CLINICAL FEATURES OF ADRENOCORTICAL CARCINOMA, PROGNOSTIC FACTORS, AND THE EFFECT OF MITOTANE THERAPY
JEAN-PIERRE LUTON, M.D., SONIA CERDAS, M.D., LINE BILLAUD, M.D., GUY THOMAS, M.D., BRIGITTE GUILHAUME, M.D., XAVIER BERTAGNA, M.D., MARIE-HÉLÈNE LAUDAT, M.D., ALBERT LOUVEL, M.D., YVES CHAPUIS, M.D., PHILIPPE BLONDEAU, M.D., ANDRÉ BONNIN, M.D., AND HENRI BRICAIRE, M.D.
Abstract Adrenocortical carcinoma is a rare tumor, and only limited information is available about its natural histo- ry and the effects of therapy.
We studied 105 patients (75 female and 30 male; mean age, 46 years) with adrenocortical carcinoma who were referred to us between 1963 and 1987. The average du- ration of symptoms before diagnosis was 8.7 months. At the time of diagnosis, 68 percent of the patients had endocrine symptoms, and 30 percent had distant metas- tases. Hormonal studies showed that 79 percent of the tumors were functional. Eighty patients underwent surgery, and 59 also received the adrenal cytotoxic agent mitotane. The median disease-free interval after surgery was 12.1 months (range, 1 to 175). Tumor dissemination
A’ DRENOCORTICAL carcinoma is a rare tumor1,2 with a poor prognosis.3-6 Its true incidence is un- known,7 but according to the Third National Cancer Survey of 19758 the incidence is approximately 0.5 to 2 cases per million people per year, accounting for 0.2 percent of the deaths from cancer in the United States. 1,2,9,10 The infrequent occurrence of this carci- noma precludes extensive clinical experience with its diagnosis and treatment, and reports of more than 50 patients from single or multiple centers are excep- tional.3-5,9,11-13
Surgical resection is the treatment of choice,5,9,14-16 but the role of adjuvant chemotherapy is controver- sial.5,11,17-20 Mitotane (o,p’-DDD or 1,1-dichlorodi- phenyldichloroethane), which was introduced in 1960,21 is the only drug that has been associated with cures, long-term remissions, regression of metastases, and increases in survival.9,14,21-23 However, there have been no randomized or controlled studies of its effica- cy in prolonging survival. Furthermore, its use after apparently successful surgery, as has been proposed by some investigators,9,12,23,24 is controversial.
The objectives of this study were to describe the clinical and hormonal characteristics of 105 patients with adrenocortical carcinoma who were referred to a single center, to review the outcomes in the 88 patients who were treated and followed up at this same center, to identify potential prognostic factors, and to evalu-
From the Clinique des Maladies Endocriniennes et Métaboliques (J .- P.L., S.C., L.B., B.G., X.B., H.B.), the Laboratoire d’Hormonologie (M .- H.L.), the Service d’Anatomo-Pathologie (A.L.), the Clinique Chirurgicale (Y.C.), and the Service de Radiologie A (A.B.), Hôpital Cochin; the Département de Toxicologie Clinique, Hôpital Fernand Widal (G.T.); and Chirurgie IV, Hôpital Broussais (P.B.), all in Paris. Address reprint requests to Dr. Luton at the Clinique des Maladies Endocriniennes et Métaboliques, Hôpital Cochin, 27, rue du Faubourg Saint Jacques, 75674 Paris Cedex 14, France.
occurred in 82 percent of the patients, most commonly to the lung, liver, and adjacent organs. The median survival time was 14.5 months (range, <1 to 175), and the five- year survival was 22 percent. Age over 40 years and the presence of metastases at the time of diagnosis were the only factors recognized as indicating a poor prognosis. Mitotane controlled hormonal secretion in 75 percent of the patients. Eight mitotane-treated patients had partial tumor regression, but the drug did not have a significant effect on survival.
We conclude that adrenocortical carcinoma carries a poor prognosis. Mitotane therapy may offer transient benefits, particularly in controlling endocrine symptoms. (N Engl J Med 1990; 322:1195-201.)
ate the effect of mitotane on the outcome in these patients.
METHODS
Patients
One hundred five patients with adrenocortical carcinoma were referred to the Endocrine Department of the Hôpital Cochin from 1963 through 1987. These patients underwent clinical, radiologic, and hormonal evaluation. The diagnosis in each patient was con- firmed by histologic examination of adrenal tumor tissue.
Follow-up
Eighty-eight patients were followed up at the same hospital by a single group of physicians for a mean of 24.9 months (median, 12; range, <1 to 175). Each patient underwent clinical evaluation four times a year, and hormonal and radiologic studies were performed twice a year.
