Hypoaldosteronism Accompanied by Normal or Elevated Mineralocorticosteroid Pathway Steroid: A Marker of Adrenal Carcinoma*
B. AUPETIT-FAISANT, C. BATTAGLIA, M. ZENATTI, N. EMERIC-BLANCHOUIN, AND J. C. LEGRAND
Service de Biochimie (B.A-F., M.Z., N.E-B., J.C.L.), CHU Pitié-Salpétrière, Paris, Cedex 13, France; and Medecine 8 Hôpital Broussais (C.B.), Paris, Cedex 14, France
ABSTRACT
In order to find a biochemical marker to assist the physician in the difficult differential diagnosis between malignant and nonmalignant adrenal tumors, plasma levels of the mineralocorticosteroids (deoxy- corticosterone, 18-hydroxydeoxycorticosterone, corticosterone, 18-hy- droxycorticosterone, and aldosterone) were determined. The same method (RIA which is preceded by a crucial separation step) was used to measure all these steroids including aldosterone. The subjects in- cluded 15 adults presenting various clinical signs of adrenocortical tumors (histopathologically: 6 with adrenal carcinoma, 1 with a history of adrenal carcinoma, 1 with adrenal metastasis from other forms of cancer, 6 with adenoma, and 1 with hyperplasia). The results show that
both presurgery and during a recurrence of adrenal carcinoma, hypoal- dosteronism occurs which contrasts with the normal or even elevated levels of some aldosterone precursors. In the 7 cases of adrenal cortical carcinoma, this dysfunction of the aldosterone pathway was detected regardless of the impairment of the other steroidogenesis pathways, whereas it was never found with a nonmalignant tumor. Despite the limited number of cases so far available, these findings suggest that detection of abnormalities of the aldosterone pathway, and particularly the detection of hypoaldosteronism by an assay method involving a crucial steroid separating step, could contribute to a differential diag- nosis between benign and malignant adrenocortical tumor and between adrenal metastasis and other forms of cancer. (J Clin Endocrinol Metab 76: 38-43, 1993)
S OME ADRENAL masses induce clinical and biological signs (hyperaldosteronism, Cushing’s syndrome, virili- zation) whereas others, whether detected fortuitously or otherwise, are asymptomatic: incidentalomas, but also car- cinomas. Some malignant masses [31% according to Luton et al. (1) and 28% according to Cohn et al. (2)] are described as “nonfunctional” since they may develop over an extended period (3) without causing any recognizable clinical or bio- logical signs until they reach the stage of being incurable. Adrenal carcinoma is thought to be very rare (1-4), but its exact incidence in the general population is still not known (5).
Adrenal masses were formerly unusual and were diag- nosed only when they caused symptoms (6). The advent of computed tomographic techniques is changing this situation (5). The management of adrenal masses now poses new dilemmas for clinicians, since it must be remembered that in the case of adrenal carcinoma, early resection provides the only chance of cure (2, 7, 8).
Particularly when faced with a fortuitously detected mass, the physician has to answer two questions: is it malignant, and is it currently hormonally active (6)?
The aim of this study was to show that for various histo-
Address all correspondence and requests for reprints to: B. Aupetit- Faisant, Service de Biochimie, CHU Pitié-Salpétrière, 91 Bd de l’Hopital 75634, Paris, Cedex 13, France.
* This study was supported by grants from the ministère de l’Educa- tion Nationale and from the Assistance Publique (Contrat de recherche clinique de l’Assistance Publique No. 126).
pathological types of adrenal masses the findings of extensive hormonal exploration of mineralocorticoid pathway corre- lated with histopathological findings.
The 15 patients explored included 6 with an adrenal carcinoma, 1 with a history of an adrenal carcinoma, 1 with an adrenal metastasis from other forms of cancer, 6 with an adenoma, and 1 with an hyperplasia. Plasma levels of ste- roids in the mineralocorticosteroid pathway were deter- mined: deoxycorticosterone (DOC), 18-hydroxydeoxycorti- costerone (18-OH DOC), corticosterone (B), 18-hydroxycor- ticosterone (18-OH B) and aldosterone.
For the sake of convenience, all these steroids will be referred to below as “aldosterone precursors” even though aldosterone is, in fact, derived only from the steroids formed in the zona glomerulosa and they are not all recognized by all authors as true aldosterone precursors (18-OH B and 18- OH DOC) (9-14). All these steroids do belong to aldosterone synthesis pathway and for convenience be referred to below as “aldosterone precursors.”
Subjects and Methods
Subjects
Control subjects. The normal values for our laboratory were determined in 30 adult subjects (18 women and 12 men between 22 and 63 yr of age with a mean of 37 yr) who were on a normal sodium diet, were normotensive and normokaliemic, and were not receiving any treatment likely to have any impact on the renin-aldosterone axis. Blood samples were taken at 0800 h successively when subjects were either supine or in upright position, after being up for 2 h. The normal range after ACTH was determined in 5 adult subjects in the same conditions.
