The source of estradiol-17β in trophoblastic neoplasia

The source of estradiol-17β in trophoblastic neoplasia

The source of estradiol- 17J in trophoblastic neoplasia M. YUSOFF DAWOOD, M.D., CH.B. (SHEFF), M.MED. (S’PORE), M.R.C.O.G. New York, New York...

440KB Sizes 0 Downloads 24 Views

The source of estradiol- 17J in trophoblastic neoplasia M.

YUSOFF

DAWOOD,

M.D.,

CH.B.

(SHEFF),

M.MED.

(S’PORE),

M.R.C.O.G.

New York, New York

Unconjugated and conjugated estradiol-I 7J (E 2) were measured in the sera of four patients with hyiydatidzform mole at frequent regdar intervals after removal of the mole, in the left and right ovarian vein blood in a patient with hydatidi$orm mole, in the peripheral sera and serowfluid of the molar vesicles in seven cases of hydattizyorm mole, and in the fluid from th e left and right theta lutein cyst (TLC) of a molar pregnancy, by a radioimmunoassay method. Patients with highly elevated serum Ez had a rapid clearance of ES within 24 to 36 hours after removal of the mole; those with minimal E2 elevation had a slower clearance of the hormone from the circulation. Mole vesicleJEuid had undetectable Ez but a wide range of unconjugated E2 which was usually lower tkn in the peri@eral blood. Both ovarian vein blood and TLCJuid have higher E, concentrations than the peripheral blood. The significance of thesejndings is discussed in relation to the contribution of the trophoblast and the ovaries to the circulating Ez

MOLAR PREGNANCY exhibits a pregnancy situation with hyperplacentosis, often with hyperstimulated ovaries, but usually absence of an identifiable fetus. Hence it is a unique experimental endocrine model in human subjects for studying steroid hormones and its source in relation to the trophoblast, the ovaries, and the fetus. Dawood and associates’, 2 have shown that peripheral serum progesterone and serum estradiol-17@(E,) are significantly elevated in a high percentage of unaborted hydatidiform moles. Hydatidiform moles with theta lutein cysts (TLC) have significantly higher serum Ez than those without TLC2 In normal pregnancy, up to 60 per cent of E2 is derived from the fetoplacental unit and depends on dehydroepiandrosterone sulfate (DHEAS) supply from both maternal and fetal adrenals. In vitro incubation of labeled estradiol precursors with molar tissue demonstrated that hydatidiform moles were capable of synthesizing estrone (E,) and estradiol (E2) but not estriol (E3).3 Patients with benign hydatidiform moles converted DHEAS to E, and E2 and to a lesser extent E3.4, 5 In vitro studies with choriocarcinoma cell lines From the Department of Obstetrics and Gynecology, The New York Hospital-Cornell Medical Center. Received for publication September 19, 1974. Revised Accepted

November November

I I, 1974. 1 I, 1974.

Reprint requests: Dr. M. Yusoff Dawood, The New York Hospital-Cornell Medical Center, 525 East 68th Street, New York, New York 10021.

have revealed their ability to produce E, and E2.S* 7 Only traces but nonmeasurable amounts of estrogens were reported in the fluid obtained from TLC’s of a patient with hydatidiform mo1e.s Thus it appears that the molar tissue is comparatively more active in estrogen production than the ovarian TLC. However, there is a significant correlation between serum human chorionic gonadotropin (HCG) and serum E2 in unaborted hydatidiform moles and a significantly higher serum E2 concentration in molar pregnancies with TLC than in molar pregnancies without TLC2 Therefore elevated peripheral serum E2 could originate from the molar trophoblast and/or the ovaries besides small amounts that are contributed by the adrenals. Since the half-life of E2 is short, the fall in serum E2 concentration following removal of the molar tissue and/or the ovaries would indicate the relative contributions by these two tissues to the peripheral serum E2 levels. The purpose of the present study was to determine (1) the pattern of peripheral serum E2 decline following removal of the molar trophoblast and (2) the concentrations of E2 in molar vesicle fluid, theta lutein cyst fluid and ovarian venous blood in relation to E2 concentration of peripheral venous blood in patients with hydatidiform mole.