Definitions
Patients were considered to have functional tumors if they had abnormal hormone levels in serum, urine, or both. The secretion of glucocorticoids was assessed by measurements of urinary excretion of 17-hydroxycorticosteroids, 17-ketosteroids, and cortisol, basally and after low-dose (2 mg per day) and high-dose (8 mg per day) dexamethasone suppression, and by the measurement of plasma cortisol concentrations every four hours. Androgen secretion was evaluated by the measurement of plasma testosterone, androstene- dione, and dehydroepiandrosterone sulfate; estrogen secretion by the measurement of plasma estradiol-178; and mineralocorticoid secretion by the measurement of plasma aldosterone and 11-desox- ycorticosterone. We also measured plasma 17-hydroxyprogesterone and 11-desoxycortisol. Urinary excretion of 17-hydroxycorticoste- roids and 17-ketosteroids was measured by the Porter-Silber and Zimmerman methods, respectively, urinary and plasma cortisol were measured by a competitive protein-binding assay,25 and the other plasma steroids were measured by specific radioimmunoas- says in routine use at the time.
Staging at the time of diagnosis was based on clinical data and the results of radiologic studies, CT scanning, or both, and was corroborated by the findings at surgery and on pathological exami- nation. Localized disease was defined as disease confined to one
adrenal gland. Regional disease was defined as disease extending beyond the limits of the adrenal gland to the surrounding organs and tissues, including the regional lymph nodes. Metastatic disease was defined as disease in which there were distant metastases, whether regional disease was present or absent. In the patients who had no metastases initially, the disease-free interval was de- fined as the time between curative surgery and a local recurrence or the occurrence of distant metastases. Survival periods were calculated from the time of diagnosis. When possible, the extent of disease was measured on the roentgenograms, CT scans, or both. During follow-up, the diagnosis of distant metastases or local recurrences was based on clinical and radiologic findings and on examination at autopsy (of 67 patients who died, 13 were examined at autopsy). A partial regression was defined as a regression of the area of at least one measurable tumor deposit by more than 10 percent.
Treatment
Surgical Treatment
Of the 88 patients treated at the Hôpital Cochin, 80 underwent surgery; in 57 it was apparently curative. Eleven patients had sec- ond operations for local recurrences. Eight (seven female and one male patient; mean age, 44 years; range, 7.5 to 67) were judged to have inoperable tumors and were treated medically.
Mitotane Treatment
Fifty-nine of the 80 patients who underwent surgery received mitotane therapy at some time, after giving their informed consent (Table 1). Twenty patients with Cushing’s syndrome were treated with mitotane for one to six months before surgery, and 52 patients, including 13 of those treated preoperatively, were treated after sur- gery. In most patients mitotane therapy was continued until death, whether or not there was a response, a progression of disease, or a recurrence. Included among the 59 patients who received mitotane were 23 who received it after apparently curative surgery and 14 who received it only after recurrence of their disease.
Mitotane (Roussel-UCLAF) was given by mouth, as capsules containing 0.5 g of micronized mitotane mixed with cellulose ace- tylphthalate. The mean initial dose was 10 g per day (range, 4 to 16). The mean maintenance dose (the daily dose used for the longest period during treatment) was 7 g per day (range, 3 to 20). The daily dose was given in three divided doses. The median duration of treatment was 10.5 months (range, 1 to 96). Thirty to 40 mg of cortisol and 50 µg of 9a-fluorocortisol were given daily with the mitotane, usually starting after the third week of mitotane treatment. This corticosteroid coverage was temporarily withdrawn at the time of each subsequent hormonal evaluation, to assess functional adrenal status. Most patients took mitotane without in-
| TREATMENT | No. OF PATIENTS | MITOTANE | ADJUVANT TREATMENT | |||||
|---|---|---|---|---|---|---|---|---|
| BEFORE SURGERY | AFTER SURGERY | ON RECURRENCE | AMINO- GLUTETHI- MIDE | CHEMO- THERAPY | RADIO- THERAPY | 2ND SURGERY | ||
| no. of patients | ||||||||
| Mitotane therapy | 59 | |||||||
| 7 | 7* | 2 | ||||||
| 25 | 25 | 4 | 4 | 8 | 3 | |||
| 14 | 14 | 4 | 1 | 6 | ||||
| 13 | 13+ | 13 | 4 | 2 | ||||
| No mitotane therapy | 21 | 3 | 2 | 1 | 1 | |||
*Treatment lasted ≤ two months in five patients and six months in two patients. +Treatment lasted from one to five months.
terruptions in treatment, and they were strongly encouraged at each visit to take the drug as prescribed.
From 1963 through 1970, 4 patients also were treated with che- motherapy, 11 with radiotherapy, and 14 with aminoglutethimide (Table 1).
Statistical Analysis
The statistical associations between pairs of variables were tested by the chi-square test, the Kruskall-Wallis test, and the Spearman rank correlation coefficient, as appropriate. All P values are two- tailed.
Survival rates were calculated by the Kaplan-Meier method26 and were compared with use of the Tarone-Ware statistic.27 The prognostic factors were evaluated with use of the Cox proportional- hazards model.28
Among the patients treated with mitotane, the heterogeneity of the therapeutic regimens precluded the definition of a treated and an untreated group. To evaluate the effect of mitotane therapy on survival, an additional analysis was performed with use of the time- dependent Cox proportional-hazards model.28,29 In this model, the variable for mitotane therapy was assigned a value of 1 from the beginning of the third month of therapy with the drug until two months after the end of each period of treatment. A value of 0 was assigned during the remaining months of follow-up. This two- month period reflected our best estimate of the delay in the action of mitotane after its initiation and the persistence of action after its cessation, based on the drug’s known kinetics. The value of the variable for metastasis was 0 in the absence of metastases and 1 after a metastasis was found. This form of analysis was intended to deal with the problem of the heterogeneity of treatment periods among the treated patients, also taking into account the kinetics of mitotane.
