomplete (Table 5). Of the complete repeat resections, 43 (69%) were for distant metastases, 14 (23%) were for local recurrence, and five (8%) were for both. Of the incomplete repeat resections, three (14%) were for distant metastases, 16 (76%) were for local recurrence, and two (10%) for both. The sites of repeat resections for distant metastases are listed in Table 6. Repeat resections for local recurrence often included the kidney and portions of the liver and, depending on which the side the lesion was on, could also involve the pancreas, spleen, and stomach.
If patients undergoing repeat resection were stratified according to completeness of second resection, those undergoing a complete resection (n = 32) had a median survival of 74 months (5-year survival, 57%), and those undergoing an incomplete resection (n = 15) had median survival of 16 months (5-year survival, 0%; P < . 001; Fig. 6). If patients who had a complete second resection were stratified according to whether the repeat resection was for distant metastasis (n = 21) or for local recur- rence (n = 11), this was not a statistically significant factor (P = . 27; Fig. 7).
1
n total = 113
.9
.8
Cumulative Survival
.7
.6
.5
Complete Resection (n = 68)
.4
No Operation (n=6)
.3
.2
.1
Incomplete Resection (n=39)
0
0
12 24 36 48 60 72 84 96 108 120
Months after First Resection/Diagnosis
| Factor | n | % | Median DSS (mo) | Univariate log rank | Multivariate Cox regression |
|---|---|---|---|---|---|
| Significant factors | |||||
| Stage I + II | 57 | 50 | 101 | P < . 001 | P < . 001 |
| Stage III + IV | 56 | 50 | 15 | ||
| Complete resection | 68 | 60 | 74 | P < . 001 | P <. 001 |
| Incomplete resection | 45 | 40 | 12 | ||
| Nonsignificant factors | |||||
| Functional | 62 | 55 | 30 | P = . 16 | P = . 17 |
| Nonfunctional | 51 | 45 | 41 | ||
| Age < 43 y | 53 | 47 | 54 | P = . 06 | P = . 10 |
| Age ≥43 y | 60 | 53 | 28 | ||
| Right side | 56 | 50 | 28 | P = . 87 | P = . 92 |
| Left side | 57 | 50 | 48 | ||
| Male | 45 | 40 | 51 | P = . 25 | P = . 10 |
| Female | 68 | 60 | 28 |
DSS, disease-specific survival.
DISCUSSION
This series of patients is similar to other large series in that there tends to be a bimodal occurrence by age with one peak at less than 5 years of age and a second peak in the fourth and fifth decades.4 Most series find a slight preponderance of female patients, but some find a male predominance.1,5,6 In a review of the English-language literature between 1952 and 1992, Wooten et al.7 found 87 series containing 1891 patients. There was a slight female (4:3) predominance. The incidence of carcinoma on the left side was 52.8%, and in 2.4% it was bilateral. The overall incidence of functional lesions was 59.3%. The current series is consistent with these findings.
The cause of adrenocortical carcinoma is unknown. It has been argued that the carcinoma develops from hy- perplastic nodules in the adrenal gland that develop into adenomas, which later develop a malignant phenotype. In support of this, some patients have abnormal adrenal steroid production preceding the development of adreno- cortical carcinoma by decades.8,9 Cytogenetic analysis suggest that loss of heterozygosity on chromosomes 11p, 13q, or 17p may be important in the pathogenesis.10,11
| Resection | n | Complete resection | Incomplete resection |
|---|---|---|---|
| 2nd | 47 | 32 | 15 |
| 3rd | 19 | 17 | 2 |
| 4th | 8 | 7 | 1 |
| 5th | 4 | 4 | 0 |
| 6th | 4 | 2 | 2 |
| 7th | 1 | 0 | 1 |
| Total | 83 | 62 | 21 |
ª 30-day mortality of repeat resection was 3 of 83, or 3.6%.
