Mitotane increases the blood levels of hormone-binding proteins
Arnoud P. van Seters1 and Abraham J. Moolenaar2
Departments of Endocrinology’ and Clinical Chemistry2, University Hospital, Leiden, The Netherlands
Abstract. In 3 patients with adrenocortical carcinoma the effects of long-term mitotane therapy on the serum levels of three hormone-binding globulins and vitamin D-binding protein were studied. Within the first month of treatment cortisol-binding globulin increased two to three times, in close correlation with sex hormone-binding globulin. The rises in thyroxine-binding globulin and vi- tamin D-binding protein were considerably less. Elevated cortisol-binding protein appeared to be associated with increased binding of cortisol, whereas the binding of thy- roxine and vitamin D remained below normal. Binding proteins returned to normal in 2 patients within a year after mitotane discontinuation. This phenomenon of hor- mone-binding protein enhancement invalidates the use of total serum hormone levels to monitor the effects of mitotane on endocrine function and could provide an explanation for the increased cortisol substitution re- quirement during mitotane therapy.
1-(0-chlorophenyl)-1-(p-chlorophenyl)-2,2-dichlo- roethane (o,p’-DDD, mitotane) is still the drug of choice for the treatment of adrenocortical carci- noma (1). Its effectiveness is attributed to a toxic action on adrenocortical mitochondria and inter- ference with the cytochrome P-450-dependent steps in steroid synthesis (2). o,p’-DDD is also a potent inductor of hepatic microsomal P-450 cyto- chromes which promote the formation of highly polar steroid metabolites, such as 6ß-hydroxycor- tisol, at the expense of ring A-reduced glucuro- nides (3). These strongly hydrophilic steroids are lost during the common extraction procedures for urinary steroids. As a result, decreased steroid me- tabolite excretion may not imply suppressed ste- roid secretion during o,p’-DDD treatment. Mea- surement of blood steroids is generally considered
to be a more reliable means of monitoring steroid responses to o,p’-DDD, but its clinical value has re- mained unquestioned. Yet, various observations suggest that steroid physiology, apart from secre- tion and metabolism, is affected by the drug. For instance, signs of glucocorticoid deficiency can de- velop after a substitution dose of dexamethasone or hydrocortisone, normally used to treat Addi- son’s disease (4-6). As far as dexamethasone is con- cerned, this can be explained by an increased dis- appearance rate (4,5). However, the blood t1/2 for cortisol is either slightly prolonged (4,7) or un- changed (3) during o,p’-DDD medication. There- fore, other factors should be considered which could interfere with the action of cortisol on its target tissues. There is no clear evidence in the literature for o,p’-DDD effects on hormone-bind- ing protein levels or binding properties in human serum other than decreased thyroxine-binding (8). However, in dogs a 30% increase in corticosteroid- binding capacity has been reported (9). In addition, Schteingart et al. (10) described a group of patients with Cushing’s disease on o,p’-DDD therapy who exhibited normalized cortisol secretion but ele- vated plasma cortisol. This observation suggests an increase in serum levels of cortisol-binding globulin (CBG) which was recently reported as a side-effect of o,p’-DDD therapy (11).
Since increased CBG might have important di- agnostic and therapeutic implications we measured serum levels of CBG in 3 patients with an adreno- cortical carcinoma before, during and after with- drawal of long-term o,p’-DDD therapy. The study also included several other hormone-binding pro- teins and vitamin D-binding protein in serum.
Patients and Methods
The patients, one male and two females, received o,p’- DDD therapy immediately after incomplete removal of an adrenocortical carcinoma. The tumours could be clas- sified as follows: one feminizing carcinoma (patient No. 1, male, age 56), one mineralocorticoid-producing carci- noma (patient No. 2, female, age 35), and one clinically silent carcinoma (patient No. 3, female, age 42). In all cases, o,p’-DDD was given for 20-24 months in sufficient amounts to maintain a fasting o,p’-DDD serum level of at least 20 mg/l (4). The drug was administered daily as tablets (mitotane), except at the onset of therapy when an o,p’-DDD milk powder mixture was used to facilitate drug resorption and to achieve the required serum levels (12). Substitution with hydrocortisone at increasing doses (up to 60 mg daily) and fludrocortisone acetate (0.2-0.6 mg daily) was started within the first month of o,p’-DDD ther- apy. No medication with known effects on hormone-bind- ing proteins was used. All three patients had severe ad- verse reactions to o,p’-DDD medication, including pain- ful, bilateral gynecomastia in the male, which became manifest when if estrogens had fallen to below the detec- tion limits. Clinically all three patients remained euthy- roid.