Patients. All the patients were adults. Most of the patients were women, which was probably due to chance and related to the retrospective nature of this study, which made no attempt to investigate the sexual prevalence of these disorders in a group of 15 patients, but was intended to provide information about the disruption of the mineralocorticoid pathway which could be used to distinguish between malignant and nonmalignant masses. The laboratory tests reported here had been requested following the detection of an adrenal mass and before surgery, except for patients 13 and 14. Blood samples were taken at 0800 h in the supine position unless otherwise indicated (see Tables 5 and 7), from subjects on a normo-sodium diet (sodium-free + 6 g NaCl), before any hypokaliemia present had been corrected and after all treatment likely to have an impact on the renin-aldosterone axis had been stopped. Histopathological examinations were carried out after tumor exeresis or biopsy (patient 1). The general data for the patients are shown in Table 1. The samples from patients 13 and 14 were taken after exeresis of an adrenal carcinoma. Patient 13 was a case of contralateral recurrence of an adrenal carcinoma, who had previously undergone surgery and received to complementary medical treatment (Op’DDD). Patient 14 had an adrenal carcinoma and undergone surgery 4 yr previously (hypertension, mass measuring 6 cm detected by computerized axial tomography scan before surgery) and had since responded favorably, without medical treatment. He had shown no sign of local recurrence or metastasis and his hypertension had disappeared at the time of the biological exploration reported here. Patient 15 had shown no clinical signs of Cushing’s syndrome or virilization. He had an adrenal mass, the histopathology of which showed it to be a metastasis of a cancer of the colon for which he had undergone surgery 1 yr earlier.
All the patients with a histopathologically identified nonmalignant adrenal mass also presented with associated hypertension. This hyper- tension was sometimes of recent origin (patients 2, 3, 4, 7) and sometimes chronic (patients 1, 5, 6). In this group, patients 3-7 inclusive presented with primary aldosteronism. Patients 3, 4, and 5 showed the drastically reduced renin level characteristic of primary hyperaldosteronism. In
| Patient | Sex/age | Clinical diagnosis | Histological diagnosis and tumor size |
|---|---|---|---|
| 1 | F/66 | Long-standing HT, moderate | Unoperated mass 15 mm in diameter on CT scan |
| 2 | F/59 | Recent HT, moderate | Adenoma 20 mm in diame- ter on CT Scan |
| 3 | F/33 | Recent HT, severe | Adenoma 2 × 1.5 × 1.5 cm |
| 4 | F/42 | Recent HT, moderate | Adenoma 4 × 4.5 × 3.4 cm |
| 5 | M/49 | Long-standing HT, severe | Adenoma 8 x 4 x 2 cm |
| 6 | F/70 | Long-standing HT, severe | Hyperplasia |
| 7 | F/37 | Recent HT | Adenoma 1 × 1 × 1 (cm) |
| 8 | F/62 | Long-standing, severe HT, recent C | Carcinoma 13 × 11 x 8 cm |
| 9 | F/54 | C | Carcinoma 3 cm in diame- ter on CT Scan |
| 10 | F/26 | HT, H, and C, recent | Carcinoma 2 cm in diame- ter on CT Scan |
| 11 | F/42 | HT (long-standing, moderate) | Carcinoma 10 cm in diam- eter on CT Scan |
| 12 | F/26 | HT (recent, moder- ate) | Carcinoma 6 × 4.5 × 3.5 cm |
| 13 | F/43 | HT (long-standing, moderate) H and C | R of a carcinoma |
| 14ª | F/60 | PO of a carcinoma | |
| 15 | F/66 | AP and F | AM |
HT, Hypertension; C, Cushing’s syndrome; AP, abdominal pain; F, fever; H, hirsutism; PO, postoperative follow-up; R, postoperative recurrence of carcinoma; AM, adrenal metastasis of another cancer.
” Patient 14, who was cured at the time of the paraclinical assessment reported here, at the time the carcinoma was diagnosed only had a blood pressure value corresponding to an adrenal carcinoma measuring 6 cm on the computerized tomography scan.
patients 6 and 7, the renin levels were at the lower limit of the normal range or normal and the diagnosis of primary hyperaldosteronism was based on the absence of the stimulation of renin production by orthos- tatism, combined with high values of plasma aldosterone and hypoka- liemia.
Six out of the seven carcinoma cases also presented with hyperten- sion, which was usually chronic (if we include the case of patient 14 who had presented isolated hypertension when diagnosed), whereas the other clinical signs had developed recently.