RESULTS
Clinical Characteristics and Initial Manifestations of Disease
The distribution according to sex and age of all 105 patients with adrenocortical carcinoma is shown in Figure 1. The sex ratio (female to male) was 2.5, and the mean age was 46 years (range, 6 to 81). The male patients tended to be older than the female patients (P = 0.07). Table 2 shows the initial clinical findings in these patients. Seventy-one patients (68 percent) were referred to us because of endocrine symptoms and signs that indicated excess secretion of glucocor- ticoids (in 32 patients), androgens (in 8 patients), glucocorticoids and androgens con- comitantly (in 28 patients), miner- alocorticoids (in 2 patients), or es- trogens (in 1 male patient). Sixteen of these 71 patients also had local or systemic symptoms of neoplastic disease. In seven female patients, adrenocortical carcinoma was dis- covered during pregnancy because of glucocorticoid excess. In one pa- tient, the abdominal mass was asso- ciated with symptomatic hypogly- cemia.
The functional status of the 105 carcinomas is shown in Table 3. The sex ratio was 3.6 in the patients with functional tumors and 0.8 in
Number of Patients
(21)
20
(16)
10
(11)
(6)
(7)
(8)
(7)
(8)
(2)
(2)
(3)
(4)
(3)
(2)
(4)
(1)
≤10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
Age at Diagnosis (years)
those with nonfunctional tumors (P = 0.005). The most common abnormalities were concomitant hyper- secretion of glucocorticoids and androgens (in 42 per- cent), hypersecretion of glucocorticoids alone (in 41 percent), and hypersecretion of androgens alone (in 10 percent). Androgen-secreting tumors were more frequent in the female patients (P = 0.003).
The mean time from initial symptoms to diagnosis was 8.7 months (range, 0 to 53). This interval was similar in the male and female patients and in the patients with functional tumors and those with non- functional tumors. Fifty-two of the 105 carcinomas (50 percent) involved the left adrenal gland, and 49 (47 percent) the right; in three patients, both adrenal glands were involved. One patient presented with a carcinoma in the pelvis; presumably, it originated in ectopic adrenal tissue. Of 76 tumors, the mean tumor weight was 532 g (range, 14 to 3000).
Among the 88 patients followed up in our center, 45 were found to have local disease (51 percent), 17 re- gional disease (19 percent), and 26 metastatic disease (30 percent). Eleven patients in the last group had associated regional disease. The ratio of functional to nonfunctional tumors was similar in each of the three stages.
Follow-up Studies and Survival
Among the 57 patients who underwent curative sur- gery, the median disease-free interval was 12.1 months (range, 1 to 175). The intervals were similar in the patients with functional and those with nonfunctional tumors. The presence of regional disease at the time of diagnosis was associated with a significantly shorter disease-free interval (P = 0.024).
Among the 88 patients followed up, the most com- mon sites of tumor dissemination were the liver (43 patients), the lung (43), and organs or tissues adjacent to the adrenal glands (41). Less common sites of dis- ease were bone (13 patients), pancreas (4), small bow- el (3), brain (2), ovary (1), orbit (1), skin (1), pleura (1), spleen (1), and contralateral adrenal gland (1). The most common combination was of the lung and liver (22 patients).
Of the 67 patients who died, the majority (82 per-
cent) died with metastastic disease. In this group, eight patients had mediastinal compression, seven had an inferior vena cava syndrome, five had neurologic complications, four had gastrointestinal hemorrhage, one had hepatic coma, and one had septicemia. Eight patients (10 percent of the patients who underwent surgery) died of acute surgical complications - car- diovascular collapse in five, pulmonary edema in one, and pulmonary embolism in two.
Figure 2 shows the actuarial survival data for the 88 patients. The median survival time was 14.5 months (range, <1 to 175). At five years, 22 percent of the patients were alive. There was no statistically signifi- cant difference in survival with regard to sex or the functional status or weight of tumors. In contrast, sur- vival was significantly lower in the patients over 40 years old and in those with distant metastases at the time of diagnosis (Fig. 3).
Evaluation of Treatment
We evaluated the efficacy of mitotane in regard to hormonal response, tumor response, and survival from the time of diagnosis. Fifty-nine of the 80 pa- tients who had surgery received mitotane (Table 1) --- 45 soon after diagnosis, and 14 at the time of relapse. In the 45 patients who were treated before surgery, immediately afterward, or both, mitotane therapy was not found to be associated with the year of entry into the study, sex, tumor weight, functional-tumor status, or disease stage at diagnosis. However, the patients who received mitotane were younger than those who did not (P = 0.02). Twenty-three patients received mitotane despite having had apparently curative surgery.
Of the 59 patients treated with mitotane in whom the responses were evaluated, 47 had functional tu- mors and 12 had nonfunctional tumors. These pa- tients all had adrenal insufficiency, which was perma- nent in 75 percent of them. Fourteen patients whose tumors were initially functional had recurrences of hormone hypersecretion when distant metastases or local recurrences developed. In one patient with a mixed excess of glucocorticoids and androgens, only cortisol secretion decreased during mitotane therapy.