A fourth patient with adrenocortical carcinoma on o,p’- DDD and the available data on 9 individuals with adre- nocortical insufficiency or dexamethasone-induced hypo- cortisolemia, contributed to the cortisol “loading studies”. As none of the 9 controls had nephrosis or Cushing’s syndrome, or received estrogen therapy, their CBG serum levels could be assumed to be normal.
Assays and reference values
Cortisol-binding globulin was measured by radioimmu- noassay (RIA) (Medgenix, Fleurus, Belgium; inter-assay variability 8.4%). Thyroxine-binding globulin (TBG) was also measured by RIA (Orie Industrie, Gif-Sur-Yvette, Cedex, France, inter-assay variability <10%), sex hor- mone-binding globulin (SHBG) by immunoradiometric assay (Farmos, Oulunsala, Finland; inter-assay variability 3%) and vitamin D-binding protein by radial immunodif- fusion (RID) according to the method of Bouillon et al. (13); inter-assay variability <3%). In a number of samples CBG was also measured by RID (14). Fasting serum o,p’- DDD levels (at least 14 h after the last dose) were assayed by gas-liquid chromatography with a detection limit of 0.5 mg/l and inter-assay variability of <4.3% (15). o,p’-DDD resorption from different vehicles was compared by mea- suring serum o,p’-DDD levels for 24 h following the oral administration of 2 g of o,p’-DDD.
In a study, which was designed to compare the in vivo cortisol-binding capacities of CBG in patients and their controls, one oral dose of cortisol (10 mg and 30 mg, respectively) was given to steroid-fasted subjects; blood
was collected at time 0 and 180 min (“loading studies”). Cortisol and T4 were measured by a fluorescence energy- transfer immunoassay (Syva-Advance, Palo Alto, CA; de- tection limit 0.05 umol/l, inter-assay variability 2.9%) and a fluorescence polarization immunoassay, respectively (Abbott-TDX, North Chicago, IL; detection limit 10 nmol/l, inter-assay variability 2.3%). The following hor- mones were measured by RIA: ACTH (Incstar, Still- water, MN; inter-assay variability 7-9%), free T4 (Dade, Cambridge, MA; detection limit 1.9 pmol/l, inter-assay variability <15%), PTH (Incstar, measuring the intact molecule, detection limit 0.6 pmol/l, interassay variability 7-9%), estradiol (E2, Farmos, Oulunsala, Finland; detec- tion limit 40 pmol/l, inter-assay variability 7%), estrone (E1, Miles Scientific, Naperville, IL; detection limit 70 pmol/l, inter-assay variability 18%), and testosterone (Byk, Dietzenbach, Germany, detection limit 0.2 nmol/l, inter-assay variability 6.8%). Estimations of estriol (E3, Dr C. Longcope, MA) and 25-hydroxyvitamin D (25-OHD, Mr I. Jans, in the laboratory of Prof R. Bouillon, Leuven, Belgium) were kindly performed by RIA and competitive protein-binding (16) respectively; 1,25-dihydroxyvitamin D (1,25-(OH)2D) was measured by a radioreceptor assay (17). TSH, LH and FSH were measured by a time-re- solved fluorimetric assay, with an inter-assay variability of <5.0% (Pharmacia, Uppsala, Sweden).