Several recent studies have implicated p53 in the patho- genesis of sporadic adrenocortical carcinoma.12-14 This is probably a late event as mutations in the conserved regions of p53 are detected more frequently in carcino- mas than in adenomas.
Adrenocortical carcinoma presents either as a func- tional tumor or secondary to mass effect. The functional lesions are broadly divided into three groups (steroid, sex hormone, or mineralocorticoid excess) on the basis of the type of hormone they produce. Steroid excess is the most common and more readily recognized presentation of the functional adrenocortical carcinomas. Patients who present with steroid excess have the classic stigmata of Cushing’s syndrome including truncal obesity, rounded facies, buffalo hump, stria, hypertension, glucose intol- erance, thinning of the skin, and osteoporosis. They sometimes have renal calculi, and if the hormonal excess is high, they may have psychiatric problems. Patients who present with sex hormone excess present with either excessive virilization in women or feminization in men. Patients with nonfunctional lesions usually present with
| n | Distant mets | Local recurrence | Both | |
|---|---|---|---|---|
| Repeat resections (n = 83) | ||||
| Complete | 62 | 43 (69%) | 14 (23%) | 5 (8%) |
| Incomplete | 21 | 3 (14%) | 16 (76%) | 2 (10%) |
| Resections involving distant metastases (n = 53) | ||||
| Lung | 29 (55%) | |||
| Liver | 15 (28%) | |||
| Bone | 6 (11%) | |||
| Brain | 1 (2%) | |||
| Omentum | 1 (2%) | |||
| Skin | 1 (2%) | |||
mets, metastases.
1
n total = 47
.9
.8
Cumulative Survival
Complete Second Resection (n = 32)
.7
.6
.5
.4
.3
.2
.1
Incomplete Second Resection (n = 15)
0
0
12 24 36 48 60 72 84 96 108 120
Months after Second Resection
a large mass. The associated symptoms are mainly re- lated to size, pressure, and invasion of contiguous struc- tures. Associated symptoms include unexplained fever, anemia, and weight loss in the presence of obesity.
In the present series, 62 patients (55%) presented with functional lesions. In this group, 35 (56%) had Cushing’s syndrome, 13 (21%) had virilization, and 6 (10%) had both Cushing’s syndrome and virilization. A minority of the patients presented with feminization (n = 5, or 8%) or hyperaldosteronism (n = 3, or 5%) (Fig. 2). Fifty-one patients presented with nonfunctional lesions, and pain, palpable mass, and weight loss were the most frequent findings (Table 1).
Unfortunately, many of the adrenocortical carcinomas, especially those that are nonfunctional, present as large masses. In the present series, the median size of the
1
n total = 32
.9
.8
Distant Metastasis (n = 21)
Cumulative Survival
.7
.6
.5
.4
.3
.2
Local Recurrence (n = 11)
.1
0
0 12 24 36 48 60 72 84 96 108 120
Months after Second Resection
carcinomas at presentation was 14 cm (range, 4-25 cm) and the median weight was 750 g (range, 4 - 2600 g).
Once an adrenocortical carcinoma is suspected, ab- dominal computed tomography (CT) is the radiology study of choice because of its availability and relatively low cost. It provides information about the extent of local invasion and the presence of metastatic disease in the liver, lymph nodes, or other sites in the peritoneal cavity. Abdominal magnetic resonance imaging is also used and gives information similar to that of CT, but may better delineate planes between the tumor and adjacent organs.
For an extent-of-disease workup, a chest x-ray or CT along with careful evaluation of the liver by the imaging modality used to evaluate the adrenal gland is usually sufficient. Further extent-of-disease workup should be tailored to the patient’s symptoms. Laparoscopy is not commonly used. Its use is somewhat tempered by the increased difficulty in visualizing the retroperitoneum and the fact that the presence of a metastatic site does not preclude resection.
A patient is considered unresectable if he or she has diffuse metastatic disease or major arterial blood vessel involvement such as the aorta. In general, an aggressive approach to local recurrence (as well as metastatic dis- ease) in a medically fit patient who has major symptoms from the disease can be justified.