Tumor response was evaluated in 37 patients with measurable disease who were treated with mitotane
| FEATURE | No. OF PATIENTS | % |
|---|---|---|
| Endocrine symptoms and signs | 71 | 68 |
| Abdominal pain | 29 | 28 |
| Weight loss, anorexia, fever, weakness | 24 | 23 |
| Abdominal mass | 8 | 8 |
| Metastases | 6 | 6 |
| Chance symptoms | 1 | 1 |
for at least two months. Twenty-seven of these pa- tients had tumor progression, two had stabilization of their disease (during 36 to 56 months of therapy, re- spectively), and eight had tumor regression. In all eight the regression was transient. Details of the re- sponses of these eight patients are shown in Table 4; none of them received other therapy. Tumor regres- sion occurred only in patients who received mitotane therapy.
To evaluate the influence of treatment on survival, the Cox proportional-hazards model was used, with treatment and the presence of metastasis as time- dependent covariates. The results of this analysis are shown in Table 5. Only age (P = 0.06) and the pres- ence of metastasis (P = 0.0001) were found to in- fluence cumulative survival independently. The effect of mitotane therapy on survival was not significant (P = 0.24).
Side Effects
Twelve patients had side effects of sufficient sever- ity to require a reduction or temporary withdrawal of mitotane therapy; in three patients the drug was with- drawn permanently. The major side effects were gas- trointestinal symptoms, such as anorexia, nausea, and vomiting (eight patients), and neurologic disturb- ances, such as ataxia, speech difficulty, somnolence, vertigo, and lethargy (eight patients). Most patients had hepatic disturbances, including increases in se- rum levels of alkaline phosphatase, aspartate amino- transferase, alanine aminotransferase, and gamma glutamyl transferase activity; increases in the serum levels of corticosteroid-binding globulin and sex-hor- mone-binding globulin; hypercholesterolemia; and hypouricemia; some patients had low serum concen- trations of total thyroxine, free thyroxine, or both, but normal concentrations of thyroid-stimulating hor- mone. Gynecomastia developed in 7 of the 21 male patients who were treated with mitotane.
DISCUSSION
Our study of 105 patients with adrenocortical carci- noma from a single center confirms previous re-
| FUNCTIONAL STATUS/STEROIDS SECRETED | No. OF TUMORS | % |
|---|---|---|
| Functional tumors | 83 | 79 |
| Glucocorticoids | 34 | 41 |
| Glucocorticoids and androgens | 35 | 42 |
| Androgens | 8 | 10 |
| 17-Hydroxyprogesterone | 2 | 2 |
| Aldosterone | 1 | 1 |
| Glucocorticoids and 17-hydroxyprogesterone | 1 | 1 |
| Glucocorticoids, 17-hydroxyprogesterone, and aldosterone | 1 | 1 |
| Glucocorticoids, androgens, and estradiol | 1 | 1 |
| Estradiol | 0 | 0 |
| Nonfunctional tumors | 22 | 21 |
100
80
Actuarial Survival (%)
60
40
20
0
4
20
40
60
80
100
175
Months
ports4,9,11,24,30,31 that the majority of patients with adre- nocortical carcinoma are female (sex ratio, 2.5). This tumor occurs at all ages; however, the male patients tended to be older than the female patients. In chil- dren, adrenocortical carcinomas are uncommon,31 often manifested by a virilizing syndrome,32-37 and sometimes associated with other cancers or congenital malformations, such as hemihypertrophy or viscero- megaly.38-40
Whereas some reports indicate the predominant site to be the left adrenal gland, 16,24,41 we found the num- bers of tumors in each adrenal gland to be simi- lar.3,19,42 Ectopic43 and bilateral4,10,15,29 carcinomas are extremely rare. The tumors are generally large44-46; in our series, 84 percent weighed more than 100 g. On the basis of hormonal assays, 79 percent of our pa- tients had functional tumors. Among them were 12 patients who had no clinical manifestations of hor- mone hypersecretion, some of whose tumors produced only steroid precursors.47 Estrogen-secreting adreno- cortical carcinomas are rare, and occur almost exclu- sively in male patients48,49; only one of our patients had such a tumor, and it secreted other hormones in addition to estrogen.
The most common clinical presentation was an en- docrine syndrome, usually Cushing’s syndrome, either in isolation or associated with virilizing features.4,9,24,42 Hyperaldosteronism is unusual.50-55 The clinical pres- entation was nonendocrine in 31 percent of the pa- tients, which is a lower proportion than has been de- scribed in some reports5,15,16,24 but higher than has been described in others.11,56 In the literature, the mean time from the onset of symptoms to the diagno- sis of tumor ranges from 6 to 14 months2,11,16,40; in our series, this interval was 9 months. The interval was not shorter in patients with functional tumors, as
has been proposed previously5,24; and as reported by Bertagna and Orth,11 this interval was similar in female and male patients.