Interference with the assays by the drug was excluded by adding appropriate amounts of o,p’-DDD and its main metabolite ortho-para-dichloro-diphenyl acetic acid (o,p’- DDA) to control sera. o,p’-DDA was synthesized from technical grade o,p’-DDE (1-(0-chlorophenyl)-1-(p-chlo- rophenyl)-2,2 dichloroethylene) by the method of Grum- mit et al. (18). A number of serum samples of patients on o,p’-DDD with elevated CBG and SHBG levels were di- luted, up to ten times, with buffer provided by the man- ufacturer, to confirm parallelism with their standard curves. Blood was collected between 08.00 and 11.00 h; plasma and serum samples were kept at -20℃ until pro- cessing. All protein hormone assays were calibrated to the appropriate WHO reference preparations.
Reference values
Cortisol-binding globulin 0.6-0.9 umol/l; SHBG 40-80 (fertile women), 20-55 nmol/l (men); TBG 12-28 mg/l; vitamin D-binding protein 5.95±1.03 (SD) umol/l; cortisol 0.2-0.6 umol/l (08.00 h); ACTH 20-90 ng/1 (08.00 h); T4 70-160 nmol/l; free T4 7.7-20.6 pmol/l; TSH 0.1-6 mU/l; E2 70-450 (females, follicular phase), 70-150 pmol/l (men); E, 93-930 (females), 110-185 pmol/l (men); E3 65- 120 pmol/l (2 men); LH 2-15 U/1, FSH 2-10 U/l (females, follicular phase), respectively 2-9 U/l and 2-10 U/l in men; testosterone 14-52 nmol/l (men); 25-OHD 31-129 nmol/l; 1,25-(OH)2D 40-140 pmol/l; PTH <7.5 pmol/l, in the presence of normal albumin-corrected serum calcium levels.
Statistics
Correlation analysis was performed by calculating the Pearson correlation coefficient. Orthogonal regression analysis was employed for the correlations between CBG, SHBG, TBG and o,p’-DDD serum levels (19), calculated according to Harff & Helversteijn (20). To compare two regression slopes we calculated their difference and the standard error of this difference. The quotient of this difference to its standard error was analysed statistically according to a standard normal distribution.
Ethics
As the study included no procedures beyond standard therapy and close patient care, no special instructions by the local Ethical Committee were involved. It was carried out by the highest ethical standards and in accordance with the laws of the Netherlands.
Results
Serum levels of hormone-binding proteins and other proteins, as changed by o,p’-DDD treatment As shown in Table 1, normal pretreatment serum levels were obtained for the three binding proteins.
Cortisol-binding globulin and SHBG rapidly in- creased during medication: this effect was already pronounced at the end of the first month of treat- ment. Cortisol-binding levels by RIA (x), up to 5.25 umol/l, agreed perfectly with CBG values obtained by RID (y); the regression equation was y= 1.02 x + 0.08 (r=0.99, p<0.001, N=8). Diluted serum sam- ples showed close parallelism to standard curves for CBG and SHBG. The increments in CBG and SHBG for the 3 patients ranged from 500 to 604% and from 567 to 841%, respectively, after 6-7 months of therapy. Within this period, CBG reached 78-97% of its maximal value (Fig. 1). TBG increments were much smaller, not exceeding 80% of the basal values. Serum levels of CBG were closely related to SHBG and TBG levels through- out therapy. These relationships could be de- scribed by an orthogonal linear regression, with highly significant coefficients of regression (p<0.001) ranging from 0.94 to 0.96 for SHBG versus CBG (day 0-722) and 0.91 to 0.99 for TBG versus CBG (day 53-707), respectively. The slopes of the regression lines varied from 59.6 to 92.2