Materials and methods Venous unaborted

blood was collected from four patients with hydatidiform mole prior to evacuation of

Volume Number

Estradiol-17p

123 3

the uterus or hysterectomy, and at regular intervals (1, 6, 12, and 24 hours after surgery) up to 24 to 36 hours after operation. The serum was separated within 2 hours of blood collection and stored at -20” C. until ready for hormonal assay. Estradiol concentrations in the serous fluid obtained from hydatidiform mole vesicles in seven cases of hydatidiform mole and in the serous fluid of a left and right ovarian TLC in another case of hydatidiform mole were also measured. Fresh molar vesicles removed either at dilatation and curettage or hysterectomy were relieved of any extravesicular fluid with tissue paper. The intravesicular fluid was then aspirated with a syringe and such fluid pooled from 10 to 20 vesicles was regarded as representative for the mole from that patient. Fluid from TLC’s was obtained by aspirating the TLC at the time of abdominal hysterectomy. In order to determine the contribution to the elevated peripheral serum estradiol made by the ovaries in molar pregnancies, blood was collected from the ovarian veins in a patient with unaborted hydatidiform mole at the time of abdominal hysterectomy. The Fallopian tube together with the vessels in the mesosalpinx of each side was doubly clamped to cut off blood flow from the uterus to the ovaries. After 15 minutes to permit equilibration and stabilization, the ovarian vein was cannulated, and ovarian blood was collected. The serum was removed within the next 2 hours and stored at -20” C. until ready for assay. Hormonal assay. Both unconjugated and conjugated estradiol-17/3 were measured in the serum, molar vesicle fluid, and ovarian TLC fluid by a radioimmunoassay method as described by Hotchkiss and associatesg using a speciftc antiserum raised in sheep against 17/3-estradiol-6-(0-carboxymethyl) oxime coupled to bovine serum albumin. This antiserum has been shown to have less than 1 per cent cross-reaction with estrone and estriol.‘O Thus preliminary chromatographic separation of the extract was not essential. The method of extracting estradiol from the serum in hydatidiform mole and subsequent modifications of the assay have been described elsewhere by Dawood and associates2* I1 Briefly, 100 ~1 of serum or the biological fluid to be assayed is diluted with double-distilled water to 500 ~1 and then extracted twice with freshly redistilled diethyl ether to give the extract containing unconjugated estradiol. The remaining serum or biological fluid is then hydrolyzed with 15 per cent hydrochloric acid and then reextracted with diethyl ether again. This extract contains the conjugated estradiol. The extracts are dried and redissolved in phosphate buffer pH 7.4. A suitable

i

i HOURS

in trophoblastic neoplasia

l-2 AFTER

l-1

2-4

$8

287

8-8

OPERATION

Fig. 1. Decline of serum unconjugated estradiol-I?‘/3 in four patients with hydatidiform mole following surgery. aliquot of the extract is taken for radioimmunoassay. Whenever the aliquot contained too much or too little estradiol to fall within the standard curve, the assay was repeated using a larger or smaller volume of aliquot. All extraction losses were monitored by the addition of estradiol-2,4,6,7-3H (specific activity 372 mCi. per milligram; Amersham, England) which acted as internal standard. All assays were carried out in duplicate and, whenever the duplicates varied by morr than 10 per cent, the assay was repeated. The lowest amount of E2 that could be reliably detected in an assay tube was 20 pg. The interassay and intra-assay variability was less than 10 per cent. Serial assays on any single patient were always performed in a Gngle run. All results expressed were corrected for extraction losses. Results Decline of Ez after removal of a male. The clearance of serum unconjugated Ez, conjugated E,, and total serum Ez following dilatation and curettage and following hysterectomy in four patients with hydatidiform mole is shown in Figs. 1 to 3. In two patients who had markedly elevated serum unconjugated E2 (cases 1 and

288

Dawood

4 T. H.

I

12

11

24

HOURS Fig. 2. Decline of serum evacuation or removal of Fig. 3. Decline of serum evacuation or removal of

I * ’ I 03 I I 12 AFTER OPERATION

SE 01

conjugated

ll

estradiol-17P in four cases of hydatidiform with the mole in situ. total estradiol-17P in four patients with hydatidiform the uterus with the mole in situ.