The diagnosis of adrenocortical cancer was made in 14 women dur- ing pregnancy. The discovery of Cushing’s syndrome during preg- nancy is uncommon, presumably because it often results in infertility. Since Cushing’s disease is the most frequent cause of hypercortisolism in young women, the preponder- ance of adrenal tumors as the cause of this condition in pregnant wom- en is surprising.57,58
A high proportion of the patients described (50 to 80 percent) have had metastatic disease at the time of diagnosis5,12,16; in the present se- ries, the proportion was somewhat smaller (30 per- cent). Metastatic disease ultimately developed in the majority of patients (82 percent), frequently involving the lung, liver, and organs or tissues adjacent to the
| PATIENT No. | DURATION OF SURVIVAL | DURATION OF MITOTANE TREATMENT* | TUMOR REGRESSION | OTHER METASTASESŤ | ||
|---|---|---|---|---|---|---|
| SITE | PERCENT | DURATION | ||||
| months | months | |||||
| 1 | 50 | 2 | Adrenal gland | 10 | 12 | Liver |
| 2 | 12 | 4 | Lung | 80 | 6 | - |
| 3 | 18 | 6 | Lung | 80 | 5 | Liver, adrenal gland |
| 4 | 53 | 6 | Lung | 60 | 12 | - |
| 5 | 33 | 7 | Lung | 50 | 25 | Bone, liver |
| 6 | 96 | 12 | Lung | 80 | 24 | - |
| 7 | 91 | 14 | Liver | 30 | 6 | - |
| 8 | 96 | 48 | Lung | 25 | 24 | Liver |
*Before tumor regression.
+Denotes the occurrence or progression of other metastases during the period of tumor regression at the site indicated.
100
100
80
Age ≤40 years
Age >40 years
80
Without metastases
Actuarial survival (%)
With metastases
60
60
40
40
20
20
0
0
-
20
40
60
80
100
175
20
40
60
80
100
175
Months
Months
For the comparison of the younger and older groups, P = 0.035 and the Tarone-Ware statistic27 = 4.467. For the comparison of patients with and without metastases, P = 0.007 and the Tarone-Ware statistic = 7.245.
adrenal gland. Among the 57 patients in this series who underwent curative surgery, the median disease- free interval was 12.1 months (range, 1 to 175), which is similar to the interval reported by Nader et al.5 In three patients the interval was exceptionally long - more than 10 years.59,60
Like others,4,5,7,12,16 we found the prognosis of pa- tients with adrenocortical carcinoma to be poor, with the five-year survival rates ranging between 20 and 30 percent in the more recent studies.5,15,16 Didolkar et al.16 identified sex and tumor stage as prognostic fac-
tors. Our results confirm the prognostic value of stag- ing. We also found an influence of age. Like others, however, we found no effect of sex or of the weight5,41 or functional status5,14,41 of the tumor.
The first therapeutic approach to adrenocortical carcinoma is surgical resection,7,9,14-16,22,61 even in pa- tients with extensive metastatic disease. Radiotherapy is ineffective and should be considered only as pal- liative treatment for metastatic disease.10,22,24,41,45,62 Adjuvant chemotherapy has been used to control hor- monal excess and induce tumor regression. The expe- rience with cytotoxic agents other than mitotane is limited, and their effects are uncertain.20,63-65 Drugs that block the synthesis of steroids, such as metyra- pone, aminoglutethimide, and more recently keto- conazole, or that block the action of steroids in their target tissue, such as antiandrogen agents, antiminer- alocorticoid agents, and more recently antiglucocorti- coid agents such as mifepristone (RU 486), may be of benefit in controlling the clinical manifestations in- duced by hypersecretory tumors; none have been re- ported to inhibit tumor growth. Mitotane has been used extensively because of its specific cytotoxic effect. In most retrospective studies, it was used only in pa- tients with inoperable tumors.9,14 Its antineoplastic ac-
| VARIABLE | REGRESSION COEFFICIENT | P VALUE |
|---|---|---|
| Year of entry into the study | 0.008±0.023 | 0.74 |
| Age | 0.017±0.0019 | 0.06 |
| Sex | 0.030±0.376 | 0.94 |
| Weight of tumor | -0.0003±0.0002 | 0.27 |
| Functional status | 0 | 1.0 |
| Regional disease | -0.365±0.284 | 0.20 |
| Metastasis | 1.624±0.408 | 0.0001 |
| Mitotane therapy | -0.928±0.790 | 0.24 |
*Plus-minus values are means ±SD.
tion has been considered to be unpredictable, incon- sistent, restricted, and transient.5,11,17,19,44,66 With the exception of van Slooten et al., 12 who measured serum mitotane levels, to our knowledge no previous investi- gators have documented improved survival as a result of mitotane therapy.2,7 The main limitations of previ- ous reports9,14,21 have been the absence of a control group and the lack of details about staging, previous treatment, and the time of the initiation of mitotane therapy.