| Patient | Time of treatment (months) | o,p'-DDDª dose level | CBG (umol/l) | SHBG TBG serum levels | Y-GT (U/I) | ||
|---|---|---|---|---|---|---|---|
| (g/day) | (mg/l) | (nmol/l) | (mg/l) | ||||
| 1 | 0 | 0 | 0 | 0.88 | 34 | 28.2 | 10 |
| 1.0 | 6.0b | 30.0 | 2.54 | 122 | - | 87 | |
| 2.1 | 6.0 | 29.8 | 2.79 | 172 | 30.2 | 112 | |
| 7.2 | 9.0 | 26.6 | 5.28 | 320 | 41.8 | 186 | |
| 13.0 | 8.0 | 18.9 | 4.28 | 207 | 38.4 | 192 | |
| 20.0 | 7.5 | 28.9 | 4.84 | 307 | 39.6 | 148 | |
| 2 | 0 | 0 | 0 | 0.80 | 54 | 24.3 | 24 |
| 0.9 | 6.0b | 17.5 | 2.07 | 150 | - | 40 | |
| 2.3 | 6.0 | 21.5 | 2.92 | 235 | 25.4 | 110 | |
| 6.7 | 4.0 | 28.8 | 4.69 | 360 | 37.2 | 61 | |
| 12.4 | 1.5 | 30.6 | 5.60 | 397 | 43.7 | 80 | |
| 21.0 | 2.5 | 32.6 | 5.30 | 495 | 42.0 | 60 | |
| 3 | 0 | 0 | 0 | 0.63 | 33 | 29.6 | 5 |
| 1.4 | 8.0b | 15.6 | 1.63 | 82 | - | 107 | |
| 2.1 | 8.0 | 18.1 | 2.29 | 110 | 29.4 | - | |
| 7.6 | 8.0 | 31.0 | 4.44 | 241 | 39.6 | 129 | |
| 12.9 | 4.5 | 35.7 | 4.92 | 251 | 42.1 | 67 | |
| 24.0 | 2.0 | 15.2 | 3.65 | 225 | 38.1ª | 53 | |
a: factor for conversion to umol/l: 3.125; b: o,p’-DDD in milk powder; c: 23 months
o, p’- DDD (g/day)
0
5
10
pat. 1
40
30
6
20
4
10
2
0
0
0
5
pat. 2
o,p’-DDD ( mg/l)
10
40
CBG ( umol/l)
30
6
20
4
10
2
0
0
0
5
10
pat. 3
40
30
6
8
o
20
4
10
2
0
60
120
180
240
300
360
420
480
540
600
660
720
0
Days of treatment
(SHBG versus CBG) and from 5.0 to 6.2 (TBG versus CBG). There were no interindividual dif- ferences, with the exception of a significantly higher slope of SHBG for patient No. 2 compared with patient No. 3 (p<0.001).
Levels of vitamin D-binding protein prior to o,p’- DDD medication were also within normal limits. By the 4-8th month of therapy they had risen from 7.1 to 9.1 umol/l (13.8 umol/l at month 20) in patient No. 1, from 6.3 to 11.2 umol/l in patient No. 2 and from 7.5 to 11.8 umol/l in patient No. 3. Within the first month of o,p’-DDD therapy, normal pretreat- ment y-glutamyl transferase rose steeply (Table 1) with normal transaminase and total serum protein levels throughout the study. Alkaline phosphatase was only slightly increased in patient No. 1 (72 U/l, reference <60) prior to the removal of a solitary hepatic metastasis which remained stationary de-
spite suppressed estrogen production. In all 3 pa- tients C-reactive protein, angiotensin converting enzyme, lysozyme and thyroglobulin remained un- changed.
Relationship of binding proteins to o,p’-DDD levels For patients No. 2 and 3, the three binding proteins were significantly related to o,p’-DDD levels during more than one year of treatment; coefficients of regression (r) were: for CBG 0.93 and 0.95, for SHBG 0.88 and 0.91, for TBG 0.89 and 0.94 (p<0.001). However, for patient No. 1 the corre- lations for both CBG and SHBG (r=0.97 and 0.93, p<0.01) were lost after the second month of treat- ment. After this short period of time, in which the slopes of the regression lines for CBG and SHBG versus o,p’-DDD were more than 50% less than in the two other patients (p<0.001, p<0.05, respec-
tively), CBG continued to increase, whereas the o,p’-DDD levels remained almost constant (Fig. 1). Estimations of TBG were too few to calculate its relationship to o,p’-DDD for this patient.