In multiple large series, the overall 5-year survival in treated patients with adrenocortical carcinoma ranges from 16% to 35% (Table 7). These series consist of patients who were treated with surgery and or medical therapy, most often consisting of 1,1-dichloro-2-(o-chlo- rophenyl)-2-(p-chlorophenyl)ethane (mitotane). In the same series, if the patients’ treatment included a com- plete resection of all disease, the 5-year survival ranged from 38% to 62%. If the patients were incompletely resected, the 5-year survival ranged from 0% to 9%. These survival statistics are compatible with those in the current study. These series demonstrate that surgical resection is the primary and only potentially curative treatment for adrenocortical carcinoma. It is important to note that 5-year survival is not equivalent to cure. Most patients surviving at 5 years are alive with disease, and 85% of patients resected for cure will develop recurrence or distant metastases. 15
As listed in Table 4, the factors that were statistically significant in univariate and multivariate analysis in re- lation to survival were stage at presentation and com- pleteness of primary resection. Patients undergoing a complete primary resection had a median survival of 74 months (5-year survival, 55%), whereas patients under- going an incomplete primary resection had a median survival of 12 months (5-year survival, 5%; P < . 001).
| Study | Institution/group | Year | n | 5-y Survival (%) | ||
|---|---|---|---|---|---|---|
| Overall | Complete resection | Incomplete resection | ||||
| Soreide et al.6 | Norway | 1991 | 99 | 16 | 62 | 0 |
| Icard et al.16 | French Endocrine Surgeons | 1992 | 156 | 34 | 42 | 0 |
| Zografos et al.17 | Roswell Park | 1994 | 53 | 19 | 38 | 0 |
| Haak et al.18 | Holland | 1995 | 96 | 27 | 49 | 9 |
| Crucitti et al.19 | ACC Italian Registry | 1996 | 129 | 35 | 48 | 7 |
ACC, adrenal cortical carcinoma.
The functionality, age of the patient, side of the lesion, and sex of the patient were insignificant factors in sur- vival of the patient on univariate and multivariate anal- ysis.
The studies reviewed in Table 7 also have tried to identify clinical factors that are of prognostic signifi- cance. Soreide et al.6 demonstrated that the factors im- portant for prognosis were early stage and curative re- section. Icard et al.16 demonstrated that, in addition to stage and curative resection, age and type of hormonal secretion influenced survival. In this study, mean sur- vival times of patients with stage I, II, III, and IV tumors were 34, 40, 22, and 8 months, respectively. This trans- lated into 5-year actuarial survival rates of 33% in stage I, 55% in stage II, 24% in stage III, and 0% in stage IV. They were also able to show that the survival of patients younger than 40 years and patients with androgen-secret- ing tumors was better and that these variables were statistically significant. Zografos et al.,17 much like Sore- ide et al.,6 demonstrated that the only two factors of prognostic significance were early stage and curative resection. In this study, median survival times of patients with stage I, II, III, and IV tumors were 46, 84, 8, and 7 months, respectively. The 5-year survival rates were 33%, 57%, 18%, and 7%, respectively. Haak et al.18 also concluded that the best chance of survival for a patient with adrenocortical carcinoma is when a complete tumor resection can be performed. Crucitti et al.19 were not able to relate age or hormonal activity to survival. They again demonstrated that early stage and curative resection were the two factors of prognostic significance. In this series, the actuarial 5-year survival for patients with stage I or II disease was 54%, stage III disease was 21%, and stage IV was 7%.