Our retrospective analysis has some obvious, un- avoidable limitations. In this study, which started 26 years ago, treatment was not assigned at random. Therefore, our first step was to search for any possible factor that could bias the assessment of survival. The two groups of patients were comparable, with the sole exception of age. The use of the time-dependent Cox proportional-hazards model was prompted by the need to solve the problem of group assignment. This approach should reduce bias but is unlikely to eradi- cate it. A positive bias may result from the fact that the patients treated soon after diagnosis were younger than those who were not treated until they had a recurrence. Conversely, a negative bias is likely to re- sult from the use of mitotane in the patients who were doing worse (those who had metastases). We found that mitotane prolonged survival slightly but not significantly. Moreover, eight patients had objec- tive tumor regression, as has been reported by oth- ers.2,23,59,67-74 No regression occurred in patients not treated with the drug. These observations suggest that in some patients, at least, mitotane is beneficial. It controlled endocrine hypersecretion in 75 percent of the patients, as reported elsewhere.9,17,67 Patient com- pliance was high, as indicated by the frequency of adrenal insufficiency and abnormalities in liver func- tion or serum lipid or uric acid concentrations. The side effects of mitotane were minimal, contrary to what others have reported.9,14,17,75-77 This difference in side effects may be explained by differences in drug formulations. 62,78
In view of these findings, and despite the lack of a significant effect on survival, we believe that mitotane should be used as adjuvant therapy after aggressive surgery and as a reference treatment in controlled tri- als with other drugs in patients with adrenocortical carcinoma.
We are indebted to Dr. Christine Bertagna for careful review of the manuscript and to Ms. Dominique Guiban, Ms. Michèle Le Scouarnec, Ms. Marie-Claude Rizzo, and Ms. Laurette Vasselier for assistance in the preparation of the manuscript.
REFERENCES
1. Steiner PE. Cancer: race and geography: some etiological, environmental, ethnological, epidemiological, and statistical aspects in Caucasoids, Mon- goloids, Negroids and Mexicans. Baltimore: Williams & Wilkins, 1954:25.
2. Plager JE. Carcinoma of the adrenal cortex: clinical description, diagnosis, and treatment. Int Adv Surg Oncol 1984; 7:329-53.
3. 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.
4. Hutter AM Jr, Kayhoe DE. Adrenal cortical carcinoma: clinical features of 138 patients. Am J Med 1966; 41:572-80.
5. Nader S, Hickey RC, Sellin RV, Samaan NA. Adrenal cortical carcinoma: a study of 77 cases. Cancer 1983; 52:707-11.
6. Heinbecker P, O’Neal LW, Ackerman LV. Functioning and nonfunctioning adrenal cortical tumors. Surg Gynecol Obstet 1957; 105:21-33.
7. Brennan MF. Adrenocortical carcinoma. CA 1987; 37:348-65.
8. Biometrics Branch, National Cancer Institute. Cutler SJ, Young JL, eds. Third National Cancer Survey: incidence data. National Cancer Institute monograph 41. Washington, D.C .: Government Printing Office, 1975. (DHEW publication no. (NIH) 75-787.)
9. Lubitz JA, Freeman L, Okun R. Mitotane use in inoperable adrenal cortical carcinoma. JAMA 1973; 223:1109-12.
10. Bradley EL III. Primary and adjunctive therapy in carcinoma of the adrenal cortex. Surg Gynecol Obstet 1975; 141:507-11.
11. Bertagna C, Orth DN. Clinical and laboratory findings and results of therapy in 58 patients with adrenocortical tumors admitted to a single medical center (1951 to 1978). Am J Med 1981; 71:855-75.
12. van Slooten H, Moolenaar AJ, van Seters AP, Smeenk D. The treatment of adrenocortical carcinoma with o,p’-DDD: prognostic implications of serum level monitoring. Eur J Cancer Clin Oncol 1984; 20:47-53.
13. Dix VW. Tumours of the adrenal cortex. Br J Urol 1963; 35:356-66.
14. Hutter AM Jr, Kayhoe DE. Adrenal cortical carcinoma: results of treatment with o,p’DDD in 138 patients. Am J Med 1966; 41:581-92.
15. Greenberg PH, Marks C. Adrenal cortical carcinoma: a presentation of 22 cases and a review of the literature. Am Surg 1978; 44:81-5.
16. Didolkar MS, Bescher RA, Elias EG, Moore RH. Natural history of adrenal cortical carcinoma: a clinicopathologic study of 42 patients. Cancer 1981; 47:2153-61.
17. Hoffman DL, Mattox VR. Treatment of adrenocortical carcinoma with o,p’-DDD. Med Clin North Am 1972; 56:999-1012.
18. Ostuni JA, Roginsky MS. Metastatic adrenal cortical carcinoma: document- ed cure with combined chemotherapy. Arch Intern Med 1975; 135:1257- 8.
19. Hajjar RA, Hickey RC, Samaan NA. Adrenal cortical carcinoma: a study of 32 patients. Cancer 1975; 35:549-54.
20. Haq MM, Legha SS, Samaan NA, Bodey GP, Burgess MA. Cytotoxic chemotherapy in adrenal cortical carcinoma. Cancer Treat Rep 1980; 64:909-13.
21. Bergenstal DM, Hertz R, Lipsett MB, Moy RH. Chemotherapy of adreno- cortical cancer with o,p’DDD. Ann Intern Med 1960; 53:672-82.
22. Orth DN, Liddle GW. Results of treatment in 108 patients with Cushing’s syndrome. N Engl J Med 1971; 285:243-7.
23. Bricaire H, Luton J-P. L’Op’DDD possède-t-il une action antimitotique? Réflexions à propos de son emploi dans le traitement des adénocarcinomes surrénaliens. Nouv Presse Med 1977; 6:3650.