Hormone assay method
Plasma levels of DOC, 18-OH DOC, B, 18-OH B, and aldosterone were determined in duplicate by RIA method using rabbit polyclonal antibodies. The cross-reactions are shown in Table 2. One milliliter of plasma was taken for each assay to which was added a known quantity of each tritiated steroid in order to calculate recovery. The sample was extracted with 15 mL of dichloromethane containing 0.1% triethylamine for assay of the 18-hydroxylated compounds. Each steroid was separated by paper chromatography (Whatman No. 2, previously washed with methanol during distillation in a Soxlett apparatus) in a system adapted to the performance of the antibody which was used for the assay of a given steroid (the characteristics of the chromatography system used for each steroid are shown in Table 3). The recoveries after extraction and chromatography ranged from 40-60%, and the results were corrected for recovery. After elution, the quantity of steroid was determined by RIA in a 0.25 mol/L (pH 7) phosphate buffer containing sodium azide (2 g/L) and gelatin (1 g/L). As usual, the accuracy of each series of assays was checked by adding various known quantities of cold steroid to steroid-free plasma, and in any series in which the highest control deviated by 20% or more from the expected value were not taken into consideration. The intraassay and interassay coefficients of variation are reported in Table 4. The lowest detectable concentration of aldosterone in 1 mL plasma was 0.5 ng/100 mL. Antibodies with similar specificities to that reported in Table 2 are commercially available.
Results
Table 5 shows that in patients bearing a nonmalignant corticoadrenaloma, the nonmalignancy of which had been demonstrated by histological findings, aldosterone levels were proportional to precursor levels, regardless of the actual plasma aldosterone levels found: normal (patient 2), upper limit of normal (patients 1, 3, and 4) or above normal (patients 5, 6, and 7). In contrast, in all the cases of adrenal carcinoma or recurrence of carcinoma (patients 8-13) plasma aldosterone levels were below the threshold of detection. This contrasted with the normal or increased plasma levels of the precursors. In patients 8 and 9, an undetectable urinary level of aldosterone excretion value confirmed the very low plasma aldosterone level. The data shown in Tables 1 and 5 demonstrate that there was a link between the malignancy and dysfunction of the aldosterone biosynthesis pathway, regardless of the size of the mass.
Tables 5 and 6 show that in the cases of adrenal carcinoma in this group, the steroid profile of the aldosterone synthesis pathway mirrored the clinical development; it returned to normal when the patient was cured (patient 14, Table 5), but persisted or was even aggravated in cases of recurrence or extension (patient 13, Table 5 and patient 8, Table 7) even if treatment with Op’DDD was administered (patient 13). In adrenal carcinoma, aldosterone returned to normal several months after surgery and in patient 8 (Table 6) the plasma aldosterone level was once more undetectable when hepatic metastasis appeared shortly before death (D0 + 15 months).
| Cross-reacting steroid | Deoxycorticosterone | 18-Hydroxydeoxycorticosterone | Corticosterone | 18-Hydroxycorticosterone | Aldosterone |
|---|---|---|---|---|---|
| Deoxycorticosterone | 100.0 | 0.08 | 31.4 | <0.1 | 0.000002 |
| 18-Hydroxydeoxycorticosterone | 1.12 | 100 | 0.01 | 0.5 | 0.000084 |
| Corticosterone | 0.5 | 0.02 | 100 | <0.01 | 0.073 |
| 18-Hydroxycorticosterone | <0.056 | <0.08 | 0.08 | 100 | 0.000017 |
| Aldosterone | <0.056 | <0.08 | 0.02 | 2.92 | 100 |
| Progesterone | 14.36 | 0.18 | 1.90 | <0.01 | <0.0058 |
| 11-Deoxycortisol | 20.7 | <0.008 | 0.039 | <0.01 | 0.0000007 |
| Cortisone | 0.8 | 0.003 | <0.52 | <0.018 | <0.03 |
| Cortisol | 0.12 | <0.001 | 3.29 | <0.001 | 0.02 |
| 17-Hydroxyprogesterone | <0.002 | <0.038 | 0.06 | <0.0013 | 0.0000001 |
| Testosterone | <0.0007 | 0.025 | 0.04 | <0.0036 | 0.000002 |