In addition to identifying clinical factors, many studies have tried to identify pathological factors of prognostic significance. A recent study from MSKCC reviewed 46 patients from the prospective data base who underwent curative resection for adrenocortical carcinoma.20 Of the 16 pathologic factors analyzed, tumor size, abnormal
mitotic figures, and the presence of intratumoral hemor- rhage were independent prognostic factors. Patients pre- senting with primary tumors of at least 12 cm had a worse outcome compared with those with smaller tumors (5-year survival of 22% versus 53%). The presence of more than six abnormal mitotic figures was a negative prognostic feature when compared with the presence of zero to six abnormal mitotic figures per high-powered field (5-year survival of 13% versus 51%). Hemorrhage into the tumor was also a negative prognostic factor when compared with lesions without intratumoral hem- orrhage (5-year survival of 22% versus 53%). Patients with none of these risk factors demonstrated a favorable 5-year survival of 83%. No patient with all three patho- logic risk factors was alive at 5 years despite being completely resected. Patients with one or two of the pathologic risk factors had intermediate 5-year survival rates of 42% and 33%, respectively.
Medical Therapy
The patients in this study were referred from various institutions to MSKCC. As a consequence, they had various forms of medical therapy. Many of the patients received chemotherapy before or after surgery. The che- motherapeutic agent most often used to treat adrenocor- tical carcinoma is mitotane (Table 8). This agent was developed as an insecticide and is taken up by the adrenal cortex and causes necrosis of the adrenal cortex. Its specific cytotoxic effect on the adrenal cortex has led to its use in adrenocortical carcinoma. It is used both in primary therapy as well as adjuvant therapy settings. Venkatesh et al.21 reported on 72 patients treated with mitotane, and found an objective partial response rate of 29%. However, the 2-year survival of this group of patients was only 13%. They concluded that early diag- nosis and complete surgical excision offered the best prospects for long-term survival and that the efficacy of adjuvant mitotane needed to be evaluated further. Luton et al.22 reported on 59 patients who received mitotane, and found partial tumor regression in 14%. They did not
| Study | Institution/Group | Year | n | Response (%) | Comment |
|---|---|---|---|---|---|
| Venkatesh et al.21 | M. D. Anderson | 1989 | 72 | 29 | PRs only |
| Luton et al.22 | France | 1990 | 59 | 14 | PRs only |
| Pommier and Brennan15 | MSKCC | 1992 | 29 | 24 | PRs only |
| Wooten and King7 | English-literature review | 1993 | 551 | 35 | PRs + CRs |
| Haak et al.18 | Holland | 1995 | 55 | 27 | PRs + CRs |
CR, complete objective response; PR, partial objective response; M. D. Anderson, The University of Texas M. D. Anderson Cancer Center; MSKCC, Memorial Sloan-Kettering Cancer Center.
find a significant increase in survival in patients treated with mitotane. They concluded that mitotane therapy might offer transient benefits, particularly in controlling endocrine symptoms. Pommier and Brennan15 reported on 29 patients who received mitotane, and found partial objective responses in seven of these (24%). They did not find a significant increase in survival in patients treated with mitotane. They concluded that the value of adjuvant therapy after complete resection remains un- proven. Wooten and King7 reviewed the English-lan- guage literature from 1952 to 1992 for reports of patients with adrenocortical carcinoma. They paid particular at- tention to the use of mitotane and patient outcomes. They found 64 reports (551 patients) that had patients treated with mitotane. Of these, 194 (35%) patients had at least a partial response to mitotane. They pointed out, how- ever, the definition of partial response was varied and broad. Haak et al.18 reported on 55 patients who received mitotane and had assessable tumor because they were not operated on, only underwent a subtotal resection or had tumor recurrence. They reported that seven (13%) of these patients had a partial objective response and that eight (15%) of these patients had a complete objective response when subsequently treated with mitotane. They found increased survival times in patients treated with mitotane if serum levels were continuously kept at greater than 14 mg/liter. This difference in survival was statistically significant. In patients with serum levels kept at greater than 14 mg/liter, objective responses were found in 15 of the 27 (55%) assessable patients. They concluded that mitotane treatment for patients with in- operable, recurrent, or metastatic adrenocortical carci- noma is effective, if serum levels of the drug are main- tained at greater than 14 mg/liter.