24. Lipsett MB, Hertz R, Ross GT. Clinical and pathophysiologic aspects of adrenocortical carcinoma. Am J Med 1963; 35:374-83.
25. Murphy BEP. Some studies of the protein-binding of steroids and their application to the routine micro and ultramicro measurement of various steroids in body fluids by competitive protein-binding radioassay. J Clin Endocrinol Metab 1967; 27:973-90.
26. Kaplan EL, Meier P. Nonparametric estimation from incomplete observa- tions. J Am Stat Assoc 1958; 53:457-81.
27. Tarone RE, Ware J. On distribution-free tests for equality of survival distri- butions. Biometrika 1977; 64:156-60.
28. Cox DR. Regression models and life-tables. J R Stat Soc [B] 1972; 34:187- 220.
29. Kalbfleisch JD, Prentice RL. The statistical analysis of failure time data. New York: John Wiley, 1980.
30. Anderson EE. Nonfunctioning tumors of the adrenal gland. Urol Clin North Am 1977; 4:263-71.
31. Ibanez ML. The pathology of adrenal cortical carcinoma: study of 22 cases. In: University of Texas, M.D. Anderson Hospital and Tumor Institute. Endocrine and non-endocrine hormone-producing tumors. Chicago: Year Book Medical, 1971:231-9.
32. Young JL Jr, Miller RW. Incidence of malignant tumors in U.S. children. J Pediatr 1975; 86:254-8.
33. Fisher DA, Panos TC, Melby JC. Therapy of adrenocortical cancer with o,p’DDD in two children. J Clin Endocrinol Metab 1963; 23:218- 21.
34. Korth-Schutz S, Levine LS, Roth JA, Saenger P, New MI. Virilizing adre- nal tumor in a child suppressed with dexamethasone for three years: effect of o,p’-DDD on serum and urinary androgens. J Clin Endocrinol Metab 1977; 44:433-9.
35. Lee PD, Winter RJ, Green OC. Virilizing adrenocortical tumors in child- hood: eight cases and a review of the literature. Pediatrics 1985; 76:437- 44.
36. Rappaport R, Schweisguth O, Cachin O, Pellerin D. Corticosurrenalome malin avec metastases: exerese et traitement par o,p’DDD suivi de guerison. Arch Fr Pediatr 1978; 35:551-4.
Vol. 322 No. 17 ADRENOCORTICAL CARCINOMA AND EFFECT OF MITOTANE THERAPY - LUTON ET AL. 1201
37. Gabrilove JL, Seman TA, Sabet R, Mitty HA, Nicolis GL. Virilizing adre- nal adenoma with studies on the steroid content of the adrenal venous efflu- ent and a review of the literature. Endocr Rev 1981; 2:462-70.
38. Levine GW. Adrenocortical carcinoma in two children with subsequent primary tumors. Am J Dis Child 1978; 132:238-40.
39. Lewinsky BS, Grigor KM, Symington T, Neville AM. The clinical and pathologic features of “non-hormonal” adrenocortical tumors: report of twenty new cases and review of the literature. Cancer 1974; 33:778-90.
40. Li FP, Tucker MA, Fraumeni JF Jr. Childhood cancer in sibs. J Pediatr 1976; 88:419-23.
41. King DR, Lack EE. Adrenal cortical carcinoma: a clinical and pathologic study of 49 cases. Cancer 1979; 44:239-44.
42. Harrison JH, Mahoney EM, Bennett AH. Proceedings: tumors of the adre- nal cortex. Cancer 1973; 32:1227-35.
43. Morimoto Y, Hiwada K, Nanahoshi M, et al. Cushing’s syndrome caused by malignant tumor in the scrotum: clinical, pathologic and biochemical studies. J Clin Endocrinol Metab 1971; 32:201-10.
44. Freeman DA. Steroid hormone-producing tumors in man. Endocr Rev 1986; 7:204-20.
45. Schteingart DE, Oberman HA, Friedman BA, Conn JW. Adrenal cortical neoplasms producing Cushing’s syndrome: a clinicopathologic study. Can- cer 1968; 22:1005-13.
46. Tang CK, Gray GF. Adrenocortical neoplasms: prognosis and morphology. Urology 1975; 5:691-5.
47. Lejonc J-L, Brochard P, Schaeffer A, Portos J-L. Les cortico-surrénalomes malins non sécrétants. Nouv Presse Med 1978; 7:3147.
48. Gabrilove JL, Sharma DC, Wotiz HH, Dorfman RI. Feminizing adrenocor- tical tumors in the male: a review of 52 cases including a case report. Medicine (Baltimore) 1965; 44:37-79.
49. Pugeat M, Merchak A, Revol A, Tourniaire J. Un cas de corticosurrénalome féminisant chez l’homme. Ann Endocrinol 1980; 41:107-18.
50. Arteaga E, Biglieri EG, Kater CE, Lopez JM, Schambelan M. Aldosterone- producing adrenocortical carcinoma: preoperative recognition and course in three cases. Ann Intern Med 1984; 101:316-21.
51. Cathelineau G, Poizat R, Legrand JC, Canivet J. Corticosurrénalome malin avec syndrome féminisant et hyperaldosteronisme primaire. Ann Endocrinol 1980; 41:65-6.
