Low-Dose Monitored Mitotane Treatment Achieves the Therapeutic Range with Manageable Side Effects in Patients with Adrenocortical Cancer*
MASSIMO TERZOLO, ANNA PIA, ALFREDO BERRUTI, GIANGIACOMO OSELLA, ANNA ALì, VALENTINO CARBONE, ELIO TESTA, LUIGI DOGLIOTTI, AND ALBERTO ANGELI
Dipartimento di Scienze Cliniche e Biologiche, Medicina Interna I (M.T., A.P., G.O., A.A., A.A.) and Oncologia Medica (A.B., L.D.), Università di Torino, 10043 Torino; and Laboratorio Medico Cesare Battisti (V.C., E.T.), 10023 Chieri, Italy
ABSTRACT
Eight patients with adrenocortical cancer were treated with low doses of mitotane (2-3 g daily) while monitoring drug plasma levels. When the mitotane concentrations reached the therapeutic range (defined as mitotane plasma levels between 14-20 µg/mL), a dose reduction was performed to avoid toxicity. Thereafter, the mitotane dose was tailored according to plasma levels. A progressive increase in plasma mitotane concentrations was observed during treatment, and a highly significant linear correlation was found between plasma drug levels and the total mitotane dose. The therapeutic threshold was reached in all patients after 3-5 months and a total mitotane dose
of 283-387 g/days (median, 363). The duration of treatment was 8-40 months (median, 9). Toxicity was manageable in all but one patient, who discontinued treatment. It is therefore possible to design a stan- dard low dose schedule, e.g. 3 g/daily for about 3-4 months with following dose adjustments guided by the monitoring of plasma mi- totane levels. This approach is able to provide therapeutic mitotane concentrations and limit the unwanted effects. The present data pro- vide a rationale to change the approach to mitotane treatment in patients with adrenocortical carcinoma from high dose to low dose regimens. (J Clin Endocrinol Metab 85: 2234-2238, 2000)
M ITOTANE, 1,1-dichloro-2-(o-chlorophenyl)-2-(p-chloro- phenyl) ethane, is a compound derived from the insecticide dichloro-diphenyl-dichloroethane, which has po- tent adrenotoxic effects (1, 2) and is able to block cortisol synthesis by inhibiting 11ß-hydroxylation and cholesterol chain cleavage (3). For these reasons, mitotane was widely used in the treatment of adrenocortical cancer (ACC) (4-6). After initial enthusiasm for the use of mitotane in ACC based on studies performed in the seventies and eighties, more recent large trials raised doubts about its efficacy in obtaining objective responses and, more important, in prolonging pa- tient survival (7-9). The toxicity of mitotane has been a major limit to its suitability in the treatment of ACC patients. At the currently employed doses of 6-10 g daily, severe side-effects, of either the gastrointestinal or the nervous system, have been frequently reported, and many patients are not able to take the drug regularly (2, 10). A low compliance with treat- ment may ultimately limit its effectiveness.
In a retrospective study, van Slooten et al. (11) suggested that the achievement of appropriate serum mitotane levels (~14 µg/mL) is mandatory to fully exploit its therapeutic action, whereas drug levels exceeding 20 µg/mL should be viewed as toxic. Haak et al. (12) confirmed the value of this
approach, demonstrating a positive effect of mitotane ther- apy on the survival of ACC patients when the threshold value of 14 µg/mL was reached.
The pharmacokinetics of mitotane are characterized by prevalent accumulation in adipose tissue and lower circu- lating levels (13). The semilogarhythmic relationship ob- served between plasma and tissue concentrations during therapy may explain the narrow therapeutic range (13). These studies emphasize the need for blood monitoring, but a mitotane assay is available only in research settings.
The aim of the present study was to perform a prospective evaluation of the tolerance of a low dose mitotane regimen in patients with ACC. The treatment was adjusted by mon- itoring plasma mitotane levels to maintain therapeutic drug concentrations between 14-20 µg/mL according to the above-mentioned studies (11, 12). The working hypothesis was that the greater compliance in taking regularly a better tolerated, low dose treatment could provide more consis- tently therapeutic mitotane levels.