Patient No. 1 also differed from the others as serum o,p’-DDD levels, on a daily dose of 6-8 g of o,p’-DDD in milk powder, were built up signifi- cantly faster than in patients No. 2 and 3 (p<0.001). A value of 25 mg/l was attained after only 21 days in patient No. I versus 89 and 146 days, respectively, in patients No. 2 and 3. Also his serum o,p’-DDD levels remained quite unstable de- spite a higher dose of o,p’-DDD than in the other patients (Fig. 1). The latter phenomenon in patient No. 1 was associated with a relatively short half- time (t1/2) for o,p’-DDD (Fig. 2) and poor drug re- sorption from tablets, the latter being 10% of the resorption from milk powder (12). Moreover, he developed persistent diarrhea during o,p’-DDD maintenance therapy.
Hormonal changes during o,p’-DDD therapy; cortisol-loading studies
To examine the serum cortisol binding capacity, loading-studies were performed with exogenous cortisol in 4 patients on o,p’-DDD with eleva-
tedCBG levels (3.01-4.44, mean value 3.88 umol/l). Serum cortisol levels rose from 0.11-0.24 to 1.32- 1.76 umol/1 3 h after oral administration of cortisol (10 mg). In the control patients, 30 mg of cortisol induced a rise from <0.05-0.07 to 0.28-0.87 umol/l; the increments averaged 1.34 and 0.59 umol/l, re- spectively (p<0.001, unpaired t-test). In patient No. 1, a normal basal cortisol level (0.25 umol/l) coincided with elevated plasma ACTH (470 ng/l), which became adequately suppressed when cortisol substitution was increased from 30 to 60 mg daily.
During o,p’-DDD therapy, in the 3 patients, total serum T4 decreased to 38% of the respective normal pretreatment levels (113-125 nmol/l), vis à vis increased TBG concentrations and normal levels of TSH (0.1-3.2 mU/l) and free T4 (12.1-19.3 pmol/l).
Despite increased levels of vitamin D-binding protein, the available values for the ligand, serum 1,25-(OH)2D, were certainly not elevated: 17 pmol/l (patient No. 2 at month 13 of therapy) and 69 pmol/l (patient No. 3 at month 8); during o,p’- DDD medication 25-OHD, PTH and serum crea- tinine remained within normal limits.
Prior to o,p’-DDD, gonadotropins and gonadal steroids were normal in all 3 patients, with the ex-
1000
pat. 1
pat. 2
pat. 3
Serum concentration
100
10
1
0.5
0
100
200
300
0
100
200
300
0
100
200
300
Withdrawal ( days )
ception of a slightly elevated E2 level (283 pmol/l) in patient No. 1. By the 9-12th month of therapy LH had increased to 21.6-30.2 U/l, whereas FSH re- mained within normal limits. In patient No. 1, E2 and E1 levels fell below the detection limit; serum E3 and testosterone did not exceed 63 pmol/l and 2.0 nmol/l, respectively, throughout the remainder of therapy, despite progressive gynecomastia. In patient No. 2, serum E2 and E1 also fell below the detection limit at month 9. Subsequently, these values fluctuated, with maximal values of 512 and 319 pmol/l, until complete estrogen suppression was effected with lynestrenol and buserelin, which was administered from day 556 to 730 of therapy. This therapeutic procedure had no apparent effect, independent of o,p’-DDD, on the levels of CBG (Fig. 1), SHBG and TBG. In the second year of treatment, patient No. 3 also exhibited great fluctuations in preovulatory E2 levels, ranging from <40 to 1584 pmol/l. She received no busere- lin.
Recovery after discontinuation of mitotane
After discontinuation of o,p’-DDD therapy, o,p’- DDD levels slowly decreased in the 3 patients, but remained well above the detection limit for over one year (Fig. 2). Half-times for o,p’-DDD were 76, 96 and 122 days, which is in accordance with a previous study (12). A log/linear analysis of the re- lation of concentration to time resulted in highly significant coefficients of regression for the three hormone-binding proteins studied. Their half- times ranged from 74-195 days, whereas the t1/2 values in normal conditions, estimated by radioac- tive tracers, are approximately 5 days. Levels of hormone-binding proteins and o,p’-DDD were also closely related (Fig. 2), but the slopes of the regres- sion lines during withdrawal and maintenance therapy differed by up to a factor of 3. In patients No. 1 and 3 CBG and TBG levels normalized within 9 months, but SHBG in patient No. I re- mained slightly elevated; at this time their serum E2 were 53 and 736 pmol/l, respectively. In patient No. 2 TBG also became normal, but CBG and SHBG levels were still elevated at the end of ob- servation (preovulatory serum E2 2406 pmol/l).