| Dexametasona | 0.05 | <0.001 | 0.047 | <0.0011 | 0.0000001 |
| Prednisone | 0.0003 | <0.013 | <0.21 | <0.0011 | <0.001 |
| Prednisolone | 0.003 | <0.0095 | 0.39 | <0.0011 | <0.02 |
| Androstenedione | 0.6 | 0.034 | 0.01 | <0.0013 | 0.1 |
| 9a-Fludrocortisone | <0.005 | <0.001 | <0.003 | <0.0011 | <0.0008 |
| Pregnenolone | 0.0001 | <0.0064 | <0.0028 | <0.0011 | <0.00086 |
| Metyrapone | <0.008 | <0.0016 | <0.0028 | <0.0011 | <0.000001 |
| 21-Deoxycortisol | 0.002 | <0.004 | <0.002 | <0.007 | 0.000002 |
| Purification | RTª | ||||
|---|---|---|---|---|---|
| Deoxycorticosterone | Corticosterone | 18-Hydroxycorticosterone | 18-Hydroxydeoxycorticosterone | Aldosterone | |
| Deoxycorticosterone | 1 | 2.6 | 20 | 3.3 | 8 |
| Corticosterone | 0 | 1 | 7.6 | 0.78 | 2.6 |
| 18-Hydroxycorticosterone | 0 | 0.13 | 1 | 0.16 | 0.3 |
| Aldosterone | 0 | 0.37 | 3 | 0.46 | 1 |
| 18-Hydroxydeoxycorticosterone | 0 | 0.78 | 6 | 1 | 2.1 |
| Progesterone | >40 | 3 | 23 | 3.83 | 11.3 |
| 11-Deoxycortisol | 0.04 | 1.04 | 8 | 1.33 | 2.6 |
| Cortisone | 0 | 0.39 | 3 | 0.5 | 1.16 |
| Cortisol | 0 | 0.17 | 1.3 | 0.22 | 0.5 |
| 17-Hydroxyprogesterone | 0.78 | 2.3 | 17.6 | 2.94 | 8 |
| Testosterone | 0.95 | 2.26 | 17.3 | 2.8 | 6.1 |
| Dexametasona | 0 | 0.21 | 1.6 | 0.27 | 0.5 |
| Prednisone | 0 | 0.34 | 2.6 | 0.4 | 1 |
| Prednisolone | 0 | 0.17 | 1.33 | 0.2 | 0.3 |
| 9a-Fludrocortisone | 0 | 1.04 | 8 | 1.33 | 2.6 |
| Composition | Methyl ciclo hexane methanol eau 10/8-2 | Benzene heptane methanol eau 7-3/8-2 | Benzene heptane methanol eau 7-3/8-2 | Benzene heptane methanol eau 7-3/8-2 | Benzene heptane methanol eau 7-3/8-2 |
” Ratio of the distance covered by a compound to be identified to that of a reference compound.
| Steroid | DOC | 18-OH DOC | B | 18-OH B | Aldosterone | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Concentration in | nmol/L 0.37 | 0.75 | 1.5 | 3 | 0.14 | 0.72 | 1.45 | 14.5 | 29 | 58 | 0.67 | 1.35 | 2.7 | 0.14 | 0.7 | 1.4 |
| (1) | 6 | 7 | 5 | 5 | 25 | 22 | 17 | 8 | 6 | 5 | 18 | 15 | 12 | 8 | 6 | 5 |
| (2) | 13 | 12 | 8 | 6 | 15 | 12 | 8 | 10 | 11 | 12 | 13 | 8 | 10 | 14 | 15 | 12 |
Neither the steroid profile before surgery, nor the postoper- ative outcome typical of the adrenal carcinoma was found in metastasis of another carcinoma type (patient 15). Metastatic invasion of the adrenal results in an abolition of the entire aldosterone pathway.
Table 7 shows that a plasma level of aldosterone below the threshold of detection, which is characteristic of malig- nant corticoadrenaloma, cannot be reliably identified if the
assay of aldosterone was not preceded by a separation step. The accumulation of precursors or their derivatives probably accounts for the different findings of these two methods of assay. The separation procedure probably eliminated not only the steroids which are known to produce cross-reactions with a given antibody, but also other compounds which are not shown in Table 2. Even with an antibody which does not show any cross-reactions with the steroids included in
| Normal values (nmol/L) | DOC 0.12-0.54 | 18-OH DOC 0.09-0.52 | B 23-52 | 18-OH B 1.08-2.16 | Aldo 0.08-0.28 | B/DOC |
|---|---|---|---|---|---|---|
| Patient | ||||||
| 1 | 0.34 | 0.49 | 47 | 1.98 | 0.30 | 138.2 |
| 2 | 0.26 | 0.25 | 20 | 0.78 | 0.18 | 76.9 |
| 3 | 0.43 | 0.18 | 20 | 1.97 | 0.37 | 46.5 |
| 4 | 2.25 | 0.73 | 43 | 2.45 | 0.24 | 19.1 |
| 5 | 15.21 | 3.48 | 149 | 9.61 | 2.05 | 9.8 |
| 6 | 0.67 | 1.9 | 68 | 5.86 | 1.02 | 101.5 |
| 7 | 0.84 | 0.81 | 65 | 6.97 | 1.96 | 77.4 |
| 8 | 0.63 | 0.14 | 34 | 0.91 | NDª | 54 |
| 9 | 1.8 | 0.36 | 68 | 2.69 | ND | 37.8 |
| 10 | 2.89 | 0.11 | 28 | 0.75 | ND | 9.7 |
| 11 | 3.96 | 1.88 | 46 | 2.89 | ND | 11.6 |
| 12 | 7.17 | 0.12 | 37 | 0.34 | ND | 5.2 |
| 13 | 15 | 128 | 1.62 | ND | 8.5 | |
| 14 | 0.78 | 0.9 | 48 | 1.51 | 0.34 | 61.5 |
| 15 | 0.24 | 0.03 | 10 | 0.27 | ND | 41.7 |
Blood samples were taken at 0800 h in the supine position except for patients 4, 13, and 14 (recumbent position at 0800 h). For other conditions see “Subjects and Methods.”