There is significant toxicity associated with mitotane therapy.23 Almost all patients experience gastrointestinal symptoms consisting of nausea and anorexia. Neuromus- cular toxicity including lethargy, dizziness, somnolence, and depression occur in 40% to 60% of patients. Many patients are unable to perform normal work or leisure activities because of the side effects. Adrenal insuffi-
ciency is almost universal in patients taking mitotane secondary to its toxicity against normal adrenal tissue.
Radiotherapy
Radiotherapy is of very limited benefit in adrenocor- tical carcinoma. Some of the patients received radiother- apy. It was most often used for the palliation of bony metastases. It does not prolong survival, but it does appear to provide significant palliation for pain from bone metastases.4,5,24
Reoperative Treatment
Even for patients who undergo complete resection, recurrent and metastatic disease are extremely common. The only effective treatment for recurrent adrenocortical carcinoma is reoperation. Resection of recurrent and metastatic disease when complete is associated with long-term survival and provides excellent palliation in cases associated with symptomatic steroid production. In the current study, 47 of the 113 patients underwent a second resection for locally recurrent or metastatic dis- ease (Table 5). Some patients also underwent a third, fourth, fifth, sixth, and even seventh resection. There were 83 repeat resections in this group of patients. Thirty of these patients underwent repeat resection for local recurrence. If the local recurrence was on the right side, the resection potentially involved the right kidney, liver, colon, or diaphragm. On the left side, however, the resection potentially involved the left kidney, liver, co- lon, pancreas, or stomach. Fifty-three patients underwent resection of metastatic disease (Table 6). In these repeat resections, patients with distant metastases more often underwent a complete resection, when compared with those with local recurrence. A majority of the repeat resections for distant metastases were for pulmonary lesions (55%), liver lesions (28%), or bone lesions (11%). These 83 repeat resections were performed with a 30-day mortality of 3.6%. Other authors have also re- ported that successful complete repeat resections are associated with long-term survival. Icard et al.16 reported on 22 patients who underwent reoperation for local re-
currence. The 5-year actuarial survival was 27% when calculated from the time of the original operation and 16% when calculated from the time of reoperation. When the reoperation was curative, the 5-year actuarial survival calculated from the time of reoperation was 28%. Cru- citti et al.19 reported on 11 patients who underwent reoperation for recurrent disease. They reported no op- erative mortality or morbidity from these reoperations. Ten of the patients underwent tumor debulking, three underwent nephrectomy, one underwent splenectomy, and one underwent liver metastasectomy. The mean sur- vival in the patients who underwent reoperation was 42 months. In this series, the patients with recurrence who did not undergo reoperation had a mean survival time of 16 months. It is clear that patients with recurrent or metastatic disease should undergo reoperation if they have potentially resectable disease and can withstand an operation.
CONCLUSIONS
From this and other studies, it is clear that patients with adrenocortical carcinoma who present with early stage or who undergo complete primary resection have improved survival. Medical therapy and radiotherapy for this disease has modest impact, at best, in terms of long-term survival. Patients with recurrence will have improved survival if the recurrence, whether local or metastatic, can be completely removed. Complete repeat resection is more readily accomplished in discrete distant metastatic lesions compared with bulky local recur- rences.
Bulky local recurrences, however, are sometimes ame- nable to complete repeat resection. Imaging modalities such as abdominal CT and sometimes magnetic reso- nance imaging (if better delineation of tissue planes is required) should be used to identify potentially resect- able patients. Aggressive repeat resection of local recur- rence or metastatic disease is justified in the medically fit patient if complete resection is possible. A repeat resec- tion on the right side potentially involves a combination of the adrenal bed, kidney, liver, colon, and diaphragm. A repeat resection on the left side potentially involves the adrenal bed, left kidney, liver, colon, pancreas, stom- ach, and diaphragm. The decision to aggressively resect metastatic disease should involve the number of sites and organs involved, as well as the disease-free interval from last resection and the severity of symptoms the patient has.
REFERENCES
1. Brennan MF. Adrenocortical carcinoma. CA Cancer J Clin 1987; 37:348-65.
2. 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.
3. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81.
4. Norton JA. Adrenal tumors. In: De Vita VT, Hellman S, Rosenberg SA, eds. Cancer: principles and practice of oncology. 5th ed. Philadelphia: Lippincott-Raven, 1997:1659-77.
5. Cohn K, Gottesman L, Brennan MF. Adrenocortical carcinoma. Surgery 1986;100:1170-7.
6. Soreide JA, Brabrand K, Thoresen SO. Adrenal cortical carcinoma in Norway, 1970-1984. World J Surg 1992;16:663-8.
7. Wooten MD, King DK. Adrenal cortical carcinoma. Cancer 1993; 72:3145-355.
8. Plager JE. Carcinoma of the adrenal cortex: clinical description, diagnosis, and treatment. Int Adv Surg Oncol 1984;7:329-53.
9. Hamwi GJ, Serbin RA, Kruger FA. Does adrenocortical hyperplasia result in adrenocortical carcinoma? N Engl J Med 1957;257:1153-7.
10. Yano T, Linehan M, Anglard P, et al. Genetic changes in human adrenocortical carcinoma. J Natl Cancer Inst 1989;81:518-23.
11. Henry I, Grandjovans S, Couillin P, et al. Tumor-specific loss of 11p15.5 alleles in del11p13 Wilms tumor and in familial adreno- cortical carcinoma. Proc Natl Acad Sci U S A 1989;86:3247-51.
12. McNicol AM, Nolan CE, Struthers AJ, Farauharson MA, Hermans J, Haak HR. Expression of p53 in adrenocortical tumours: clini- copathological correlations. J Pathol 1997;181:146-52.
13. Ohgaki H, Kleihues P, Heitz PU. p53 mutations in sporadic adre- nocortical tumors. Int J Cancer 1993;54:408-10.
14. Reincke M, Karl M, Travis WH, et al. p53 mutations in human adrenocortical neoplasms: immunohistochemical and molecular studies. J Clin Endocrinol Metab 1994;78:790-4.
15. Pommier RF, Brennan MF. An eleven-year experience with adre- nocortical carcinoma. Surgery 1992;112:963-71.
16. Icard P, Chapuis Y, Adnreassian B, Bernard A, Proye C. Adreno- cortical carcinoma in surgically treated patients: a retrospective study on 156 cases by the French Association of Endocrine Sur- gery. Surgery 1992;112:972-80.
17. Zografos GC, Driscoll DL, Karakousis, CP, Huben RP. Adrenal adenocarcinoma: a review of 53 cases. J Surg Onc 1994;55:160-4.
18. Haak HR, Hermans J, van de Velde CJ, et al. Optimal treatment of adrenocortical carcinoma with mitotane: results in a consecutive series of 96 patients. Br J Cancer 1994;69:947-51.
19. Crucitti F, Bellantone R, Ferrante A, Boscherini M, Crucitti P, The ACC Italian Registry Study Group. The Italian registry for adrenal cortical carcinoma: analysis of a multiinstitutional series of 129 patients. Surgery 1996;119:161-70.
20. Harrison LE, Gaudin P, Brennan MF. Pathologic features of prog- nostic significance for adrenocortical carcinoma after curative re- section. Ann Surg Oncol 1999; in press.
21. Venkatesh S, Hickey RC, Sellin RV, Fernandez JF, Samaan NA. Adrenal cortical carcinoma. Cancer 1989;64:765-9.
22. Luton J, Cerdas S, Line B, et al. Clinical features of adrenocortical carcinoma, prognostic factors, and the effect of mitotane therapy. N Engl J Med 1990;322:1195-201.
23. Pommier RF, Brennan MF. Management of adrenal neoplasms. Curr Probl Surg 1991;28:657-739.
24. Percarpio B, Knowlton AH. Radiation therapy of adrenal cortical carcinoma. Acta Rad Ther Phys Biol 1976;15:288-92.