Discussion
The present study demonstrates that o,p’-DDD therapy induces significant increases in CBG, SHBG, TBG and vitamin D-binding protein in
human peripheral blood. A similar but less striking effect on CBG was reported for dogs as opposed to guinea pigs, rats and chickens (9). In recent human studies increased CBG and SHBG serum levels were mentioned as side-effects of o,p’-DDD ther- apy in adrenocortical carcinoma (11). Since C-re- active protein and non-hepatocyte-derived pro- teins (lysozyme and thyroglobulin) remained unal- tered in our study, this response does not represent a “generalized reaction” of protein synthesis to the drug. The close correlation between serum levels of o,p’-DDD and binding proteins, both during ther- apy and o,p’-DDD withdrawal, suggests a causal role for the drug. Buserelin and lynestrenol, which were applied as adjuvants in this study, do not affect CBG and SHBG, whereas glucocorticoids only depress binding proteins when administered in excess. Other explanations for increased SHBG, such as weight loss and liver damage, do not seem to be applicable in our patients.
Among the various factors which influence bind- ing proteins, estrogen excess and pregnancy are the only known conditions that increase both CBG, SHBG, TBG and vitamin D-binding protein in the blood (21,22). Feminizing phenomena are striking clinical sequellae of o,p’-DDD treatment. They could be related to increased synthesis of estrogens, as suggested by Schteingart et al. (10) or represent an estrogen-like action of the drug. Related chlo- rinated hydrocarbons, such as DDT and the ß-iso- mer of lindane, were shown to have estrogenic properties (23-25), also in ovariectomized-adrena- lectomized rats. Increased endogenous estrogen activity cannot be excluded in our patients, as es- timations of serum free estrogens were not avail- able. However, this explanation seems to be less likely, since gynecomastia became a dominant clin- ical symptom in the male patient while blood levels of estrogens were severely suppressed. Also, the persistance of markedly increased hormone-bind- ing protein levels, despite long-term suppression of E2 and E, by the GnRH superagonist in one of our female patients, is not compatible with increased estrogen secretion.
The increments of binding proteins (SHBG > CBG > TBG) are fairly similar to those seen in pregnancy (21,26,27), but the enhancement of CBG is at least twice as high during o,p’-DDD ther- apy. Beastall et al. (28) found slight but significant increases in CBG and SHBG during phenytoin ad- ministration. Consequently, enzyme induction by o,p’-DDD, which is expressed by the rise in y-glu-
tamyl transferase in this study, could represent an- other factor in binding protein stimulation. Eleva- tion of CBG could result from increased synthesis or reduced disposal owing to increased sialylation, as has been shown for TBG during estrogen ther- apy (29), or a combination of the two. Since our investigation was carried out in a therapeutic set- ting, no data could be obtained on clearance and synthesis of CBG.
In this limited study no indication was obtained that CBG during o,p’-DDD treatment differed from pretreatment CBG with regard to its cortisol- binding properties. The CBG increases were shown by two different techniques, with close par- allelism for the RIA method, and drug-induced artifacts appear to have been properly excluded. In comparison to control patients the serum in vivo binding capacity for cortisol was greatly increased (see “loading studies”), whereas normal serum cor- tisol was associated with elevated plasma ACTH in one patient. Moreover, double cortisol substitution induced no clinical signs of Cushing’s syndrome.
On the other hand, TSH and free T4 were normal despite greatly decreased total T4 and in- creased TBG levels. Similar changes in parameters of thyroid function are seen during treatment with various other drugs, in the presence of normal TBG levels (30). Therefore, o,p’-DDD could com- pete for bindings sites on TBG, as suggested by Marshall & Tompkins (8) or alter its T4-binding characteristics. Similar considerations apply to vi- tamin D-binding protein, but from this study no conclusions can be drawn regarding the steroid- binding properties of SHBG during o,p’-DDD therapy.