” Not detected.
Table 2, the steroid levels of aldosterone for example, deter- mined in the same laboratory with and without chromato- graphic separation did not give the same results either for adenoma or for carcinoma (from 25-70% higher). It can therefore be concluded that apart from possible hypotheses which could be advanced concerning the steroids eliminated during separation, the results show that this is due to so-far unidentified steroids which differ from those listed in Table 2, but which can be eliminated using the aldosterone purifi- cation method described here. The conventional wisdom that adrenal carcinomas produce a wide range of steroids, opens a wide field for future exploration of the identity of these compounds.
All the patients presented frank hypokaliemia or a hypo- kaliemic tendency, but in adrenal carcinoma cases this was consistent with neither the hypoaldosteronism nor with the renin values. The weak mineralocorticoid activity of aldoste- rone precursors as defined above cannot account entirely for this hypokaliemia. Concerning renin or PRA level in adrenal carcinoma with high level of cortisol, it would be interesting to explore, in a further study, the relationship between renin, cortisol and stimulation of synthesis of hepatic angiotensi- nogen (15, 16).
Table 8 shows the response of adrenal mineralocorticos- teroid function to changes in posture and to Synacthen in a malignant and a nonmalignant mass; it can be seen that hypoaldosteronism in adrenal carcinoma remains obvious regardless of posture or Synacthen.
Discussion
A theoretical distinction is drawn in adrenal carcinomas between “functional” and “nonfunctional” masses (4, 6). The vast majority of functional adrenal carcinomas result from abnormalities of the cortisol or adrenal androgen pathways, whereas very few cases accompanied by hyperaldosteronism have been reported in the literature (17-20). Isolated cases of adrenal carcinoma presenting hypersecretion of some precursors of aldosterone (21-24) have been reported, sug- gesting the concept of a DOC tumor or an 18-OH DOC tumor, associated with hypoaldosteronism for the three pa- tients reported by Biglieri et al. (24).
Our results show that for all carcinoma cases in this study, there was a dysfunction of the aldosterone pathway with an undetectable aldosterone level accompanied by normal or elevated levels of various aldosterone precursors. This was found in all cases of adrenal carcinoma regardless of the clinical signs (Cushing’s syndrome, virilization, hyperten- sion) or of the size of the mass and the treatment received. This metabolic dysfunction of aldosterone pathway was never seen in nonmalignant masses or in a case of cured carcinoma (patient 14). Moreover, the dysfunction of the aldosterone pathway mirrors the development of the tumor (Table 6). This metabolic dysfunction in adrenal carcinoma could be explained either by an acquired deficiency or by an inhibition. The hypothesis of an 11-hydroxylase deficiency in the aldosterone synthesis pathway, analogous to that demonstrated by Lipsett et al. (25, 26) in the cortisol pathway, has been suggested by Doerr et al. (21) for adrenal carcinoma.
However, the following facts suggest that our observations do not correspond to a blockade analogous to congenital enzyme deficiency at 118 or type II carboxy methyl oxidase level. The accumulation of DOC is not associated with the disappearance of the subsequent steroids. The hypoaldoster- onism which occurred in all the cases of carcinoma, was not associated with an increase in either 18-OH B or B. Although it was not relieved by a change of posture or by synacthen
| Normal values (nmol/L) | Patient 8 (adrenal carcinoma) | Patient 11 (adrenal carcinoma) | Patient 15 (metastasis of colonic cancer) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Before surgery Do | Do + 1 month | Do + 8 months | Do + 15 months | Before surgery Do | Do + 5 months | Do + 8 months | Before surgery Do | Do + 13 days | ||
| DOC | 0.12-0.54 | 0.63 | 0.08 | 0.88 | 1.08 | 3.96 | 1.31 | 0.43 | 0.24 | 0.34 |
| 18-OH DOC | 0.09-0.52 | 0.14 | 0.05 | 0.03 | 1.88 | 0.27 | 0.53 | 0.03 | 0.16 | |
| B | 23-52 | 34 | 12 | 65 | 46 | 42 | 67 | 10 | 40 | |
| 18-OH B | 1.08-2.16 | 0.91 | 0.10 | 0.51 | 2.89 | 1.66 | 0.27 | 0.86 | ||
| Aldosterone | 0.08-0.28 | NDª | ND | 0.23 | ND | ND | ND | 0.32 | ND | 0.15 |
Blood samples were taken at 0800 h in supine position. For other conditions see “Subjects and Methods.” ” Not detected.