Subjects and Methods
Subjects
From 1994 to the present 15 patients with ACC were treated with mitotane at our institution. In 7 of them, the drug was given in asso- ciation with etoposide, doxorobucin, and cisplatin (EDP regimen) ac- cording to the treatment schedule of a prospective phase II trial (14). The remaining 8 patients (4 females and 4 males; aged 45-62 yr; median age, 52 yr) were not enrolled in this trial for ineligibility (poor performance status in cases 1 and 8, previous chemotherapy in cases 5 and 6, and disease-free status in cases 4 and 7), or chemotherapy refusal (cases 2 and 3). These latter cases were considered in the present study. Pertinent clinical data are given in Table 1. The study design was approved by the
Received September 21, 1999. Revision received November 2, 1999. Rerevision received January 12, 2000. Accepted February 29, 2000.
Address all correspondence and requests for reprints to: M. Terzolo, M.D., Unitá Operativa Autonoma a Direzione Universitaria Medicina Interna I, Azienda Sanitaria Ospedaliera San Luigi, Regione Gonzole 10, 10043 Orbassano (TO), Italy. E-mail: terzolo@usa.net.
* Supported in part by Associazione Italiana per Ricerca sul Cancro (AIRC), Milan, Italy.
| n, sex | Age (yr) | BMI (kg/m2) | ACC stage | Previous treatment | Hormone secretion | Treatment duration (months)ª | Toxicity | Tumor response | Hormone response |
|---|---|---|---|---|---|---|---|---|---|
| 1, F | 47 | 18 | IV | Surgery | No | 8+ | CNS I | SD | |
| 2, F | 48 | 25 | III | Surgery | T, DHEAS, A4 | 8 | CNS II, A II | SD | CR |
| 3, F | 62 | 22 | IV | No | 17-OHP, DHEAS, A4, S, T | 10 | Gi I, A I | P | PR |
| 4, M | 49 | 20 | DF | Surgery | No | 9+ | Gi I, A I, Gyn | DF | |
| 5, M | 54 | 19 | IV | Surgery, chemo | 17-OHP, S | 8+ | Gi I, A I | SD | NC |
| 6, M | 62 | 26 | IV | Surgery, chemo | No | 40+ | No | SD | |
| 7, M | 55 | 24 | DF | Surgery | No | 27+ | No | DF | |
| 8, F | 45 | 23 | IV | Surgery | T, S | 8 | Gi I, A I | P | CR |
Staging of ACC according to Brennan (15) (stage I, tumor ≤5 cm; stage II, tumor >5 cm; stage III, tumor with local infiltration or metastatization in the regional lymph nodes; stage IV, distant metastases). Toxicity and tumor response was graded according to WHO criteria (15), and hormone response was defined as the normalization of altered biochemical parameters; partial hormone response was defined as a more than 50% reduction of altered biochemical parameters. DF, Disease free; chemo, chemotherapy; G.I., gastrointestinal; A, asthenia; CNS, central nervous system; gyn, gynecomastia; SD, stable disease; CR, complete response; PR, partial response; NC, no change. ” The plus sign means that the patient was taking mitotane at the last follow-up.
local ethical committee, and informed consent was obtained from all participants.
Mitotane (Lysodren, Bristol-Myers Squibb Co. Canada Inc., Montréal, Canada; 500-mg tablets) was administered orally at a starting dose of 1 g daily, with progressive weekly increments up to 3 g daily or the highest tolerated dose. When mitotane reached therapeutic concentrations, de- fined by van Slooten et al. (11) as between 14-20 µg/mL, patients were given 1-2 g daily, with further individual dose adjustments guided by the results of mitotane measurement. Mitotane administration was pur- sued until disease progression or occurrence of severe toxicity. Each patient underwent a regular follow-up, including routine biochemistry, endocrine work-up, computerized tomography scan, and physical ex- amination every 3 months. Endocrine work-up included measurement of serum and urinary cortisol, plasma ACTH, serum dehydroepiandro- sterone sulfate, androstenedione, 17-hydroxyprogesterone, 11-deoxy- cortisol, and testosterone. Hormone measurements were performed as previously described (16). No patient had impairment of liver or renal function at the start of treatment or developed clinically significant renal or liver insufficiency. All patients received concomitant steroid replace- ment therapy to prevent adrenal insufficiency. Blood samples for mi- totane assay were collected every month until reaching the therapeutic threshold and thereafter every 3 months. Blood was collected in a K3 ethylenediamine tetraacetate tube; plasma was immediately separated by centrifugation for 15 min at -4 C and was stored at -20 C until assayed.