52. Foye LV, Feichtmeir TV. Adrenal cortical carcinoma producing solely min- eralocorticoid effect. Am J Med 1955; 19:966-75.
53. Greathouse DJ, McDermott MT, Kidd GS, Hofeldt FD. Pure primary hyperaldosteronism due to adrenal cortical carcinoma. Am J Med 1984; 76:1132-6.
54. Sakashita S, Kashiwagi A, Maru A, et al. Primary aldosteronism due to adrenal cortical carcinoma. J Urol 1984; 132:959-61.
55. Stone NN, Janoski A, Muakkassa W, Shpritz L. Mineralocorticoid excess secondary to adrenal cortical carcinoma. J Urol 1984; 132:962-5.
56. Cassan P, Baglin A, Coulbois J, Saigot T, Kuhn J-M, Betourne C. Cortico- surrénalomes malins non-sécrétants révélés par une fièvre prolongée: quatre cas. Nouv Presse Med 1978; 7:2153-6.
57. Bevan JS, Gough MH, Gillmer MD, Burke CW. Cushing’s syndrome in pregnancy: the timing of definitive treatment. Clin Endocrinol (Oxf) 1987; 27:225-33.
58. Koerten JM, Morales WJ, Washington SR III, Castaldo TW. Cushing’s syndrome in pregnancy: a case report and literature review. Am J Obstet Gynecol 1986; 154:626-8.
59. Benaily M, Schweisguth O, Job J-C. Les tumeurs cortico-surrénales de l’enfant: étude rétrospective de 34 cas observés de 1954 à 1973. Arch Fr Pediatr 1975; 32:441-53.
60. Page DL, De Lellis RA, Hough AJ. Tumors of the adrenal. Washington, D.C .: Armed Forces Institute of Pathology, 1985:115-49.
61. Karakousis CP, Uribe J, Moore J. Adrenal adenocarcinomas: diagnosis and management. J Surg Oncol 1981; 16:385-9.
62. Moolenaar AJ, van Slooten H, van Seters AP, Smeenk D. Blood levels of o,p’-DDD following administration in various vehicles after a single dose and during long-term treatment. Cancer Chemother Pharmacol 1981; 7:51- 4.
63. Johnson DH, Greco FA. Treatment of metastatic adrenal cortical carcinoma with cisplatin and etoposide (VP-16). Cancer 1986; 58:2198-202.
64. Schick M. Survival with adrenal carcinoma. JAMA 1973; 224:1763.
65. Tattersall MH, Lander H, Bain B, et al. Cis-platinum treatment of metastat- ic adrenal carcinoma. Med J Aust 1980; 1:419-21.
66. Molnar GD, Mattox VR, Bahn RC. Clinical and therapeutic observations in adrenal cancer: a report on 7 patients treated with o,p’-DDD. Cancer 1963; 16:259-68.
67. Hogan TF, Citrin DL, Johnson BM, Nakamura S, Davis TE, Borden EC. o,p’-DDD (mitotane) therapy of adrenal cortical carcinoma: observations on drug dosage, toxicity, and steroid replacement. Cancer 1978; 42:2177- 81.
68. Becker D, Schumacher OP. o,p’DDD therapy in invasive adrenocortical carcinoma. Ann Intern Med 1975; 82:677-9.
69. Boven E, Vermorken JB, van Slooten H, Pinedo HM. Complete response of metastasized adrenal cortical carcinoma with o,p’-DDD: case report and literature review. Cancer 1984; 53:26-9.
70. Cariou G, Richard F, Turpin G, Chatelain C, Kuss R. Corticosurrénalomes malins: 9 observations. Nouv Presse Med 1982; 11:2821-5.
71. Charbonnel B, Guillon J, Chupin M. Régression par OP’DDD d’un cortico- surrénalome malin et de ses métastases. Nouv Presse Med 1980; 9:239- 41.
72. Downing V, Eule J, Huseby RA. Regression of an adrenal cortical carcino- ma and its neovascular bed following mitotane therapy: a case report. Can- cer 1974; 34:1882-7.
73. Jarabak J, Rice K. Metastatic adrenal cortical carcinoma: prolonged regres- sion with mitotane therapy. JAMA 1981; 246:1706-7.
74. Levy JM, Lutz P, Wagner C, et al. Evolution favorable sous o,p’-DDD d’un corticosurrénalome malin récidivant. Ann Pediatr (Paris) 1985; 32:541-4.
75. du Rostu H, Krempf M, Mussini JM, Murat A, Guillon J, Charbonnel B. Neurotoxicité du traitement par le mitotane des cortico-surrénalomes (5 cas) et des syndromes de Cushing (7 cas). Presse Med 1987; 16:951-4.
76. Gutierrez ML, Crooke ST. Mitotane (o,p’-DDD). Cancer Treat Rev 1980; 7:49-55.
77. May CA, Garnett WR. Treatment of adrenocortical carcinoma: a case report and review of the literature. Drug Intell Clin Pharm 1986; 20:24-32.
78. Luton JP, Mahoudeau JA, Bouchard P, et al. Treatment of Cushing’s dis- ease by O,P’DDD: survey of 62 cases. N Engl J Med 1979; 300:459-64.