| Patient | Potassium (mmol/L) | Active Renin (ng/L) N = 10-35 (L) N = 15-50 (S) | PRA (ng. L-1.s-1) N1 = 0.08-0.41 (L) N2 = 0.16-0.6 (L) | Plasma aldosterone determined by RIA (nmol/L) | |||
|---|---|---|---|---|---|---|---|
| Direct | After separation | ||||||
| Normal values | Patients | Normal values | Patients | ||||
| 1 | 4 | (L) 17 | - | 0.08-0.40 | 0.59 | 0.08-0.28 (L) | 0.30 |
| 2 | 3.2 | - | (L) 0.1 (N1) | 0.14-0.28 | 0.36 | 0.08-0.28 (L) | 0.18 |
| 3 | 2.7 | (L) 5.6 | - | 0.06-0.21 | 0.87 | 0.08-0.28 (L) | 0.37 |
| 4 | 3 | (S) 13 | - | 0.21-1 | 0.33 | 0.18-0.64 (S) | 0.24 |
| 5 | 2.9 | (L) 5 | - | 0.08-0.40 | 1 | 0.08-0.28 (L) | 2.05 |
| 6 | 2.9 | (L) 10 | - | 0.08-0.40 | 0.2 | 0.08-0.28 (L) | 1.02 |
| 7 | 3.3 | (L) 21 | - | 0.08-0.40 | 2.39 | 0.08-0.28 (L) | 1.96 |
| 8 | 3.1 | - | - | - | - | 0.08-0.28 (L) | ND |
| 9 | 3.3 | - | (L) 0.31 (N2) | 0.07-0.5 | 0.25 | 0.08-0.28 (L) | ND |
| 10 | 3.6 | (L) 21) | - | 0.04-0.4 | 0.17 | 0.08-0.28 (L) | ND |
| 11 | 3.4 | (L) 8 | - | 0.2-0.7 | 0.9 | 0.08-0.28 (L) | ND |
| 12 | 3.5 | - | (L) ND (N1) | 0.14-0.28 | ND | 0.08-0.28 (L) | ND |
| 13 | 3.7 | (S) 17 | - | 0.17-0.19 | <0.03 | 0.18-0.64 (S) | ND |
| 14 | 4.8 | (S) 59 | - | 0.21-1 | 0.63 | 0.18-0.64 (S) | 0.34 |
| 15 | 4 | - | (L) 0.55 (N1) | 0.14-0.36 | 0.36 | 0.08-0.28 (L) | ND |
The patients were explored in different hospital centers and so the values reported are those for the center where the patient was hospitalized for blood levels of potassium, active renin, or PRA (N1, N2 correspond to various hospital) and aldosterone when the determination is without chromatographic separation. Normal values of cortisol at 0800 h and testosterone for female at various hospital centers were the same (cortisol: 276-552 nmol/L, testosteone: 0.7-2 nmol/L). Cortisol values for patients 8, 10, 12, 13, and 15 were, respectively: 966, 870, 612, 1735, and 340 nmol/L. Testosterone values for patients 10, 13, and 15 were, respectively: 4.2, 8.1, and 1.5 nmol/L. Result of aldosterone with chromatographic separation was performed in the laboratory of the first author.
ª -, results not available; L, supine; S, standing; ND, not detected.
| DOC | 18-OH DOC | B | 18-OH B | Aldosterone | ||
|---|---|---|---|---|---|---|
| Basal normal values (nmol/L) | ||||||
| Supine | 0.12-0.54 | 0.09-0.52 | 23-52 | 1.08-2.16 | 0.08-0.28 | |
| Recumbent | 0,15-0,70 | 0,10-0,61 | 30,8-69,6 | 1,44-4,60 | 0,18-0,64 | |
| After synacthen | 0,52-2,37 | 3,3-19 | 358-811 | 3,72-7,34 | 0,24-0,84 | |
| Synacthen test | ||||||
| (NM) patient 7 | TOª | 0.84 | 0.81 | 65 | 6.97 | 1.96 |
| T1 | 3.09 | 3.68 | 179 | 12.06 | 3 | |
| (M) patient 12 | TO | 7.17 | 0.12 | 37 | 0.34 | ND |
| T1 | 7.17 | 0.52 | 41 | 0.62 | ND | |
| Effect of posture | ||||||
| (NM) patient 2 | L | 0.26 | 0.25 | 20 | 0.78 | 0.18 |
| S | 0.37 | 0.32 | 27 | 1.67 | 0.48 | |
| (M) patient 10 | L | 2.89 | 0.11 | 28 | 0.75 | ND |
| S | 3.31 | 0.13 | 28.1 | 0.96 | ND |
Blood samples were taken at 0800 h in the supine position and at 1200 h in the standing position.
” Abbreviations: TO, before the synacthen test; T1, 1 h after synacthen injection; L, supine; S, standing.