Toxicity has been scored by means of a specific questionnaire to rate subjective symptoms while objective data were obtained as appropriate. Toxicity and tumor response was graded according to WHO criteria (17). A complete response was defined as the disappearance of all clinical evidence of tumor on physical examination and/or on radiography and the complete recalcification of all osteolytic metastases for a minimum of 4 weeks. A partial response required at least a 50% decrease in all measurable tumor size and 50% or more recalcification of osteolytic appearances of the duration of at least 4 weeks without the appearance of new lesions. Progressive disease was defined as a 25% or more increase in the size of any measurable lesion or the appearance of new lesions.
A complete hormone response was defined as the normalization of all biochemical parameters above the normal range; a partial hormone response was defined as a greater than 50% reduction in altered bio- chemical parameters compared with baseline; other results were defined as no response.
Mitotane assay
Mitotane assay was performed by a modified version of the meth- odology developed by Benecke et al. (18). Mitotane (o,p’-DDD) and its metabolite 1,1-(o,p’-dichlorodiphenyl)-2,2-dichloroethane (o,p’-DDE) were measured by a gas chromatograph (model 8500, Perkin-Elmer Corp., Palo Alto, CA) equipped with a 3H electron capture detection and fitted with a PT-5 fused silica capillary column (i.d., 30 m × 0.25 mm;
film thickness, 0.25 um). The instrument conditions were as follows: carrier gas, helium; carrier gas flow rate, 2.0 mL/min; column temper- ature, 220 C; injector temperature, 250 C; detector temperature, 350 C. Peak area integrations were performed using a Turbochrom Navigator (Perkin-Elmer Corp., Palo Alto, CA).
Mitotane and its metabolite o,p’-DDE and internal standard (p,p’- DDD) were purchased from Supelco (Bellefonte, PA). Stock standard solutions (1 mg/mL) were obtained dissolving o,p’-DDD, o,p’-DDE, and p,p’-DDD (100 mg each) in 100 mL ethanol. A series of working standard solutions containing 5, 10, 20, 40, 80, and 120 µg/mL o,p’-DDD and o,p’-DDE and a constant amount of 40 µg/mL p,p’-DDD in each solution was prepared by serial dilutions of the stock solutions with ethanol. Plasma standards were freshly prepared each day by spiking 200 LL drug-free plasma (blank plasma) with 20 uL of the working standard solutions to yield o,p’-DDD and o,p’-DDE concentrations ranging from 0.5-12 µg/mL in plasma; the concentration of the internal standard was then 4 µg/mL. The efficacy of the recovery was 95%. To each working standard solution, 100 pL acetone were added, and the mixture was shaken mechanically for 30 min. After the addition of 2 mL heptane, the drugs were extracted by mechanical shaking for 30 min. Then, 200 mg anhydrous sodium sulfate were added, and the mixture was centrifuged for 5 min at 3000 X g. About 0.5 pL of the heptane extract was inject into the chromatograph. Plasma samples (200 µL) from pa- tients were spiked with 20 uL internal standard solution and treated in the same way. Each working day, six or seven plasma standards were assayed to construct a calibration graph. Thus, the calibration curve was prepared from drug-free plasma sample spiked with known amounts of o,p’-DDD and o,p’-DDE. The peak areas of o,p’-DDD and o,p’-DDE were measured, and the ratios of o,p’-DDD/p,p’-DDD and o,p’-DDE/ p,p’-DDD were calculated. From a calibration curve, the amounts of mitotane and its metabolite present in 200 µL plasma from each patient were read. The intra- and interassay coefficients of variation were less than 5%.
Statistical analysis
To compare individual regimens that were different in terms of daily dose of mitotane and duration of treatment, the total dose of mitotane was calculated and expressed as grams/days of treatment. Linear re- gression analysis was performed between the total dose of mitotane and its plasma levels at each time point evaluated. Levels of significance were set at P < 0.05.