Finally, the slow decrease of binding protein levels after o,p’-DDD discontinuation can be mainly attributed to the slow disappearance of the drug from adipose tissue. In addition, a “rebound” en- dogenous estrogen excess may have developed on drug withdrawal in one of the female patients.
As a practical consequence, the enhancement of hormone-binding proteins imply that the use of total serum hormone levels, as the only parameter, to monitor the effects of o,p’-DDD on endocrine function is not justifiable. Particularly, the ade- quacy of cortisol substitution therapy may be re- flected more accurately by the blood levels of ACTH or free cortisol, both during o,p’-DDD ther- apy and following its withdrawal. Moreover, it seems likely that the enhanced CBG levels and the increased serum cortisol-binding capacity result in
deficient free cortisol levels, as confirmed in one patient (unpublished observation), unless more than the usual amounts of cortisol are given as re- placement therapy during o,p’-DDD administra- tion.
Acknowledgments
o,p’-DDA was kindly synthesized by Dr C. Winkel, De- partment of Organic Chemistry, Leiden. We are grateful to the nursing staff for their support, the personnel of Clinical Chemistry for their accuracy, Dr. J. Hermans for his advise on statistical analysis, Mrs M. Sjardin-van Leeu- wen for her secretarial help, and to Prof Dr D. van der Heide for his critical reading of the manuscript.
References
1. Zabbo A, Stratton RA, Montie JE. Cancer. In: Javad- pour N, ed. Principles and management of adrenal cancer. Berlin-Heidelberg: Springer-Verlag 1987: 113-20.
2. Martz F, Straw JA. Treatment with o,p’-DDD (mito- tane) decreased P-450, heme, and microsomal pro- tein content in the dog adrenal cortex in vivo. Res Commun Chem Pathol Pharmacol 1976;13:83-92.
3. Bledsoe T, Island DP, Ney RL, Liddle GW. An effect of o,p’-DDD on the extra-adrenal metabolism of cor- tisol in man. J Clin Endocrinol Metab 1964;24:1303- 11.
4. Van Slooten H, Moolenaar AJ, van Seters AP, Smeenk D. The treatment of adrenocortical carci- noma with o,p’-DDD: prognostic implications of serum level monitoring. Eur J Clin Oncol 1984; 20:47-53.
5. Robinson BG, Hales IB, Henniker AJ, et al. The effect of o,p’-DDD on adrenal steroid replacement therapy requirements. Clin Endocrinol (Oxf) 1987; 27:437-44.
6. Hague RV, May W, Cullen DR. Hepatic microsomal enzyme induction and adrenal crisis due to o,p’-DDD therapy for metastic adrenocortical carcinoma. Clin Endocrinol (Oxf) 1989;31:51-7.
7. Southren L, Tochimoto S, Isurugi K, Gordon GG, Krikun E, Stypulkowski W. The effect of 2,2-bis (2-chlorophenyl-4-chlorophenyl)-1,1-dichloroethane (o,p’-DDD) on the metabolism of infused cortisol-7- $H. Steroids 1966;7:11-29.
8. Marshall JS, Tompkins LS. Effect of o,p’-DDD and similar compounds on thyroxine binding globulin. J Clin Endocrinol Metab 1968;28:386-92.
9. Tron’ko MD, Kravchenko VI. Effect of o,p’-DDD on the binding ability of transcortin in dogs, rats, guinea pigs and chickens. Fisiol Zh (Moscow) 1971;17:245-7.
10. Schteingart DE, Tsao HS, Taylor CI, Mckenzie A, Victoria R, Therrien BA. Sustained remission of Cushing’s disease with mitotane and pituitary irradi- ation. Ann Int Med 1980;92:613-9.
11. Luton JP, Cerdas S, Billaud L, et al. Clinical features of adrenocortical carcinoma, prognostic factors and the effect of mitotane therapy. N Engl J Med 1990; 322:1195-201.