(Table 8), the hypoaldosteronism cannot be explained as the result of an enzyme deficiency limited to the tumor (aldoste- rone would be secreted by the contralateral adrenal) nor in all cases by an early bilateral biochemical impact preceding the development of the tumor process (the rise in aldosterone levels following ablation of the tumor could not then be explained, see Table 6).
The contralateral adrenal also appears to be incapable of synthesizing aldosterone before surgery, whereas it becomes active a considerable time after surgery. The post-operative recovery of the ability to synthesize aldosterone by the healthy adrenal suggests that before surgery a substance produced by the tumor reversibly inhibited aldosterone syn- thesis. The possibility of the reversibility of the global hy- poaldosteronism rules out the hypothesis of a unilateral or bilateral enzyme deficiency. So the term of “inhibition of the
aldosterone synthesis pathway” would be more appropriate to describe the findings reported here than the term “enzyme deficiency” in its conventional sense. This inhibition of al- dosterone production affected both adrenals before exeresis of the tumor, and was reversed several months after ablation. This reversibility took longer for an adrenal carcinoma than for a mass located within the adrenal but consisting of non- steroid secreting tissue. This means that for adrenal carci- noma (well known to produce a wide range of steroids) the inhibitory substance could be a corticosteroid produced by the tumor and which inhibits adrenal function in the same way as exogenous corticosteroids. Tumor development leads to a disorganization of the normal zonation of the adrenal. In vitro cortisol and testosterone inhibit aldosterone biosyn- thesis (27) and so, in vivo, they could inhibit biosynthesis which normally occurs in the zona glomerulosa. The conver-
sion of 18-OH B into aldosterone is most affected (27) since it is the most vulnerable reaction in this metabolic pathway (9-13). This inhibition of aldosterone biosynthesis pathway by cortisol or testosterone could account for the hypoaldos- teronism of adrenal carcinoma due to inhibition of one or other hydroxylase (110-hydroxylase or CMO I or CMO II), but not the hypokaliemia. The clinical situation of low al- dosterone hypermineralocorticosteroidism in adrenal carci- noma cannot be explained only by accumulation of aldoste- rone precursors because they have low mineralocorticoster- oid activity, but it may be due to derivatives of aldosterone precursors such as 19 Nor-DOC. Such compounds could have sufficient mineralocorticoid activity (28) to explain the lack of clinical expression of hypoaldosteronism in adrenal carcinoma patients and may not be regulated by the renin- angiotensin system. As a result of the progress in medical imaging, the percentage of adrenal masses detected in life is approaching that found postmortem (29, 30) and differential diagnosis is increasingly important. Our results show that, in agreement with Lejonc et al. (31) and Mc Kenna et al. (32) a corticosteroid alone cannot constitute a sign of malignancy. But extensive exploration of the entire aldosterone synthesis pathway using a single method involving a chromatographic steroid separating step, which is particularly crucial in de- tecting hypoaldosteronism, should assist the clinician in the difficult differential diagnosis.
Acknowledgments
We would like to express our gratitude to Professor Vecsei who kindly supplied us with several antibodies and Doctor Fraser who allowed B. A. F. to spend time in his laboratory to perfect the technique and his long-suffering team. We would like to thank Professors Corvol (Paris), Reville, Pinget, and Dr. Ortega (Strasbourg), and Schaison and Warnet (Paris) for their clinical data.
References
1. Luton JP, Cerdas S, Billaud L et al. 1990 Clinical features of adrenocortical carcinoma, prognostic factors, and the effect of mi- totane therapy. N Engl J Med. 322:1195-201.
2. Cohn K, Gottesman L, Brennan MF. 1986 Adrenocortical carci- noma. Surgery. 100:1170-1177.
3. Grundfeld Avellàn JA, Baltasar A. 1986 Non-functionning adrenal carcinoma. Eur Urol. 12:211-14.
4. Brennan MF. 1987 Adrenocortical carcinoma. CA. 37:348-365.
5. Hubbart MM, Husami TW, Abumrad NN. 1989 Nonfunctioning Adrenal Tumors. Am Surg. 55:516-522.
6. Penn I, Moulton J, Bracken B. 1988 Diagnosis and management of adrenal masses: 1987 Du Pont Lecture. Can J Surg. 31:105-109.
7. Nader S, Hickey RC, Selin RV, et al. 1983 Adrenal Cortical carcinoma A study of 77 cases. Cancer. 52:707-711.
8. Plager JE. 1984 Carcinoma of the adrenal cortex: Clinical descrip- tion, diagnosis and treatment. Int Adv Surg Oncol. 7:329-353.
9. Vecsei P, Lonner P, Wolff HP. 1968 The intermediate role of 18- hydroxycorticosterone in aldosterone biosynthesis. Experientia. 24:1199-201.
10. Sandor T, Fazekas AG, Lehoux GJ, Leblanc H, Lanthier A. 1972 Studies on the biosynthesis of 18-oxygenated steroids from exoge- nous corticosterone by domestic duck adrenal gland mitochondria.
