Expression of P-Glycoprotein in Relation to Clinical Manifestation, Treatment and Prognosis of Adrenocortical Cancer
Harm R. Haak, Arnoud P. van Seters, Abraham J. Moolenaar and Gert Jan Fleuren
The presence of P-glycoprotein, associated with multiple drug resistance and present in the normal adrenal cortex, was studied in 15 cases of adrenocortical carcinoma. P-glycoprotein was found in eight tumours; no correlation was found with clinical presentation, steroid production or histological index. 10 patients received mitotane. Remarkably, 3 patients with P-glycoprotein-positive tumours achieved complete remission. On the other hand, 2 patients with P-glycoprotein-negative tumours showed progression of the disease despite mitotane treatment. These findings suggest that the expression of P-glycoprotein in adrenocortical carcinoma is not related to clinical manifestations, steroid production, histological index or response to mitotane therapy. Eur } Cancer, Vol. 29A, No. 7, pp. 1036-1038, 1993.
INTRODUCTION
ADRENOCORTICAL CARCINOMA is resistant to most chemothera- peutic agents. Mitotane (o,p’-DDD) is the drug of choice for metastatic, or recurrent adrenocortical carcinoma [1]. The response rates reported for mitotane range from 29 to 61% [2, 3]. Serum levels exceeding 14 mg/l seem to correlate best with tumour response [4]. However, most patients will suffer a recurrence or develop metastases.
In a variety of tumours multiple drug resistance (MDR) is associated with expression of the MDR-1 gene, leading to the presence of P-glycoprotein. P-glycoprotein is a plasma membrane-bound transport protein that causes decreased (cytotoxic) drug accumulation within the cell [5, 6]. Since P- glycoprotein is expressed in the normal adrenal cortex [5, 7] it is possible that this effusion system is responsible for the intrinsic resistance of adrenocortical carcinoma to cytotoxic agents.
We investigated the relationship between the presence of P- glycoprotein, clinical manifestations, morphological parameters and treatment of adrenocortical cancer.
PATIENTS AND METHODS
Patients
15 adrenocortical carcinoma patients (12 female, 3 male) were enrolled in the study. All patients had undergone surgery. The operation was considered radical in only 4 of the 15 patients.
10 patients received mitotane therapy. Therapy was started at 4-8 g/day in four equal doses. Mitotane serum concentrations of over 14 mg/l, determined according to the method described by Moolenaar [8], were aimed at.
Evaluation of response was based on the standard criteria of complete remission (no detectable disease), partial response
Correspondence to H.R. Haak at the Department of Internal Medicine, Diaconessenhuis P.O. Box 90052, 5600 PD Eindhoven, The Nether- lands.
H.R. Haak and A.P. van Seters are at the Department of Endocrinology; A.J. Moolenaar is at the Department of Pathological Chemistry; G.J. Fleuren is at the Department of Pathology, University Hospital Leiden, The Netherlands.
Revised 16 Oct. 1992; accepted 3 Nov. 1992.
(≥ 50% reduction), minimal response (≥25 -< 50% reduction), stable disease and progressive disease (≥ 25% increase).
Immunohistochemistry
Fresh-frozen tumour tissue (-75℃) was stained immunohis- tochemically with the murine monoclonal antibody JSB-1 [9] against P-glycoprotein, according to the immunoperoxidase amino-ethyl-carbazole (AEC) method [10]. The presence of P- glycoprotein was scored as - (negative), + (10-24% cells positive), ++ (25-49% cells positive), +++ (50-74% cells positive), and ++++ (75% or more cells positive). A normal adrenal gland served as a positive (cortex) as well as negative control (medulla).
The patients were divided in two groups: P-glycoprotein- negative (group 1) and P-glycoprotein-positive (group 2).
Histological index
Seven parameters (regressive changes, preservation of normal structure, nuclear atypia, nuclear hyperchromasia, structure of nucleoli, mitotic activity and invasion of the capsule and/or vascular wall), each with a specific discriminating value, deter- mine the histological index according to van Slooten et al. [11]. The maximal index is 28.4. A tumour with an index above 8 is considered malignant.
Steroid production
Steroid production was assessed by determination of the pretreatment serum level of dehydroepiandrosterone sulphate and the excretion of 17-oxo and 17-oxogenic steroids in 24-h urine corrected for 10 mmol creatinine excretion.
Statistical analysis
The Mann-Whitney test was used for statistical analysis.
RESULTS
Expression of P-glycoprotein
P-glycoprotein was present in eight tumours (three ++++, two ++ +, two + + and one +), and absent in seven tumours.
| expression | ||
|---|---|---|
| P-glycoprotein- negative (group 1) n = 7 | P-glycoprotein- positive (group 2) n = 8 | |
| Dehydroepiandrosterone sulphate (umol/l) | 7.9 ± 10 | 8.1 ± 11 NS |
| 17-Oxo-steroids (pmol/24-h/10 mmol creatinine) | 155 ± 191 | 97 ± 49 NS |
| 17-Oxogenic-steroids (p.mol/24-h/10 mmol creatinine) | 68 ± 64 | 44 ± 28 NS |
| Histological index | 21.4 ± 5.6 | 25.0 ± 4.3 NS |
Values are expressed as mean ± S.D.