Results
Mitotane plasma concentrations increased progressively during treatment (Fig. 1), and in each patient a highly sig- nificant linear correlation (r > 0.8) was found between plasma mitotane concentrations and the total mitotane dose. The therapeutic threshold was reached in all patients after
3-5 months and a total mitotane dose ranging between 283- 387 g/days of treatment (median, 363). Daily mitotane doses and the corresponding circulating levels of two illustrative cases (patients 4 and 6) are given in Fig. 2. The duration of mitotane treatment was 8-40 months (median, 9). Tumor and hormone responses are shown in Table 1. Toxic mitotane levels were reached in the first two patients treated because the mitotane dose was not tapered off. Patient 1 experienced confusion and vertigo, which resolved after dose reduction, whereas patient 2 experienced confusion, sleepiness, and ataxia and discontinued mitotane. The remaining patients experienced only moderate unwanted effects (nausea, an- orexia, and asthenia). Two other patients discontinued mi- totane for ACC progression (cases 3 and 8); five patients are still taking mitotane at a dose of 1-2 g daily targeted to maintain therapeutic drug concentrations. In these latter pa-
900
700
Total mitotane dose
500
300
100
-100
-5
0
5
10
15
20
25
30
35
Mitotane levels
tients no tumor progression was demonstrated. During treat- ment, it was necessary to increase the dose of steroid cov- erage in three patients, who displayed ACTH concentrations 25 times higher than the upper normal limit. In all but one patient serum total cholesterol increased, reaching levels above 350 mg/dL in four of them. y-Glutamyl transferase levels rose steeply in all patients, without alterations in other liver function tests (Table 2).
Discussion
The present study is the first prospective evaluation of the value of monitoring plasma mitotane levels to guide treat- ment of ACC patients. This study demonstrates that thera- peutic concentrations of mitotane, as defined by van Slooten et al. (11), can be reached by a sustained low dose treatment. Mitotane doses of 1-3 g daily are much better tolerated than the classic high doses as demonstrated in the present series that also includes compromised patients and in a recent study that revived interest in mitotane therapy in ACC (19). Low dose regimens might be as effective as high dose ones even if the concept of a critical threshold of mitotane levels for antineoplastic activity needs to be confirmed. This study was not designed to evaluate the activity of mitotane; how- ever, mitotane pharmacokinetics, characterized by a pro- longed half-life and accumulation in lipid stores (21), make this drug well suitable for a treatment schedule consisting of a low dose taken over a sustained period of time. As Dick- stein underlined (21), it is possible that the therapeutic effect of low doses might be better than that of high doses because of a greater compliance with a continuous treatment.
The monitoring of plasma mitotane levels allows a poten- tial threshold to be reached with the lowest effective dose, thus limiting side-effects. A better tolerance is achieved with stepwise dose increments continuing (or reducing to) a lower dose until gastrointestinal symptoms resolve. It is also ef- fective to promptly initiate steroid supplementation that, in some patients, requires doses higher than those usually em- ployed (22). This is an important point because symptoms of
4
4
3
g/day
3.
g/day
2
2
1
1
20
20
15
15
4 g/ml
10
u g/ml
10
5
5
0
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
Months
Months
| Patient no. | Months | |||
|---|---|---|---|---|
| 0 | 3 | 6 | 9 | |
| 1 | ||||
| ACTH (pg/ml) | 76 | 20ª | 58ª | 49ª |
| Cholesterol (mg/dL) | 150 | 325 | 415b | 301b |
| Triglycerides (mg/dL) | 171 | 243 | 324 | 184 |
| yGT (U/L) | 102 | 312 | 379 | 388 |
| 2 | ||||
| ACTH (pg/mL) | 56 | 119ª | 1150ª | 1450℃ |
| Cholesterol (mg/dL) | 167 | 202 | 182 | 174 |
| Triglycerides (mg/dL) | 157 | 122 | 124 | 161 |
| yGT (U/L) | 148 | 347 | 270 | 221 |
| 3 | ||||
| ACTH (pg/mL) | 7.5 | 36ª | 78ª | 408ª |