J Steroid Biochem. 3:661-682.
11. Ulick S. 1976 Diagnosis and nomenclature of the disorders of the terminal portion of the aldosterone biosynthetic pathway. J Clin Endocrinol. 43:92-96.
12. Fraser R, Lantos CP. 1978 18-hydroxycorticosterone: A review. J Steroid Biochem. 9:273-286.
13. Aupetit B, Accarie C, Emeric N, Vonarx V, Legrand JC. 1983 The final step of aldosterone biosynthesis requires reducing power: It is not a dehydrogenation. Biochim Biophys Acta. 752:73-8.
14. Grekin RJ, Dale SL, Melby JC. 1973 The role of 18-hydroxydeox- ycorticosterone as a precursor in human adrenal tissue in vitro. J Clin Endocrinol Metab. 37:61-4.
15. Krakoff LW. 1973 Measurement of plasma renin substrate by radioimmunoassay of angiotensin I: Concentration in syndromes associated with steroid excess. J Clin Endocrinol Metab. 37:110-117.
16. Clauser E, Gaillard I, Wei L, Corvol P. 1989 Regulation of angi- otensinogen gene. Am J Hypertens. 2:403-410.
17. Foye LV, Feichtmeir TV. 1955 Adrenal cortical carcinoma produc- ing solely mineralocorticoid effect. Am J Med. 19:966-975.
18. Arteaga E, Biglieri EG, Kater CE, Lopez M, Schambelan M. 1984 Aldosterone producing adrenocortical carcinoma: Preoperative rec- ognition and course in three cases. Ann Intern Med. 101:316-321.
19. Greathouse DJ, Mc Dermott MT, Kidd GS, Hofeldt FD. 1984 Pure primary hyperaldosterism due to adrenal cortical carcinoma. Am J Med. 76:1132-136.
20. Farge D, Chatellier G, Pagny JY, Jeunemaitre X, Plouin PF, Corvol P. 1987 Isolated clinical syndrome of primary aldosteronism in four patients with adrenocortical carcinoma. Am J Med. 83:635- 640.
21. Doerr HG, Sippell WG, Drop SLS, Bidlingmaier F, Knorr D. 1987 Evidence of 11 ß hydroxylase deficiency in childhood adrenocortical tumors. Cancer. 60:1625-629.
22. Isles CG, Mac Dougall IC, Lever F, Fraser R. 1987 Hyperminer- alocorticoidism due to adrenal carcinoma: plasma corticosteroids and their response to ACTH and angiotensin II. Clin Endocrinol. 26:239-251.
23. Ishikawa SE, Saito T, Kaneko K, Okada K, Fukuda S, Kuzuya T. 1988 Hypermineralocorticism without elevation of plasma aldoste- rone: Deoxycorticosterone-producing adrenal adenoma and hyper- plasia. Clin Endocrinol (Oxf). 29:367-375.
24. Biglieri EG, Irony I, Kater CE. 1989 Identification and implications of new types of mineralocorticoid hypertension. J Steroid Biochem. 32:199-204.
25. Lipsett MB, Wilson H. 1962 Adrenocortical cancer: steroid biosyn- thesis and metabolism evaluated by urinary metabolites. J Clin Endocrinol Metab. 22:906-915.
26. Lipsett MB, Hertz R, Ross G. 1963 Clinical and pathophysiologic aspects of adrenocortical carcinoma. Am J Med. 35:374-383.
27. Emeric-Blanchouin N, Zenatti M, Vonarx V, Schaison G and Aupetit-Faisant B. 1992 Cortisol and testosterone: Inhibitory agents of the mineralocorticoid pathway. Is there a physiopathological meaning in adrenal tumors? J Steroid Biochem Mol Biol. 41:823- 826.
28. Azar ST, Melby JC, Wilson TE, Nichols R, Schwartz JH. 1991 19- Nordeoxycorticosterone synthesis by rat kidney inner medullary collecting duct cells. Steroids. 56:78-80.
29. Soma R, Miyamori I, Nakagawa A, et al. 1989 Possible association of aldosterone producing adenoma and non-functioning adrenal tumor. J Endocrinol Invest. 12:183-186.
30. Maegher AP, Hugh TB, Casey JH, et al. 1988 Primary adrenal tumors-A ten-year experience. Aust NZ J Surg. 58:457-462.
31. Lejonc JL, Brochard P, Schaeffer A, Portos JL. 1978 Les cortico- surrenalomes malins non secretants. Nouv Presse Med. 7:3147.
32. McKenna TJ, O’Connell Y, Cunningham S, McCabe M, Culliton M. 1990 Steroidogenesis in an estrogen-producing adrenal tumor in young woman: Comparison with steroid profiles associated with cortisol-and androgen-producing tumors. J Clin Endocrinol Metab. 70:28-34.