12. Moolenaar AJ, van Slooten, van Seters AP, Smeenk D. Blood levels of o,p’-DDD following administration in various vehicles after a single dose and long-term treatment. Cancer Chemother Pharmacol 1981; 7:51-4.
13. Bouillon R, van Baelen H, De Moor P. The measure- ment of the vitamin D-binding protein in human serum. J Clin Endocrinol Metab 1977;45:225-31.
14. Van Baelen H, De Moor P. Immunochemical quan- tification of human transcortin. J Clin Endocrinol Metab 1974;39:160-3.
15. Moolenaar AJ, Niewint JWM, Oei IT. Estimation of o,p’-DDD in plasma by gas-liquid chromatography. Clin Chim Acta 1977;76:213-8.
16. Bouillon R, van Herck E, Jans I, Biauw Keng Tan, van Baelen H, De Moor P. Two direct (non-chro- matographic) assays for 25-hydroxyvitamin D. Clin Chem 1984;30:1731-6.
17. Mudde AH, van der Berg H, Boshuis PG, Breedveld FC, et al. Ectopic production of 1,25-dihydroxy-vi- tamin D by B-cell lymphoma as a cause of hypercal- cemia. Cancer 1987;59:1543-6.
18. Grummit O, Buck A, Egan R. Di-(p-chlorophenyl) acetic acid. In: Horning EC, ed. Org. Synth. Organic Syntheses Collective Coll. Vol. 3. New York: Wiley and Sons, 1955:270-1.
19. Schellens JHM, van der Wart JHF, Danhof M, van der Velde EA, Breimer DD. Relationship between the metabolism of antipyrine, hexobarbitone and threophylline in man as assessed by a “coctail” ap- proach. Br J Clin Pharmacol 1988;26:373-84.
20. Harff GA, Helversteijn WT. Computer program for comparison of methods evaluation. Clin Chem 1987; 33:1262.
21. Skjöldebrand L, Brundin J, Carlström A, Pettersson T. Thyroid associated components in serum during normal pregnancy. Acta Endocrinol (Copenh) 1982; 100:504-11.
22. Westphal U. Steroid-protein interactions II. Berlin- Heidelberg: Springer-Verlag 1986;27. (Monographs on Endocrinology).
23. Hrdina PD, Singhal RL, Ling GM. DDT and related chlorinated hydrocarbon insecticides: pharmacolog- ical basis of their toxicity in mammals. Adv Pharma- col Chemother 1975;12:13-88.
24. Loeber JG, van Velzen FL. Uterotrophic effect of B-HCH, a food chain contaminant. Food Addit Contam 1984;1:63-6.
25. van Velzen FL. The estrogenicity of the environmen- tal contaminant ß-hexa-chlorocyclohexane (Thesis). Utrecht, The Netherlands: State University of Utrecht, 1986.
26. Doe RP, Fernandez R, Seal US. Measurement of cor- ticosteroid binding globulin in man. J Clin Endocri- nol Metab 1964;24:1029-39.
27. Uriel J, Dupiers M, Rimbaut C, Buffe D. Maternal serum levels of sex steroid-binding protein during pregnancy. Br J Obstet Gynaecol 1981;88:1229-32.
28. Beastall GH, Cowan RA, Gray JMB, Fogelman I. Hormone binding globulins and anticonvulsant ther- apy. Scott Med J 1985;30:101-5.
29. Ain KB, Mori Y, Refetoff S. Reduced clearance rate of thyroxine-binding globulin (TBG) with increased sialylation: a mechanism for estrogen-induced eleva- tion of serum TBG concentration. J Clin Endocrinol Metab 1987;65:689-96.
30. Stockigt JR, Lim CF, Barlow JW, et al. Interaction of furosemide with thyroxine-binding sites: in vivo and in vitro studies and comparison with other inhibitors. J Clin Endocrinol Metab 1985;60:1025-31.
Received June 21st, 1990. Accepted December 13th, 1990.
Dr Arnoud P. van Seters, Department of Endocrinology, University Hospital, P.O. Box 9600, NL-2300 RC Leiden, The Netherlands.