21

88

S8

mole following

the uterus

4) prior to surgery, the decline in serum Ez was very marked during the first 6 hours following dilatation and curettage and subsequently showed a much slower decline up to 24 hours after surgery. In the other two patients (cases 2 and 3, Fig. l), the preoperative unconjugated Ez levels, which were 2.35 and 2.74 ng. per milliliter, were not significantly elevated and showed a very remarkably slow fall up to 24 hours after evacuation (Table I). The pattern of decline of total serum Ez (Fig. 3) in the four patients showed a pattern similar to the decline in unconjugated Ez. In contrast, although serum conjugated Ez showed some similarity in its decline after operation in the four patients, the decline was much more spread out over the period of 24 hours. Molar vesicle fluid. The concentration of unconjugated and conjugated & in molar vesicle fluid and in the peripheral venous blood in seven patients with hydatidiform mole is shown in Table II. In all the cases studied, molar vesicle fluid had consistently nonmeasurable levels of conjugated estradiol whereas the concentration of unconjugated E, was between 0.3 and 62.0 ng. per milliliter. The vesicle/serum unconjugated Ez ratio was between 1: 0.16 and 1: 121.7. TLC fluid and ovarian vein blood. In one patient with molar pregnancy, the left ovarian vein had a serum unconjugated E, concentration of 31.07 ng.

mole following

per milliliter and conjugated Ez concentration of 12.40 ng. per milliliter whereas the corresponding values from the right ovarian vein were 46.53 and 15.70 ng. per milliliter (Table III). This was about 1.26 times higher in the left ovarian vein and about 1.8 times higher in the right ovarian vein than the concentration in the peripheral vein. Unconjugated E, concentration was 18.4 and 22.1 ng. per milliliter in the serous fluid from the left and right TLC of a hydatidiform mole, whereas the peripheral serum unconjugated Ez was 12.0 ng. per milliliter. Conjugated Ez was 12.5 and 10.8 ng. per milliliter in the left and right TLC, respectively, whereas the corresponding level in the peripheral serum was 10.2 ng. per milliliter.

Comment Peripheral serum Ez concentrations in unaborted hydatidiform moles with and without TLC and in aborted moles showed elevated levels in intact moles and suggest the source of the hormone to be both the trophoblast and the ovaries, especially when TLC’s are present.2 In the present study, two basic patterns of E2 clearance in the peripheral blood are seen following removal of the molar trophoblast. In patients with elevated peripheral blood Ez levels, the initial fall during the first 6 hours is very rapid in both the unconjugated E2 and total EE. A much slower secondary

296

Dawood

r chest x-rays tomograms ~~~\~~

PULMONARY

8

CHORIOCARCINOMA cl%@2

5 rt. middle

7970

tobectomy

1971

Fig. 6. Serum progesterone and serum HCG during chemotherapy of pulmonary choriocarcinoma. Pulmonary choriocarcinoma recurred 5 years after initial chemotherapy for postmolar choriocarcinema in 1965. Total hysterectomy was performed in 1965. Poor response to methotrexate necessitated lohectomy. The elevated serum progesterone levels in November and December, 1970, and January and March, 197 I, could possibly be due to spontaneous corpus luteum activity in the absence of HCG.

October

1970

november

1970

dec.

Fig. 7. Serum progesterone and serum HCG during treatment of a nonmoiar choriocarcinoma. Uterine chori~arcinoma was preceded by a stillbirth 2 months earlier. Both ovaries had theta lutein cysts and were conserved. M’lX = intramuscular methotrexate.