Clinical presentation and steroid production
Clinical hormonal activity was observed in 11 patients. The expression of P-glycoprotein was not related to the clinical symptoms or to steroid production (Table 1).
Histological index
The histological index of the tumours ranged from 13.6 to 28.4 in the 15 patients. In 3 patients with an index of 28.4 the tumour was P-glycoprotein-positive: 25→ 75% of the cells were positive. 1 patient with the maximum index showed no P- glycoprotein immunoreactivity. The other 6 patients without P- glycoprotein in the tumour had a histological index between 13.7 and 24.2. The mean histological index in group 1 did not differ statistically significantly from the mean histological index in group 2 (Table 1).
Response to mitotane therapy: P-glycoprotein-positive tumours (n = 7) (Table 2)
3 patients (numbers 5, 6 and 8) achieved a complete remission. A minimal tumour response was seen in patient number 10. 2
| Patient no. | P-glycoprotein presence | Duration of mitotane therapy (months) | Mitotane level (mg/l) | Response |
|---|---|---|---|---|
| 1 | - | 2 | <14 | PD |
| 2 | - | 4 | <14 | PD |
| 3 | 9 | >14 | PR, 5 months | |
| 4 | + | 12 | >14 | ND |
| 5 | 6 | >14 | CR, 2 months | |
| 6 | 24 | >14 | CR, 60 months* | |
| 7 | 5 | >14 | PD | |
| 8 | 12 | >14 | CR, 12 months | |
| 9 | 4 | >14 | PD | |
| 10 | 3 | >14 | MR, 2 months |
CR = Complete response, PR = partial response, MR = minimal response, PD = progressive disease, ND = not determinable, mitotane prophylactically.
*Ongoing remission.
patients did not respond to mitotane therapy. The results obtained in patient number 4 could not be evaluated because mitotane had not been received adjuvantly; the patient is tumour-free 1 year after starting mitotane.
Response to mitotane therapy: P-glycoprotein-negative tumours (n = 3) (Table 2)
Patient number 3 (mitotane serum level over 14 mg/l) had a partial response lasting 5 months. Patients number 1 and 2 had progressive disease despite mitotane (serum levels below 14 mg/l).
DISCUSSION
Large variations in expression of the MDR-1 gene, indicated by the presence of P-glycoprotein, were observed. The presence of P-glycoprotein in adrenocortical carcinoma was not correlated with the histological index. Apparently, less well differentiated adrenocortical carcinomas do not lose the ability to express P- glycoprotein.
In the adrenals, P-glycoprotein is expected to be related to transport of steroids out of cells [5]. If P-glycoprotein plays an important role in steroid excretion, we would expect patients with a tumour lacking this protein to exhibit hormonal inactivity. Patients with pronounced P-glycoprotein expression would show overt hormonal activity. Our results, however, offer no support for the hypothesis that P-glycoprotein plays an important role in hormone transport out of adrenocortical cells.
2 patients with a tumour lacking P-glycoprotein who received mitotane therapy had progressive disease despite this therapy. Unfortunately, mitotane therapy in these cases was not optimal since the serum levels were below 14 mg/l. In 3 patients with P- glycoprotein expression on mitotane therapy, complete remission was achieved. These results indicate that the response of adrenocortical cancer to mitotane is not related to the presence or absence of P-glycoprotein. In fact, it was recently found that mitotane reverses MDR in adrenocortical carcinoma in vitro [12]. In adrenocortical carcinoma, as well as adrenocortical adenoma or hyperplasia, mitotane was found in all assayed samples [13]. Mitotane can inhibit steroid synthesis in normal and pathological adrenal glands [2, 14]. In view of the presence of P-glycoprotein in healthy and pathological adrenal cortices, it may be concluded that mitotane is not pumped out of the cells by the P-glycoprotein transport system, at least not in substantial amounts.
Research on adrenocortical cancer is limited by the rarity of this tumour. The results in our group of patients suggest, however, that resistance of adrenocortical cancer to mitotane is not regulated via the P-glycoprotein transport system. The clinical presentation, steroid excretion and histological index or degree of differentiation of the tumour are not related to the presence of P-glycoprotein. We are inclined to conclude at this time that expression of the MDR-1 gene is probably not important for determining the prognosis of patients with adrenocortical cancer.
1. Samaan NA, Hickey RC. Adrenal cortical carcinoma. Sem Oncol 1987, 14, 292-296.
2. Guteirrez ML, Crooke ST. Mitotane (o,p-DDD). Cancer Treat Rev 1980, 7, 49-55.
3. Venkatesh S, Hickey RC, Sellin RV, et al. Adrenal cortical carci- noma. Cancer 1989, 64, 765-769.
4. van Slooten H, Moolenaar AJ, van Seters AP, Smeenk D. The treatment of adrenocortical carcinoma with o,p-DD: prognostic
implications of serum level monitoring. Eur J Clin Oncol 1984, 20, 47-53.