| Cholesterol (mg/dL) | 230 | 379 | 400 | 344 |
| Triglycerides (mg/dL) | 105 | 133 | 180 | 193 |
| yGT (U/L) | 30 | 52 | 160 | 167 |
| 4 | ||||
| ACTH (pg/mL) | 25 | 82ª | 352ª | 774€ |
| Cholesterol (mg/dL) | 205 | 302 | 226b | 176b |
| Triglycerides (mg/dL) | 150 | 130 | 195 | 182 |
| yGT (U/L) | 32 | 119 | 559 | 303 |
| 5 | ||||
| ACTH (pg/mL) | 122 | 47ª | 35ª | |
| Cholesterol (mg/dL) | 239 | 337 | 3026 | 310b |
| Triglycerides (mg/dL) | 120 | 135 | 189 | 173 |
| yGT (U/L) | 65 | 113 | 270 | 278 |
| 6 | ||||
| ACTH (pg/mL) | 55 | 79ª | 64ª | 50ª |
| Cholesterol (mg/dL) | 240 | 344 | 323b | 384b |
| Triglycerides (mg/dL) | 184 | 1037 | 449ª | 507ª |
| yGT (U/L) | 45 | 198 | 410 | 477 |
| 7 | ||||
| ACTH (pg/mL) | 35 | 174ª | 620ª | 870€ |
| Cholesterol (mg/dL) | 214 | 310 | 335 | 290 |
| Triglycerides (mg/dL) | 110 | 140 | 175 | 180 |
| yGT (U/L) | 60 | 350 | 550 | 600 |
| 8 | ||||
| ACTH (pg/mL) | 28 | 178ª | 195ª | 197ª |
| Cholesterol (mg/dL) | 188 | 250 | 270 | 300 |
| Triglycerides (mg/dL) | 110 | 148 | 165 | 180 |
| yGT (U/L) | 50 | 150 | 215 | 188 |
Normal values: ACTH, 10-60 pg/mL; cholesterol, 130-220 mg/dL; triglycerides, 35-160 mg/dL; yGT, 10-45 U/L.
” On 37.5 mg/day cortisone acetate.
b On statins.
€ On 50 mg/day cortisone acetate.
d On omega-3 fatty acids.
adrenal insufficiency may be interpreted as mitotane side- effects. A likely explanation may be the mitotane-induced enhancement of steroid metabolism by microsomial liver enzymes (23). Pertinently, high y-glutamyl transferase levels were observed in all patients, as previously reported (24). Total cholesterol frequently increases with treatment and reaches levels so high that specific therapy may be advised in some patients. This finding has been only occasionally reported (25).
Another novel finding of the present study is that it is pos- sible to predict with sufficient accuracy the total dose needed to attain the therapeutic concentrations, at least with body pro- portions like those of our patients. The monitoring of plasma
mitotane should be initiated when the patient has taken a total dose of about 300 g and then continued on a monthly basis when approaching the critical threshold. When the target is reached, adequate plasma levels of the drug can probably be maintained by regularly taking 1-2 g and measuring mitotane every 3 months to make individual dose adjustments.
To conclude, it is possible to design a standard schedule for mitotane treatment consisting of 3 g daily for about 3-4 months (or, if not tolerated, 2 g for 5-6 months) and then tapering to 1-2 g to be taken chronically. Mitotane dose reduction after having attained the threshold is critical to avoid toxicity as demonstrated by the first two patients who were not offered this option. This protocol is generally well tolerated, and it is able to provide therapeutic concentrations of the drug. The monitoring of mitotane levels helps to make individual dose adjustments, thus reducing toxic effects.
The present data provide a rationale to change the ap- proach to mitotane treatment in ACC patients from high dose to low dose regimens. This alternative treatment modality is tantalizing for adjuvant mitotane treatment, which was in- troduced by Schteingart in 1982 (4) but has not gained broad acceptability, and for a combination treatment of mitotane with other chemotherapeutic agents, an approach that looks promising for advanced ACC (13, 26).
References
1. Hoffman DL, Mattox VR. 1972 Treatment of adrenocortical carcinoma with o,p’DDD. Med Clin North Am. 56:999-1012.
2. Lubitz JA, Freeman L, Okun R. 1973 Mitotane use in inoperable adrenal cortical carcinoma. JAMA. 223:1109-1112.
3. Trainer PJ, Besser M. 1994 Cushing’s syndrome. Therapy directed at the adrenal glands. Endocrinol Metab Clin North Am. 23:571-584.
4. Schteingart DE, Motazedi A, Noonan RA, Thompson NW. 1982 Treatment of adrenal carcinomas. Arch Surg. 117:1142-1146.
5. Boven E, Vermorken JB, van Slooten H, Pinedo HM. 1984 Complete response of metastatized adrenal cortical carcinoma with o,p’-DDD. Case report and literature review. Cancer. 53:26-29.