290

Dawood

decline is seen over the next 18 hours. With unconjugated Ez the fall over the 24 hour period is less markedly diphasic. The second pattern of E2 clearance after molar evacuation which showed a remarkably slow clearance of E2 from the peripheral blood is seen in two cases of molar pregnancy with insignificantly elevated peripheral blood E, levels. These results indicate that in hydatidiform moles with elevated peripheral serum Ez, the principal source of the hormone is the molar trophoblast. The rapid initial clearance of both unconjugated Ez and total Ez following removal of the trophoblast is consistent with the short half-life of Ez and its origin from the trophoblast. The more gradual fall in conjugated Ez is attributable to the continuous replacement of freshly conjugated Ez by the hepatic conjugation of the unconjugated E, that is being cleared. The continuous basal level contribution by the ovaries to the peripheral Ez levels is responsible for the slower fall of Ez many hours after removal of the hydatidiform mole and its persistence 24 hours later. The two cases of hydatidiform mole with TLC had elevated unconjugated Ez levels prior to surgery, thus suggesting that the ovary would be responsible for the elevated levels. However, unconjugated Ez was only 0.54 ng. per milliliter, 36 hours after evacuation of the uterus, although the TLC was still present and a substantial amount of HCG was still present in the circulation at that time. Hence the trophoblast is the more probable primary source of the elevated peripheral Ez.

REFERENCES

1. Dawood, M. Y.: AM. J. OBSTET. GYNECOL. 119: 911,1974. 2. Dawood, M. Y., Ratnam, S. S., and Teoh, E. S.: AM. J. OBSTET. GYNECOL. 119: 904, 1974. 3. Van Leusden, H. A., and Villee, C. A.: Clin. Endocrinol. 26: 842: 1966. 4. Barlow, J. J., Goldstein, D. P., and Reid, D. E.: J. Clin. Endocrinol. 27: 1028, 1967. 5. Jungmann, R. A., and Schweppe, J. S.: J. Clin. Endocrinol. Metab. 27: 1151. 1967. 6. Patillo, R. A., Hussa, R. 01, Huang, W. Y., Delfs, E., and Mattingly, R. F.: J. Ciin. Endocrinol. 34: 59, 1972.

No record of any previous study on the estrogen concentration of the molar vesicle fluid exists in the literature. Conjugated Ez was not detectable in the molar vesicle fluid but unconjugated E, showed a wide variation in concentration and a loo-fold difference in its ratio to the peripheral blood in all cases studied. The wide range in unconjugated E, concentration in molar vesicle fluid may reflect either (1) the variability in Ez production in hydatidiform mole, as has been observed by MacDonald and Siiteri,” or (2) the variability in the secretion of E2 synthesized by the molar trophoblast into the peripheral circulation. That the ovaries are active in Ez production in patients with hydatidiform mole is clearly demonstrated by the higher levels of Ez in ovarian venous blood in comparison to the peripheral venous blood. Besides the large amounts of Ez secreted into the circulation via the ovarian veins, a substantial amount is either secreted into or retained in the TLC fluid which has a higher concentration of Ez than the peripheral blood. The finding of substantial amounts of Ez in both TLC’s of a patient with hydatidiform mole is in direct contrast to the only previously published study of estrogens in TLC where nonmeasurable amounts of estrogens were reported.* The differences in results could be due to the less sensitive method of hormone assay used in that study. Thus the ovaries are active in steroidogenesis in trophoblastic disease and contribute to the peripheral blood E2 levels.

7. Kohler, P. O., Bridson, W. E., Hammond, J. M., van Thiel, D. H., and Kirschner, M. H.: In vitro 6: 384, 1971. 8. Coutts, 1. R. T., MacNaughton, M. C., Ross, P. E., and Walker,J.: J. Endocrinol~4: 335, 1969. 9. Hotchkiss. 1.. Atkinson. I. E.. and Knobil. E.: Endocrinology 89: 177, 1971. 10. Exley, D., Johnson, M. W., and Dean, P. D.: Steroids 18: 605, 1971. 11. Dawood, M. Y., and Ratnam, S. S.: Obstet. Gynecol. 44: 194, 1974. 12. MacDonald, P. C., and Siiteri, P. K.: Steroids 8: 589, 1966. .J,