5. Pastan I, Gottesman M. Multiple-drug resistance in human cancer. New Engl J Med 1987, 316, 1388-1393.
6. Rothenberg M, Ling V. Multidrug resistance: molecular biology and clinical relevance. } Natl Cancer Inst 1989, 81, 7-10.
7. Fojo AT, Ueda K, Slamon DJ, et al. Expression of a multidrug- resistance gene in human tumors and tissues. Proc Natl Acad Sci USA 1987, 84, 265-269.
8. Moolenaar AJ, Niewind JWM, Oei IT. Estimation of o,p-DDD in plasma by gas-liquid chromatography. Clin Chim Acta 1977, 76, 213-218.
9. Scheper RJ, Bulte JWM, Brakkee JGP, et al. Monoclonal antibody JSB-1 detects a highly conserved epitope of the P-glycoprotein associated with multidrug-resistance. Int J Cancer 1988, 42, 389-394.
10. de Bruijn HWA, van Beeck Calkoen-Carpay T, Jager S, et al. The tumor marker CA125 is the common constituent of normal cervical mucus. Am J Obstet Gynecol 1986, 154, 1088-1091.
11. van Slooten H, Schaberg A, Smeenk D, Moolenaar AJ. Morphologic characteristics of benign and malignant adrenocortical tumors. Cancer 1985, 55, 766-773.
12. Bates SE, Shieh CY, Mickley CA, et al. Mitotane enhances cytotox- icity of chemotherapy in cell lines expressing a multipledrug resist- ance gene (mdr-1/P-glycoprotein) which is also expressed by adreno- cortical carcinomas. J Clin Endocrinol Metab 1991, 73, 18-29.
13. Touitou Y, Moolenaar AJ, Bogdan A, et al. O,p-DDD (mitotane) treatment for Cushing’s syndrome: adrenal drug concentrations and inhibition in vitro of steroid synthesis. Eur J Clin Pharmacol 1985, 29, 483-487.
14. Bricaire H, Luton JP. Douze ans de traitement médical de la maladie de Cushing. Nouv Presse med 1976, 5, 325-329.
Acknowledgements-The authors wish to thank Dr J. Hermans for his advice concerning statistical analyses, and Dr N.A.T. Hamdy and Prof. H.M.J. Krans for their critical review of the manuscript.
Eur J Cancer, Vol. 29A, No. 7, pp. 1038-1044, 1993. Printed in Great Britain
0964-1947193 $6.00 + 0.00 C) 1993 Pergamon Press Lid
Feature Articles
Treatment Options in Clinical Stage I Non-seminomatous Germ Cell Tumours of the Testis: A Wager on the Future? A Review
J.P. Droz and A. T. van Oosterom
INTRODUCTION
THE PROGNOSIS of metastatic germ cell tumours of the testis has dramatically improved with the introduction of cisplatin-based chemotherapy [1]. One of the most important refinements in the treatment of these tumours is the use of prognostic factors to indicate risk-adapted treatments. This is currently used in the treatment of advanced disease [1]. It is noteworthy that even in the past the cure rate of patients with tumours localised in the testis only (clinical stage 1) was high [2, 3]. The standard treatment of these patients was either retroperitoneal lymph node dissection (RPLND) or radiotherapy to the retroperitoneal region and cure rates as high as 80-90% have been reported [2]. However, the “wait and see” policy (surveillance) emerged in the early 1980s [4] when it was proven that low volume metastatic disease could be cured by chemotherapy in the majority of cases and more reliable investigations including chest and abdominal computed tomography (CT) scan, but especially serum alpha- fetoprotein (AFP) and human chorionic gonadotrophin (HCG) measurements made earlier detection possible. This policy then became popular. Unfortunately, no evaluation of both modalities
has been undertaken in a randomised trial. We review here the status of the different treatment options in localised testis cancer.
We only consider in this review non-seminomatous germ cell tumours of the testis as defined both in the World Health Organization (WHO) classification [5] and the English classifi- cation [6]. Clinical stage I disease is defined according to the Union Internationale Contre le Cancer staging system [7] as a disease without visceral and retroperitoneal lymph node metast- ases after physical examination, imaging and biochemical tests.
CLINICAL AND PATHOLOGICAL STAGE I, CLARIFYING THE PROBLEM
Clinical staging is based on clinical examination, measurement of serum tumour markers (AFP, HCG), chest radiographs and CT scan of the chest and abdomen with and without bilateral pedal lymphangiography. In clinical stage I disease there is no evidence of tumour spread demonstrated by these investigations. When postorchiectomy tumour marker levels remain elevated involvement of retroperitoneal lymph nodes or metastatic spread elsewhere is found in all cases [8]. One way to evaluate clinical staging is to compare the results of investigations with the results of surgical staging (that is RPLND). Table 1 summarises findings from different studies [9-23]. Three conclusions can be drawn from this table: (i) the false negative and false positive rates of lymphangiography and abdominal CT scan are approximately equal to 30 and 15%, respectively. (ii) From the data in the