6. Wooten MD, King DK. 1993 Adrenal cortical carcinoma. Epidemiology and treatment with mitotane and a review of the literature. Cancer. 72:3145-55.
7. Luton JP, Cerdas S, Billaud L, et al. 1990 Clinical features of adrenocortical carcinoma, prognostic factors, and the effect of mitotane therapy. N Engl J Med. 322:1195-1201.
8. Venkatesh S, Hickey RC, Sellin RV, Fernandez JF, Samaan NA. 1989 Adrenal cortical carcinoma. Cancer. 64:765-769.
9. Pommier RF, Brennan MF. 1992 An eleven-year experience with adrenocor- tical carcinoma. Surgery. 112:963-971.
10. Barzilai JI, Pazianos AG. 1989 Adrenocortical carcinoma. Urol Clin North Am. 16:457-468.
11. Van Slooten H, van Seters AP, Smeenk D, Moolenaar AJ. 1982 o,p’-DDD (mitotane) levels in plasma and tissues during chemotherapy and at autopsy. Cancer Chemother Pharmacol. 9:85-88.
12. Haak HR, Hermans J, van de Velde CJ, et al. 1994 Optimal treatment of adrenocortical carcinoma with mitotane: results in a consecutive series of 96 patients. Br J Cancer. 69:947-951.
13. Van Slooten H, Moolenaar AJ, van Seters AP, Smeenk D. 1984 The treatment of adrenocortical carcinoma with o,p’-DDD: prognostic implications of serum level monitoring. Eur J Cancer Clin Oncol. 20:47-53.
14. Berruti A, Terzolo M, Pia A, Angeli A, Dogliotti L for the Italian Group for the Study of Adrenal Cancer. 1998 Mitotane associated with etoposide, doxo- rubicin, and cisplatin in the treatment of advanced adrenocortical carcinoma. Cancer. 83:2194-2200.
15. Brennan MF. 1987 Adrenalcortical carcinoma. Can J Clin. 37:348-365.
16. Osella G, Terzolo M, Borretta G, et al. 1994 Endocrine evaluation of inci- dentally discovered adrenal masses (incidentalomas). J Clin Endocrinol Metab. 79:1532-1539.
17. WHO. 1979 WHO handbook for reporting results of cancer treatment. WHO offset publication 48. Geneva: WHO.
18. Benecke R, Vetter B, De Zeeuw RA. 1987 Rapid micromethod for the analysis of mitotane and its metabolite in plasma by gas chromatography with electron- capture detection. J Chromatogr. 417:287-294.
19. Dickstein G, Shechner C, Arad E, Best LA, Nativ O. 1998 Is there a role for
low doses of mitotane (o,p’-DDD) as adjvant therapy in adrenocortical car- cinoma? J Clin Endocrinol Metab. 83: 3100-3103.
20. Moy RH. 1961 Studies of the pharmacology of o,p’-DDD in man. J Lab Clin Med. 58:296-304.
21. Dickstein G. 1999 Is there a role for low doses of mitotane (o,p’-DDD) as adjuvant therapy in adrenocortical carcinoma? Authors’ response. J Clin En- docrinol Metab. 84:1488-1489.
22. Robinson BG, Hales IB, Henniker AJ, Luttrell BM, Smee IR, Stiel JN. 1987 The effects of o,p’-DDD on adrenal steroid replacement therapy requirements. Clin Endocrinol (Oxf). 27:437-444.
23. Hague RV, May W, Cullen DR. 1989 Hepatic microsomal enzyme induction and adrenal crisis due o,p’ DDD therapy for adrenal carcinoma. Clin Endo- crinol (Oxf). 31:51-57.
24. van Seters AP, Moolenaar AJ. 1991 Mitotane increases the blood levels of hormone-binding proteins. Acta Endocrinol (Copenh). 124:526-533.
25. Maher DM, Trainer PJ, Scoppola A, Anderson JV, Thompson GR, Besser GM. 1992 Possible mechanism and treatment of o,p’-DDD-induced hyper- cholesterolemia. Q J Med. 84:671-679.
26. Bonacci R, Gigliotti A, Baudin E, et al. 1998 Cytotoxic therapy with etoposide and cisplatin in advanced adrenocortical carcinoma. Br J Cancer. 78:546-549.